research on food deprivation

• Mission and Vision
• Scientific Advancement Plan
• Science Visioning
• Research Framework
• Minority Health and Health Disparities Definitions
• Organizational Structure
• Staff Directory
• About the Director
• Director’s Messages
• News Mentions
• Presentations
• Selected Publications
• Director's Laboratory
• Congressional Justification
• Congressional Testimony
• Legislative History
• NIH Minority Health and Health Disparities Strategic Plan 2021-2025
• Minority Health and Health Disparities: Definitions and Parameters
• NIH and HHS Commitment
• Foundation for Planning
• Structure of This Plan
• Strategic Plan Categories
• Summary of Categories and Goals
• Scientific Goals, Research Strategies, and Priority Areas
• Research-Sustaining Activities: Goals, Strategies, and Priority Areas
• Outreach, Collaboration, and Dissemination: Goals and Strategies
• Leap Forward Research Challenge
• Future Plans
• Research Interest Areas
• Research Centers
• Research Endowment
• Community Based Participatory Research Program (CBPR)
• SBIR/STTR: Small Business Innovation/Tech Transfer
• Solicited and Investigator-Initiated Research Project Grants
• Scientific Conferences
• Training and Career Development
• Loan Repayment Program (LRP)
• Data Management and Sharing
• Social and Behavioral Sciences
• Population and Community Health Sciences
• Epidemiology and Genetics
• Medical Research Scholars Program (MRSP)
• Coleman Research Innovation Award
• Health Disparities Interest Group
• Art Challenge
• Breathe Better Network
• Healthy Hearts Network
• DEBUT Challenge
• Healthy Mind Initiative
• Mental Health Essay Contest
• Science Day for Students at NIH
• Fuel Up to Play 60 en Español
• Brother, You're on My Mind
• Celebrating National Minority Health Month
• Reaching People in Multiple Languages
• Funding Strategy
• Active Funding Opportunities
• Expired Funding Opportunities
• Technical Assistance Webinars

• Community Health and Population Sciences
• Clinical and Health Services Research
• Integrative Biological and Behavioral Sciences
• Intramural Research Program
• Training and Diverse Workforce Development
• Inside NIMHD
• ScHARe HDPulse PhenX SDOH Toolkit Understanding Health Disparities For Research Applicants For Research Grantees Research and Training Programs Reports and Data Resources Health Information for the Public Science Education

• Understanding Health Disparities Series

Food Accessibility, Insecurity and Health Outcomes

• PhenX SDOH Toolkit
• Understanding Health Disparities
• For Research Applicants
• For Research Grantees
• Research and Training Programs
• Reports and Data Resources
• Health Information for the Public
• Science Education

Go to food and nutrition insecurity scientific resources

Go to food accessibility and insecurity as an SDOH

Go to NIMHD food accessibility/insecurity resources

Having access to nutritious food is a basic human need.

• Food security means having access to enough food for an active, healthy life.
• Nutrition security means consistent access, availability, and affordability of foods and beverages that promote well-being, prevent disease, and, if needed, treat disease.

On the other hand, food and nutrition insecurity is an individual-, household-, and neighborhood-level economic and social condition describing limited or uncertain access to adequate and affordable nutritious foods and is a major public health concern.

Food insecurity and the lack of access to affordable nutritious food are associated with increased risk for multiple chronic health conditions such as diabetes , obesity, heart disease, mental health disorders and other chronic diseases . In 2020, almost 15% of U.S. households were considered food insecure at some point in time, meaning not all household members were able to access enough food to support active, healthy lifestyles. In nearly half of these households, children were also food insecure ( see chart above ), which has implications for human development and school experience . Food insecurity disproportionately affects persons from racial and ethnic minority and socioeconomically disadvantaged populations:

• 20% of Black/African American households were food insecure at some point in 2021 , as were 16% of Hispanic/Latino households when compared to 7% of White households.
• Food insecurity for U.S. Hispanic/Latino adults differs by origin. Current national data is not available but from 2011-2014 food insecurity was highest among those identifying from Puerto Rico (25.3%), followed by Mexico (20.8%) Central and South America (20.7%) and Cuba (12.1%).
• In the past 20 years, American Indian/Alaskan Native (AI/AN) households have also been at least twice as likely to have experienced food insecurity when compared with White households, often exceeding rates of 25% across different regions and AI/AN communities.
• Native Hawaiian and Pacific Islander (NHPI) adults also experience a high food insecurity prevalence (20.5%) and had significantly higher odds of experiencing low and very low food security compared with White households.
• While national data on specific Asian American national origin populations is not readily available, among Asian Americans living in California from 2001-2012 , food insecurity was highest among Vietnamese households (16.4%), followed by Filipino (8.3%), Chinese (7.6%), Korean (6.7%), South Asian (3.14%), and Japanese households (2.3%), highlighting considerable variation across Asian American communities.
• Food insecurity is inextricably linked to poverty , with 35.3 % of households with incomes below the federal poverty line being food insecure.
• Although the graph below on national trends in food insecurity does not capture the full impact of the COVID-19 pandemic, food insecurity is likely to increase, and racial and ethnic disparities in food insecurity experiences could worsen.

Healthy food accessibility and insecurity is a social determinant of health.

Food and nutrition insecurity are predominantly influenced by the local environment, including surrounding neighborhood infrastructure, accessibility, and affordability barriers. Access to grocery stores that carry healthy food options (such as fresh fruit, vegetables, low-fat fish and poultry) are not located equitably across residential and regional areas in the United States.

Areas that lack access to affordable, healthy foods are known as food deserts . Food deserts are:

• Found in urban or suburban neighborhoods that lack grocery stores (supermarkets or small grocery stores) that offer healthy food options.
• Found in rural areas and neighborhoods where the nearest grocery stores are too far away to be convenient or accessible.
• More prevalent in neighborhoods that are comprised of a majority of racial or ethnic minority residents or in rural AI/AN communities.
• More likely found in areas with a higher percentage of residents experiencing poverty , regardless of urban or rural designation.

Urban, suburban, and rural areas can also be overwhelmed with stores that sell unhealthy calorie-dense and inexpensive junk foods, including soda, snacks, and other high sugar foods. This is known as a food swamp . Food swamps:

• Reduce access to nutritional foods and provide easier access to unhealthy foods .
• Are a predictor of obesity , particularly in communities where residents have limited access to their own or public transportation and experience the greatest income inequality.

Reducing food and nutrition insecurity in the U.S. will require a multifaceted approach that considers, among other possibilities:

• Strategies that engage communities in local health programs; for example, recruiting community partners to assist in addressing gaps between food access and intake .
• Interventions that utilize federal food and nutritional supplemental programs, including the Supplemental Nutrition Assistance Program ( SNAP ) and the Special Supplemental Nutritional Program for Women, Infants, and Children ( WIC ).
• Leveraging local and federal policies targeting food insecurity; for example, retail store interventions , where healthy food placement, promotion and price influence healthier choices; sweetened beverage taxes to reduce the purchase appeal to consumers; and junk food taxes balanced with removal of taxes on water and fruits and vegetables.

NIMHD is studying and addressing issues related to food and nutrition insecurity through a variety of initiatives:

NIH Publication

Research Opportunities to Address Nutrition Insecurity and Disparities Coauthored by Shannon N. Zenk, Lawrence A. Tabak and Eliseo J. Pérez-Stable, JAMA 2022

NIMHD Events on Food Insecurity

Food Insecurity, Neighborhood Food Environment, and Nutrition Health Disparities: State of the Science NIMHD co-sponsored this September 2021 workshop led by the NIH Office of Nutrition Research and its Nutrition and Health Disparities Implementation Working Group

NIMHD Hosts Senior Research Investigators to Present on Food Insecurity and Related Topics:

NIMHD co-sponsored the November 2020 virtual workshop NIH Rural Health Seminar: Challenges in the Era of COVID-19 . This workshop focused on long-standing health disparities and social inequities experienced by rural populations, and featured experts on food insecurity, including:

• Dr. Brenda Eskenazi , Professor in Maternal and Child Health and Epidemiology, Brian and Jennifer Maxwell Endowed Chair in Public Health and Director of the Center for Environmental Research and Children’s Health, University of California, Berkeley, who spoke on the topic of COVID-19 and the impact on health of Californian farmworkers .
• Dr. Alice Ammerman , Mildred Kaufman Distinguished Professor of Nutrition, Director, Center for Health Promotion and Disease Prevention, University of North Carolina, who spoke on interventions to address job loss and food security in rural communities during COVID .

NIMHD Content on Food Insecurity

Nimhd science visioning and research strategies.

As part of the Scientific Visioning Research Process , NIMHD developed a set of 30 strategies to transform minority health and health disparities research . Several of these strategies focus on issues related to food security and accessibility, including:

• Assessing how environment and neighborhood structures such as areas where people have limited access to a variety of healthy and affordable foods (or food deserts) influence health behaviors.
• Promoting multi-sectoral interventions that address the structural drivers of food deserts.
• Promoting interventions that address the social determinants of health within health care systems, including food insecurity.

Publications

Food Insecurity and Obesity: Research Gaps, Opportunities, and Challenges Dr. Derrick Tabor, NIMHD Program Officer, co-authored “Food insecurity and obesity: research gaps, opportunities” in Translational Behavioral Medicine . This review highlights NIH funding for grants related to food insecurity and obesity, identifies research gaps, and presents upcoming research opportunities to better understand the health impact of food insecurity.

NIMHD Research Framework

Native Hawaiian Health Adaptation The NIMHD Research Framework was adapted by Keawe’aimoku Kaholokula, Ph.D., University of Hawai’i at Mānoa, to reflect social and cultural influences of Native Hawaiian health . Ka Mālama Nohona (nurturing environments) to support Native Hawaiian health include strategic goals of food sovereignty and security to promote a strong foundation for healthy living.

NIMHD Articles

NIMHD Research Features

• The Osage Community Supported Agriculture Program: A Tribal Nation’s Effort Toward Food Security and Food Sovereignty
• The Navajo Nation Junk Food Tax and the Path to Food Sovereignty
• Fighting Cancer—and Reducing Disparities—Through Food Policy
• Fresh Food for the Osage Nation: Researchers and a Native Community Work Toward Improved Food Resources and Food Sovereignty

NIMHD Insights Blog

• Amplifying the Voice of Native Hawaiian and Pacific Islander Communities Amid the COVID-19 Crisis by Joseph Keawe‘aimoku Kaholokula, Ph.D.
• Racism and the Health of Every American by NIMHD Director Eliseo J. Pérez-Stable, M.D.
• The Future of Minority Health and Health Disparities Research by Tany Agurs-Collins, Ph.D., R.D., and Susan Persky, Ph.D.
• Addressing Social Needs and Structural Inequities to Reduce Health Disparities: A Call to Action for Asian American and Pacific Islander Heritage Month by Marshall H. Chin, M.D., M.P.H.
• “Insights” on Simulation Modeling and Systems Science, New Research Funding Opportunity by Xinzhi Zhang, M.D., Ph.D.

NIMHD Funding Resources and Opportunities

Funding opportunity announcements (foas).

NIMHD supports many FOAs that include topics related to food security as an area of research interest:

• Request for Information: Food Is Medicine Research Opportunities
• Notice of Special Interest: Stimulating Research to Understand and Address Hunger, Food and Nutrition Insecurity
• Community Level Interventions to Improve Minority Health and Reduce Health Disparities (R01 Clinical Trial Optional)
• Addressing Health Disparities Among Immigrant Populations through Effective Interventions (R01 Clinical Trial Optional)
• Health Services Research on Minority Health and Health Disparities (R01 Clinical Trial Optional)
• Long-Term Effects of Disasters on Health Care Systems Serving Populations Experiencing Health Disparities (R01 Clinical Trial Optional)
• Please see our list of Active NIMHD Funding Opportunities for more.

NIMHD-Supported Research Projects

See a list of active NIMHD-supported research projects studying food security and related topics .

NIMHD-Supported, NIH-Wide Initiatives

The PhenX Toolkit provides recommended and established data collection protocols for conducting biomedical research. There are PhenX protocols available for assessing and understanding food insecurity and food swamps .

The Strategic Plan for NIH Nutrition Research is the first NIH-wide strategic plan for nutrition research that highlights crosscutting, innovative opportunities to advance nutrition research from basic science to experimental design to research training. The plan emphasizes the need for studies on minority health and nutrition-related health disparities research.

Additional Resources and Data

The Centers for Disease Control’s Healthier Food Environments: Improving Access to Healthier Foods discusses CDC efforts to improve food access within the community.

The United States Department of Agriculture has two databases that compile national data on food accessibility:

• The Food Access Research Atlas provides food access data for populations within census tracts and mapped overviews of food access for low-income communities.
• The Food Environment Atlas assembles statistics on food environment indicators to stimulate research on the determinants of food choice and diet quality.

The Healthy People initiative provides 10-year, measurable public health objectives and useful tools to help track progress. The Healthy People 2030 includes food insecurity as a social determinant of health.

Page updated July 3, 2024

Page updated February 24, 2023

NIMHD Fact Sheet

Read about what is happening at NIMHD at the News and Events section

301-402-1366

[email protected]

Connect with Us

Subscribe to email updates

Staying Connected

• Funding Opportunities  
• News & Events  
• HHS Vulnerability Disclosure  
• Privacy/Disclaimer/Accessibility Policy  
• Viewers & Players  

The Psychology of Food Cravings: the Role of Food Deprivation

• September 2020
• Current Nutrition Reports

• Universität Regensburg

Abstract and Figures

Discover the world's research

• 25+ million members
• 160+ million publication pages
• 2.3+ billion citations
• Ivana Hromatko
• Maja Batorek

• Saman Sarram

• Margareth Da Silva Oliveira
• Rejane Rosaria Grecco dos Santos
• Mirna Brilmann

• Sangeeta Yadav
• Devyani Tomar
• Kajol Batta

• Sachin Kumar Sharma
• Acharya Balkrishna
• Pınar Hamurcu
• Seminur Barlak
• Lawrence Josephs
• J CLIN NURS
• Huiling Zeng
• Shuanglian Yi
• Katherene O.-B. Anguah

• Ana I. Castro

• J BEHAV MED

• Robert I. Berkowitz

• Manju Bhapkar

• Nerylee A. Watson
• Kathryn A. Dyer

• NEUROSCI BIOBEHAV R

• Recruit researchers
• Join for free
• Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up

Food Insecurity and Hunger in the United States: An Assessment of the Measure (2006)

Chapter: executive summary, executive summary.

T he United States is viewed by the world as a country with plenty of food, yet not all households in America are food secure, meaning access at all times to enough food for an active, healthy life. A proportion of the population experiences food insecurity at some time in a given year because of food deprivation and lack of access to food due to economic resource constraints. Still, food insecurity in the United States is not of the same intensity as in some developing countries.

Since 1995 the U.S. Department of Agriculture (USDA) has annually published statistics on the extent of food insecurity and food insecurity with hunger in U.S. households. These estimates are based on a survey measure developed by the U.S. Food Security Measurement Project, an ongoing collaboration among federal agencies, academic researchers, and private organizations. It is an experiential measure based on reported behaviors, experiences, and conditions in response to questions in a household survey. The measure was developed over the course of several years in response to the National Nutrition Monitoring and Related Research Act of 1990 (NNMRR). The legislation specifically called for development of a standardized mechanism and instrument(s) for defining and obtaining data on the prevalence of food insecurity in the United States and methodologies that can be used across the NNMRR programs and at the state and local levels.

The USDA estimates of food insecurity are based on data collected annually in the Food Security Supplement (FSS) to the Current Population Survey (CPS). On the basis of the number of food-insecure conditions reported, households are classified into one of three categories for purposes

of monitoring and statistical analysis of the food security of the U.S. population: (1) food secure , (2) food insecure without hunger , and (3) food insecure with hunger.

The USDA estimates, published in a series of annual reports, are widely used by government agencies, the media, and advocacy groups to report the extent of food insecurity and hunger in the United States, to monitor progress toward national objectives, to evaluate the impact of particular public policies and programs, as a standard by which the performance of USDA programs is measured, and as a basis for a diverse body of research relating to food assistance programs.

In addition, USDA has a program of research for improving the measurement and understanding of food security. Despite these efforts, some major questions continue to be raised regarding the underlying concepts, the estimation methods, and the design and clarity of the questions used to construct the food insecurity scale.

PANEL CHARGE

USDA requested the Committee on National Statistics of the National Academies to convene a panel of experts to undertake a two-year study in two phases to review at this 10-year mark the concepts and methodology for measuring food insecurity and hunger and the uses of the measures. The specific tasks to be addressed in Phase 1 of the study were:

the appropriateness of a household survey as a vehicle for monitoring on a regular basis the prevalence of food insecurity among the general population and within broad population subgroups, including measuring frequency and duration;

the appropriateness of identifying hunger as a severe range of food insecurity in such a survey-based measurement method;

the appropriateness, in principle and in application, of item response theory and the Rasch model as a statistical basis for measuring food insecurity;

the appropriateness of the threshold scores that demarcate food insecurity categories—particularly the categories “food insecure with hunger” and “food insecure with hunger among children”—and the labeling and interpretation of each category;

the applicability of the current measure of the prevalence of food insecurity with hunger for assessing the effectiveness of USDA food assistance programs, in connection with the Government Performance and Results Act performance goals for the Food and Nutrition Service; and

future directions to consider for strengthening measures of hunger prevalence for monitoring, evaluation, and related research purposes.

In Phase 2 of the study the panel was to consider in more depth the issues raised in Phase 1 relating to the concepts and methods used to measure food security and make recommendations as appropriate. In addition, the panel was asked to address and make recommendations on:

the content of the 18 items and the set of food security scales based on them currently used by USDA to measure food insecurity;

how best to incorporate and represent information about food security of both adults and children at the household level;

how best to incorporate information on food insecurity in prevalence measures;

needs and priorities for developing separate, tailored food security scales for population subgroups, for example, households versus individuals, all individuals versus children, and the general population versus homeless persons; and

future directions to consider for strengthening measures of food insecurity prevalence for monitoring, evaluation, and related research purposes throughout the national nutrition monitoring system.

The Committee on National Statistics appointed a panel of 10 experts to examine the above issues. In order to provide timely guidance to USDA, the panel issued an interim Phase 1 report, Measuring Food Insecurity and Hunger: Phase 1 Report . That report presented the panel’s preliminary assessments of the food security concepts and definitions; the appropriateness of identifying hunger as a severe range of food insecurity in such a survey-based measurement method; questions for measuring these concepts; and the appropriateness of a household survey for regularly monitoring food security in the U.S. population. It provided interim guidance for the continued production of the food security estimates. This final report primarily focuses on the Phase 2 charge. The major findings and conclusions based on the panel’s review and deliberations are summarized below, followed by the text of all of the recommendations.

FINDINGS AND CONCLUSIONS

Concepts and definitions.

The broad conceptual definitions of food security and insecurity developed by the expert panel convened in 1989 by the Life Sciences Research Office (LSRO) of the Federation of American Societies for Experimental

Biology have served as the basis for the standardized operational definitions used for estimating food security in the United States. Food security according to the LSRO definition means access at all times to enough food for an active, healthy life. Food insecurity exists whenever the availability of nutritionally adequate and safe foods or the ability to acquire acceptable foods in socially acceptable ways is limited or uncertain. Food insecurity as measured in the United States refers to the social and economic problem of lack of food due to resource or other constraints, not voluntary fasting or dieting or because of illness or for other reasons. Although lack of economic resources is the most common constraint, food insecurity can also be experienced when food is available and accessible but cannot be used because of physical or other constraints, such as limited physical functioning by elderly people or those with disabilities.

Food insecurity is measured as a household-level concept that refers to uncertain, insufficient, or unacceptable availability, access, or utilization of food. It is therefore households that are classified as food secure or food insecure. It means that one can measure and report the number of people who are in food-insecure households (even though not everyone in the household need be food insecure themselves). When a household contains one or more food-insecure persons, the household is considered food insecure.

A full understanding of food insecurity requires the incorporation of its frequency and duration because more frequent or longer duration of periods of food insecurity indicate a more serious problem. Frequency and duration are therefore important elements for USDA to consider in the concept, operational definition, and measurement of household food insecurity and individual hunger.

The LSRO conceptual definition of hunger adopted by the interagency group on food security measurement is: “The uneasy or painful sensation caused by a lack of food, the recurrent and involuntary lack of access to food. Hunger may produce malnutrition over time…. Hunger … is a potential, although not necessary, consequence of food insecurity” (Anderson, 1990, pp. 1575, 1576). This language does not provide a clear conceptual basis for what hunger should mean as part of the measurement of food insecurity. The first phrase “the uneasy or painful sensation caused by a lack of food” refers to a possible consequence of food insecurity. The second phrase “the recurrent and involuntary lack of access to food” refers to the whole problem of food insecurity, the social and economic problem of lack of food as defined above.

Unlike food insecurity, which is a household-level concept, hunger is an individual-level concept. The Household Food Security Survey Module (HFSSM) in the Food Security Supplement to the CPS measures food insecurity at the household level; it does not measure the condition of hunger at

the individual level. The HFSSM does include items that are related to being hungry. Some or all of these items are probably appropriate in the food insecurity scale, but they contribute to the measurement of food insecurity and not the measurement of hunger.

The panel therefore concludes that hunger is a concept distinct from food insecurity, which is an indicator of and possible consequence of food insecurity, that can be useful in characterizing severity of food insecurity. Hunger itself is an important concept that should be measured at the individual level distinct from, but in the context of, food insecurity.

The broad conceptual definition of household food insecurity includes more elements than are included in the current USDA measure of food insecurity. Not all elements of the consensus conceptual definition of food insecurity have been incorporated into the USDA measurement of food insecurity in the United States. It was a decision of the Food Security Measurement Project to limit the operational definition and measurement approach to only those aspects of food insecurity that can be captured in a household-level survey. The other conceptually separable aspects of food insecurity are potentially distinct empirical dimensions. For example, the measurement does not include the supply of food or its safety or nutritional quality; these additional aspects would require developing measures and fielding separate surveys to measure them. Moreover, the food supply in the United States is generally regarded as safe, and nutritional adequacy is already assessed by other elements of the nutrition monitoring system, in particular the continuing National Health and Nutrition Examination Survey. The panel therefore concludes that it is neither required nor necessarily appropriate for USDA to attempt to measure all the elements of the broad conceptual definition of food insecurity as part of the HFSSM.

The labeling used to categorize food insecurity is at the heart of the criticism of the current measurement system. In particular, the category “food insecure with hunger” has come under scrutiny because of disagreement over whether hunger is actually measured. The rationale for including hunger in the label for the classification is understandable. Hunger is a politically sensitive and evocative concept that conjures images of severe deprivation, and the HFSSM does include some items that are specifically related to hunger. However, the measurement of food insecurity rather than hunger is the primary focus of the HFSSM. As an indication of the severity of food insecurity, the HFSSM asks the household respondent if in the past 12 months she or he has experienced being hungry because of lack of food due to resource constraints. This is not the same as evaluating individual members of the household in a survey as to whether or not they have experienced hunger . The panel urges USDA to consider alternate labels to convey the severity of food insecurity without the problems inherent in the current labels.

Survey Measurement of Food Insecurity and Hunger

The panel reviewed the current questions used to measure food insecurity and hunger, considered the relationship among the three major aspects of food insecurity and hunger embodied in the questions (whether the household experienced uncertainty, the perception of insufficiency in quality of diet, and reduced food intake or the feeling of hunger), and identified several design issues in the HFSSM that should be addressed.

USDA’s food security scale measures the severity of food insecurity in surveyed households and classifies their food security status during the previous year. The frequency of food insecurity and the duration of spells of insecurity are not assessed directly in the HFSSM questions that are used to classify households by food security status. Although some of the response options do offer the choice of “often, sometimes, or never,” these response options are not sufficient measures of frequency, and they are not included in the construction of the scale. In addition to the items in the HFSSM, the full supplement includes questions that focus on duration. However, these questions are not part of the 18-item HFSSM, although they have been used in research to estimate the percentage of the population that is food insecure on a given day in a given month. A recent study undertaken by USDA researchers examined the extent to which food insecurity and hunger are occasional, recurring, or frequent in the U.S. households that experience them. The panel recommends further research on the frequency and duration of food insecurity.

The panel reviewed the 18 items that constitute the food insecurity scales as well as the entire questionnaire module in which these are embedded and found many issues of questionnaire design. Consistent terminology, clustering questions so as to focus on a specific reference person or reference group (e.g., the respondent, all adults in the household, all children in the household) and on a specific reference period (e.g., 12 months versus 30 days), and developing response options that most closely map to the respondent’s representation of the behavior or attitude are all means by which questions can be designed to reduce cognitive burden and thereby improve the validity and reliability of the measures. Inevitably, questionnaire design requires balancing multiple intents and principles, and there is no perfect questionnaire design. Nevertheless, the panel concludes that the questions in the HFSSM in particular and the FSS in general can be improved by attending to these design principles to the extent possible.

Item Response Theory

USDA uses the Rasch model, a specific type of item response theory (IRT) model, to estimate the food insecurity of households. Several issues

have been raised about the use of IRT models in the measurement of food insecurity and, in particular, the use of the Rasch model.

The panel reviewed IRT and related statistical models and discussed their use and applicability to the development of such classifications as food insecurity. The panel recommends modifications of the current IRT methodology used by USDA to increase the amount of information that is used and to make the methodology more appropriate to the types of data that are currently collected using the Food Security Supplement to the Current Population Survey.

The panel reviewed how the latent variable models are estimated and issues of identifiability of these models and how IRT models are used by USDA in the measurement of food insecurity. On the basis of this review, the panel suggests how the models might be used in better ways to accomplish this measurement and recommends a simple way to modify the existing models currently used by USDA to take into account the polytomous nature of the data collected.

Survey Vehicles to Measure Food Insecurity and Hunger

USDA bases its annual report and estimates of the prevalence of food insecurity on data collected from the Food Security Supplement to the Current Population Survey. The Household Food Security Survey Module, or a modification of it, is or has been used in several surveys. One of the main objectives of the annual food insecurity measure is to monitor the estimated prevalence of food insecurity, as well as changes in its prevalence over time, at the national and state levels to assess both program policies and the possible need for program development.

After reviewing the key features of selected national surveys—the Current Population Survey, the National Health Interview Survey, the National Health and Nutrition Examination Survey, and the Survey of Income and Program Participation—the panel compared the relative merits of each, for either carrying the Food Security Supplement or conducting research to supplement the information obtained from it. The panel recommends research and testing to understand better the strengths and weaknesses of each survey in relation to the Current Population Survey, leading to the selection of a specific survey vehicle for the Food Security Supplement, or for supplementing that information for research purposes.

Food Insecurity Estimates as a Measure of Program Performance Assessment

Currently, the Food and Nutrition Service in USDA uses trends in the prevalence of food insecurity with hunger based on the HFSSM as a mea-

sure of its annual performance to implement the Government Performance and Results Act of 1993 (GPRA). That law requires government agencies to account for progress toward intended results of their activities. It requires that specific performance goals be established and that annual measurement of these output goals be undertaken to determine the success or failure of the program. The panel was asked to comment on the applicability of these data for this purpose.

The panel concludes that an overall national estimate of food insecurity is not appropriate as a measure for meeting the requirements of the GPRA. Even an appropriate measure of food insecurity or hunger using appropriate samples would not be a useful performance indicator of food assistance programs, because their performance is only one of many factors that result in food insecurity or hunger. Consequently, changes in food insecurity and hunger could be due to many factors other than the performance of the food safety net.

The panel concludes that relying exclusively on trends in prevalence estimates of food insecurity as an indicator of program results is inappropriate. To assess program results, a better understanding is needed of the transitions into and out of poverty made by low-income households and the kind of unexpected changes that frequently bring about alterations—for good or bad—in households participating in food assistance programs.

The panel is impressed with the extensive research thus far undertaken, and with the continuing research carried out by USDA. The panel urges that the research program be continued and makes several recommendations for its direction in the future.

The panel concludes that the measurement both of food insecurity and of hunger is important. The recommendations in the report are intended to improve these measurements, so that policy makers and the public can be better informed. Toward this end, the panel has recommended research efforts that should lead to improved concepts, definitions, and measurement of food insecurity and hunger. The panel has provided a detailed discussion of the analytical methods used by USDA and made recommendations for further research to improve the accuracy of the food insecurity scale and on survey alternatives. The panel recognizes that such research will take time.

RECOMMENDATIONS

On the basis of its findings and conclusions, the panel presents recommendations in five areas: concepts and definitions, labeling of food insecu-

rity data outcomes, survey measurement, item response theory and food insecurity, and survey vehicles to measure food insecurity and hunger. The text of the recommendations, grouped according to these areas, follows, keyed to the chapter in which they appear in the body of the report.

This page intentionally left blank.

The United States is viewed by the world as a country with plenty of food, yet not all households in America are food secure, meaning access at all times to enough food for an active, healthy life. A proportion of the population experiences food insecurity at some time in a given year because of food deprivation and lack of access to food due to economic resource constraints. Still, food insecurity in the United States is not of the same intensity as in some developing countries. Since 1995 the U.S. Department of Agriculture (USDA) has annually published statistics on the extent of food insecurity and food insecurity with hunger in U.S. households. These estimates are based on a survey measure developed by the U.S. Food Security Measurement Project, an ongoing collaboration among federal agencies, academic researchers, and private organizations.

USDA requested the Committee on National Statistics of the National Academies to convene a panel of experts to undertake a two-year study in two phases to review at this 10-year mark the concepts and methodology for measuring food insecurity and hunger and the uses of the measure. In Phase 2 of the study the panel was to consider in more depth the issues raised in Phase 1 relating to the concepts and methods used to measure food security and make recommendations as appropriate.

The Committee on National Statistics appointed a panel of 10 experts to examine the above issues. In order to provide timely guidance to USDA, the panel issued an interim Phase 1 report, Measuring Food Insecurity and Hunger: Phase 1 Report. That report presented the panel's preliminary assessments of the food security concepts and definitions; the appropriateness of identifying hunger as a severe range of food insecurity in such a survey-based measurement method; questions for measuring these concepts; and the appropriateness of a household survey for regularly monitoring food security in the U.S. population. It provided interim guidance for the continued production of the food security estimates. This final report primarily focuses on the Phase 2 charge. The major findings and conclusions based on the panel's review and deliberations are summarized.

READ FREE ONLINE

Welcome to OpenBook!

You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

Do you want to take a quick tour of the OpenBook's features?

Show this book's table of contents , where you can jump to any chapter by name.

...or use these buttons to go back to the previous chapter or skip to the next one.

Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

Switch between the Original Pages , where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

To search the entire text of this book, type in your search term here and press Enter .

Share a link to this book page on your preferred social network or via email.

View our suggested citation for this chapter.

Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

Get Email Updates

Do you enjoy reading reports from the Academies online for free ? Sign up for email notifications and we'll let you know about new publications in your areas of interest when they're released.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 01 May 2024

Effects of sleep deprivation on food-related Pavlovian-instrumental transfer: a randomized crossover experiment

• Wai Sze Chan   ORCID: orcid.org/0000-0001-6829-7822 1  

Scientific Reports volume  14 , Article number:  10029 ( 2024 ) Cite this article

669 Accesses

1 Altmetric

Metrics details

• Human behaviour
• Learning and memory

Recent research suggests that insufficient sleep elevates the risk of obesity. Although the mechanisms underlying the relationship between insufficient sleep and obesity are not fully understood, preliminary evidence suggests that insufficient sleep may intensify habitual control of behavior, leading to greater cue-elicited food-seeking behavior that is insensitive to satiation. The present study tested this hypothesis using a within-individual, randomized, crossover experiment. Ninety-six adults underwent a one-night normal sleep duration (NSD) condition and a one-night total sleep deprivation (TSD) condition. They also completed the Pavlovian-instrumental transfer paradigm in which their instrumental responses for food in the presence and absence of conditioned cues were recorded. The sleep × cue × satiation interaction was significant, indicating that the enhancing effect of conditioned cues on food-seeking responses significantly differed across sleep × satiation conditions. However, this effect was observed in NSD but not TSD, and it disappeared after satiation. This finding contradicted the hypothesis but aligned with previous literature on the effect of sleep disruption on appetitive conditioning in animals—sleep disruption following learning impaired the expression of appetitive behavior. The present finding is the first evidence for the role of sleep in Pavlovian-instrumental transfer effects. Future research is needed to further disentangle how sleep influences motivational mechanisms underlying eating.

Similar content being viewed by others

Potential interactive effect of positive expectancy violation and sleep on memory consolidation in dogs

Modulating social learning-induced evaluation updating during human sleep

Individual sleep need is flexible and dynamically related to cognitive function

Introduction.

Sleep curtailment is a prevalent problem across age groups and populations around the world, and is accompanied by serious public health consequences 1 . The National Sleep Foundation recommends that adults should obtain seven to nine hours of sleep per night 2 ; however, population-based studies have found that 20–30% of adults report getting fewer than six hours of sleep per night 3 , 4 . Similarly, obesity is a public health crisis in developed countries, with obesity rates also rising at alarming rates in developing countries over the past two decades 5 . Previous research suggests a potential link between these two pressing public health issues—insufficient sleep elevates the risk of obesity 6 , 7 , 8 , with metabolic, hormonal, and behavioral mechanisms proposed to account for the association 9 .

One of the plausible mechanisms underlying the relationship between sleep deprivation and obesity is that sleep deprivation alters the way we eat. Indeed, experimentally induced partial or full sleep deprivation has been found to increase hunger, caloric consumption, and neurological responsiveness to food rewards in humans 10 , 11 , 12 , 13 , 14 . A meta-analysis showed that, on average, experimental sleep deprivation increased caloric consumption by 385 calories per day in humans 15 . Although sleep loss appears to affect eating, the underlying processes remain unclear. Some previous studies have found that sleep loss alters metabolic and hormonal mechanisms in humans, such as up-regulating ghrelin, the hormone signaling hunger, and down-regulating leptin, the hormone signaling satiety 16 , 17 ; however, mixed findings have been observed in later studies 18 . A recent partial sleep deprivation experiment in adolescents has found that sleep loss affects caloric consumption only in the evening during the extended period of wakefulness 19 , suggesting that sleep loss may increase caloric consumption merely because of the increased opportunities to eat. Previous studies have shown that the impact of sleep loss on eating in humans is stronger when food rewards are hyperpalatable, i.e., calorie-dense, high-fat, and high-carbohydrate, with the enhanced impact not fully explained by increased hunger 20 , 21 , 22 . These findings suggest that sleep deprivation may affect both homeostatic eating and hedonic eating; the former is dependent on hunger, and the latter is driven by the palatability of food with or without the presence of homeostatic needs 23 .

There are times when we eat for neither hunger nor pleasure. For instance, “I can’t resist eating everything on my plate even when I feel full and satiated” and “I find myself munching on snacks whenever I am watching television” exemplify habitual control of eating . Habitual control of behavior is defined as cue-elicited processes, learned via repeated associations, and is insensitive to outcome devaluation 24 . Although habitual control of eating is closely related to hedonic eating, the wanting of a food (i.e., a motivation to eat a food) and the liking of a food (i.e., the affective reactivity to a food) are well-differentiated constructs in appetite research with dissociable neurological substrates 25 . Although it is widely believed to influence the way we eat, habitual control of eating has been considered the culprit in problematic eating only recently 26 . As habitual control of eating is characterized by insensitivity to outcome devaluation, it is linked to overeating (eating despite the decreased value of food after satiation) and external eating (eating in response to environmental cues rather than hunger), both of which are frequently associated with obesity in observational research 27 . Indeed, individuals with a higher body mass index (BMI) were found to less sensitive to food reward devaluation than those with a lower BMI 28 .

Habitual control and goal-directed control are two parallel processes that compete or interact to direct behavior 24 , 29 , 30 . Although it may not be adaptive in the pursuit of goals, habitual control is more efficient than goal-directed control and may compete better or function more adaptively in circumstances in which cognitive resources are depleted 31 . Deficits in executive functioning following sleep loss are well-documented 32 . As such, sleep loss may tip the balance between habitual and goal-directed control in favor of more efficient, less effortful habitual control at times when cognitive resources are depleted following sleep loss. Indeed, observational research has shown that individuals who regain weight after weight loss have a shorter sleep duration than those who maintain weight loss 33 . It is possible that insufficient sleep might increase the likelihood of resorting to old eating and lifestyle habits, an indication of overreliance on habitual control of behavior. An experimental study has shown that one-night total sleep deprivation (TSD) biased responding to stimuli associated with the devalued outcome in the slips-of-action task, which the authors interpreted as overreliance on habitual control of behavior over goal-directed control 34 . However, the greater biased responding for a devalued outcome documented by Chen et al. 34 could also be explained by the confounding impact of sleep loss on information encoding 35 . Furthermore, greater responsiveness to a devalued outcome in the slips-of-action task could also be attributable to excessive goal-directed behavior rather than an overreliance on habitual control of behavior.

An increased desire to eat when one is exposed to food-associated cues is a widely observed process even in normal eaters 36 , whereas a continued desire to eat triggered by cues despite satiation is more strongly implicated in overeating and disordered eating 27 . The Pavlovian-instrumental transfer (PIT) paradigm can be used to dissociate these two processes and allows for the evaluation of the effects of sleep deprivation on each of these processes. The PIT is an experimental paradigm for examining the influence of conditioned stimuli (CS) on instrumental behavior 37 and has been used to evaluate the effect of cues on food-seeking behavior 38 , 39 , 40 . It can be used in conjunction with satiation procedures to evaluate cue-elicited food-seeking behavior in the presence and the absence of satiation. The transfer effects derived from the PIT paradigm refer to the phenomenon where the presence of CS paired with an appetitive outcome (CS+) enhances an instrumental action. There are two types of transfer effects. Specific transfer refers to the increased instrumental responding in the presence of an outcome-specific CS that signals the outcome previously trained with the instrumental action via instrumental conditioning, i.e., response-outcome (R-O) learning. Increased responding can be expressed in the increased preference for the instrumental action (e.g., greater percentage of the choice of action against alternative actions) or the increased frequency of the action; sometimes it can also be expressed in suppressed responding for the alternative actions 37 . A real-life example would be an increased desire to order pizza after seeing a pizza commercial. General transfer refers to the increase in the vigor of an instrumental action in the presence of a general CS that signals an outcome that is not previously trained with the instrumental action but generates a similar motivational state, e.g., an approach action. For instance, an increase in approach behavior towards food in the presence of a CS associated with other food. A real-life example would be an increased desire to order pizza after seeing a non-pizza food commercial.

Taken together, habitual control of food-seeking behavior, defined as being cue-elicited and insensitive to satiation, could be a potential mechanism underlying the relationship between sleep deprivation and increased risks of obesity. However, no previous studies have examined the effects of sleep deprivation on these processes. Hence, the present study aimed to evaluate whether sleep deprivation had an impact on the effects of Pavlovian cues and satiation on food-seeking behavior, measured with the PIT paradigm, using a randomized, within-individual crossover experimental design in which participants partook in one-night TSD and one-night normal sleep duration (NSD). It was hypothesized that participants’ instrumental behavior would be more strongly affected by cues, i.e., greater specific and general transfer effects, and less sensitive to satiation following one-night TSD compared to NSD.

Descriptive statistics

One-hundred-twenty-one adults who met the habitual sleep duration criterion confirmed by actigraphy enrolled in the experiment. Twenty-five of them dropped out before completing the experiment. Data collection ended after 96 participants had completed the experiment, as re-activated COVID-related social distancing measures led to a halt in further data collection. Table 1 presents the sample’s demographic and clinical characteristics. Eighteen participants pressed no keys during the PIT or failed to learn the response-outcome (R-O) or the stimulus-outcome (S-O) associations. Their data were thus excluded from the main analysis, leaving 78 participants. The excluded participants did not differ from those included in any of the demographic or clinical variables.

Table 2 presents the differences in the measures of food liking, hunger, stress, total caloric consumption during satiation, working memory, response inhibition, and delay discounting between the TSD and NSD before and after satiation (see Fig.  5 in “ Methods ” for experimental procedures). None of the variables differed between them except for subjective liking of the hyperpalatable food and stress. Participants liked the hyperpalatable food less and felt more stressed in TSD than NSD. As expected, subjective liking of most food items, hunger, and stress were significantly reduced after satiation. Unexpectedly, despite selecting food items based on similar prior ratings of subjective liking (see “ Methods ” section “ Selection of food rewards ”), subjective liking of the two hyperpalatable food (O1 and O3) was significantly lower than that of the two non-hyperpalatable food (O2 and O4) after tasting (paired t  = − 2.73, p  = 0.008). It should be noted that a larger than expected portion of data were missing on food liking, hunger, and stress due to a programming error resulting in missing data on these variables in the first 20 participants. Sporadic missing data on other variables were caused by non-systematic data file corruption for two participants.

Learning of response-outcome (R-O) knowledge

During the training phase (see Table 4 in “ Methods ” for detailed PIT procedures), one participant did not achieve R-O knowledge during the last block of instrumental training query trials in NSD, and five of them in TSD. The difference was not statistically significant ( X 2  = 0.83, p  = 0.36). All participants achieved S-O knowledge during the last block of Pavlovian training query trials in both conditions.

During the testing phase, query trials were presented to check whether participants remembered the R-O knowledge. All participants in NSD showed correct memory of the R-O knowledge, and three (4%) in TSD did not. Hence, the data from these three participants were excluded from the analysis, leaving data of 75 participants for the analysis of PIT data. The difference in R-O knowledge between NSD and TSD was not statistically significant ( X 2  = 2.06, p  = 0.36).

Specific transfer—biased responding by cue

Specific transfer was measured as the biased responding by cues in the PIT paradigm (see Table 4 in “ Methods ”), specifically, the differences in responding in the presence of CS- (no cue), CS-same (cue predicting the reward associated with the key), CS-different (cue predicting reward associated with the other key), CS-other-reward (cue predicting food rewards not associated with any key), and CS-no-reward (cue predicting the absence of reward). A 3-way (sleep × cue × satiation) repeated-measures ANOVA with the percentage of instrumental responses for the hyperpalatable food as the dependent variable was conducted. Evaluations of the residual plots suggested that the multivariate normality assumption was met. Three multivariate outliers were identified by evaluating the Mahalanobis distance, the distance between a data point and the center of a distribution, indicating unusual deviation from a multivariate normal distribution. Further evaluations found that these outliers were valid data points and removing them did not make noticeable differences to the pattern and significance of the results. Hence, the analysis of the full dataset was conducted and reported here.

The sleep × cue × satiation 3-way interaction was non-significant but marginal ( F [3,222] = 2.48, p  = 0.062). The satiation × cue 2-way interaction was significant ( F [3,222] = 2.75, p  = 0.044). The simple main effect of cue was significant before satiation ( F [2.68,199] = 6.48, p  = < 0.001) and after satiation ( F [2.58,191] = 5.71, p  = 0.002) As shown in Fig. 1 , participants preferred the non-hyperpalatable food when there was no cue (CS-)—less than 50% of responses were made to obtain the hyperpalatable food either before or after satiation. Before satiation, there was a significant elevation of biased responding for the hyperpalatable food in the presence of CS-same compared to CS- (Cohen’s d  = 0.42), CS-no-reward (Cohen’s d  = 0.42), and CS-different (Cohen’s d  = 0.39), supporting the presence of a specific transfer effect. After satiation, biased responding for the hyperpalatable food in the presence of CS-same was not significantly different from CS-, indicating that the specific transfer effect was sensitive to satiation. Nonetheless, the biased responding in the presence of CS-different was significantly suppressed compared to CS- (Cohen’s d  = − 0.41) and CS-same (Cohen’s d  = 0.32), indicating some degree of the effect of cues on biased responding even after satiation.

Percentage of Responding for the Signaled Hyperpalatable Food Before and After Satiation (Combined Sleep Conditions). Note. CS- = no cue. CS-same = cue predicting the reward associated with the key. CS-different = cue predicting reward associated with the other key. CS-no-reward = cue predicting the absence of reward. The box area covers the middle 50% of the data values. The upper whisker covers the top 25% of the data values and the lower whisker covers the bottom 25% of the data values. The values displayed in the boxes indicate the mean values. The satiation × cue 2-way interaction is significant, showing that the effect of cue is significant only before satiation. Significant post-hoc comparisons are indicated along with the mean difference and the 95% confidence interval. *Bonferroni-corrected p  < 0.05; ** p  < 0.01.

Specific transfer—rate of responding by cue

The rates of responding for the signaled food across different sleep × satiation conditions were further examined by two 3-way (sleep × cue × satiation) repeated-measures ANOVA, one with the rate of responding for the hyperpalatable food as the dependent variable, the other with the rate of responding for non-hyperpalatable food as the dependent variable. Evaluations of the residual plots suggested that the multivariate normality assumption was met with no extreme multivariate outliers. Hence, all data were analyzed.

The sleep × cue × satiation 3-way interaction on the rate of responding for the hyperpalatable food was significant ( F [3,222] = 2.83, p  = 0.039). The sleep × cue 2-way interaction effect was significant only before satiation ( F [3,222] = 3.58, p  = 0.015) but not after satiation ( F [3,222] = 0.18, p  = 0.909). Post-hoc comparisons showed that (see Fig.  2 ), before satiation and following NSD, the rate of responding for hyperpalatable food was significantly elevated in CS-same compared to CS- (Cohen’s d  = 0.32), CS-no reward (Cohen’s d  = 0.40), and CS-different (Cohen’s d  = 0.43), indicating the presence of a specific transfer effect in NSD. Contrarily, this effect was not observed in TSD.

Rate of Responding for Hyperpalatable Food by Sleep and Satiation Conditions. Note. TSD, total sleep deprivation. NSD, normal sleep duration. The box area covers the middle 50% of the data values. CS- = no cue. CS-same = cue predicting the reward associated with the key. CS-different = cue predicting reward associated with the other key. CS-no-reward = cue predicting the absence of reward. The upper whisker covers the top 25% of the data values and the lower whisker covers the bottom 25% of the data values. The values displayed in the boxes indicate the mean values. The sleep × cue × satiation interaction effect is significant. Significant post-hoc comparisons are indicated along with the mean difference and the 95% confidence interval. *Bonferroni-corrected p  < 0.05; ** p  < 0.01.

On the other hand, the sleep × cue × satiation 3-way interaction on the rate of responding for the non-hyperpalatable food was non-significant, but marginal ( F [3,222] = 2.49, p  = 0.061). The cue × satiation 2-way interaction was significant ( F [3,222] = 3.06, p  = 0.029). The effect of cue was significant before ( F [3,222] = 10.90, p  < 0.001) and after satiation ( F [2.71, 200] = 3.71, p  = 0.015). As shown in Fig. 3 , before satiation, there was not significant elevation in the rate of responding for the non-hyperpalatable food in the presence of CS-same, but suppressed rates of responding in the presence of CS-different compared to CS- (Cohen’s d  = − 0.63), CS-same (Cohen’s d  = − 0.38), and CS-no reward (Cohen’s d  = − 0.34). After satiation, the rate of responding in the presence of CS-different was significantly lower than that in CS-same (Cohen’s d  = − 0.33).

Cue × Satiation Interaction Effect on the Rate of Responding for Non-Hyperpalatable Food. Note. CS- = no cue. CS-same = cue predicting the reward associated with the key. CS-no-reward = cue predicting the absence of reward. CS-different = cue predicting reward associated with the other key. The box area covers the middle 50% of the data values. The upper whisker covers the top 25% of the data values and the lower whisker covers the bottom 25% of the data values. The values displayed in the boxes indicate the mean values. Significant post-hoc comparisons are indicated along with the mean difference and the 95% confidence interval. *Bonferroni-corrected p  < 0.05; ** p  < 0.01; *** p  < 0.001.

General transfer

General transfer was measured by the number of all key presses in the presence of CS-other reward (C3 and C4) compared to CS- and CS-no-reward. A 3-way (sleep × cue × satiation) repeated-measures ANOVA was conducted with the combined response vigor as the dependent variable. Evaluations of the residual plots suggested that the multivariate normality assumption was met with no extreme multivariate outliers. Hence, all data were analyzed. All interaction effects were non-significant, but the main effects of satiation ( F [1,74] = 23.34, p  < 0.001) and cue were significant ( F [1,74] = 71.47, p  < 0.001). As expected, response vigor was significantly suppressed after satiation (see Fig.  4 A; Cohen’s d  = − 0.26). Response vigor in the presence of CS-no-reward was significantly suppressed compared to CS- (Cohen’s d  = − 1.04) and C4 (see Fig.  4 B; Cohen’s d  = − 0.42).

Main effects of satiation and cue on response vigor. Note. CS- = no cue. CS-O3 = cue predicting another reward (hyperpalatable). CS-O4 = cue predicting another reward (non-hyperpalatable). CS-no-reward = cue predicting the absence of reward. Panel A indicates the main effect of satiation. Panel B indicates the main effect of Cue. The box area covers the middle 50% of the data values. The upper whisker covers the top 25% of the data values and the lower whisker covers the bottom 25% of the data values. The values displayed in the boxes indicate the mean values. Significant post-hoc comparisons are indicated along with the mean difference and the 95% confidence interval. *Bonferroni-corrected p  < 0.05; *** p  < 0.001.

Correlations between transfer effects and covariates

Both within- and between-individual correlations of specific transfer and general transfer effects with ratings of hunger, stress, liking of each of the food items, caloric consumption, go no go task, working memory, and delay discounting were examined. None of the correlations were significant after Bonferroni adjustments of p -values (see Table 3 ).

The present study examined the effects of one-night TSD and satiation on food-related Pavlovian-instrumental transfer effects using a within-individual randomized, crossover experimental design. Cue-elicited food-seeking responding was found to differ in different combinations of sleep and satiation conditions. This finding was novel. However, contrary to the hypothesis, specific transfer effects were found in NSD before satiation but not in TSD. Sleep by satiation interactions were not found for the general transfer effects.

The present findings did not support the hypothesis that one-night TSD would exacerbate habitual eating and lead to greater cue-elicited motivation for eating even after satiation. Rather, these findings appear to align with the literature on sleep and appetitive conditioning in animal samples. In a systematic review of animal studies, disruption of sleep following learning was found to impair the acquisition, consolidation, and extinction of appetitive behaviors when natural reinforcers such as food were employed 41 . And, this alteration was unlikely attributable to learning or memory deficits. As shown in a study of sleep deprivation in rats, sleep-deprived rats were found to achieve similar probabilities in performing a behavioral response to a rewarded outcome as well-rested rates, indicating similar degrees of learning/memory; however, only sleep-deprived rats exhibited decreased responding over time, suggesting an alteration in motivational processes 42 .

Chen et al. 34 was the only human study of the effect of sleep deprivation on appetitive conditioning, and although they found that one-night TSD increased biased responding for a devalued outcome using the slips-of-the-action task, they did not control for the impact of TSD on information encoding as learning was conducted after TSD. Indeed, in their Experiment 2 when learning occurred before sleep deprivation they did not find such an effect. Consistently, the present study found that sleep deprivation did not increase cue-elicited appetitive behavior. It should be noted that the research on the effects of sleep on reward conditioning in human is scarce. It is recognized that reward conditioning processes and habitual vs goal-directed processes are frequently regarded as a proposed mechanism underpinning appetite disorders in humans 43 . The present findings highlighted the role of sleep disruption in appetitive conditioning, but given the unexpected results, future research is needed to further disentangle the effects of sleep disruption on these processes.

Food-related specific transfer effects were quite consistently observed in previous studies 38 , 39 . Similarly, a specific transfer effect was found in NSD before satiation in the present study. The percentage of responding for the signaled food increased from 42% in CS- to 56% in CS-same and the specific transfer effect disappeared after satiation. A larger specific transfer effect was reported by Watson et al. 39 , who found greater than 30% increases in biased responding for the signaled food before satiation and this effect remained significant even after satiation. While the instrumental and Pavlovian training procedures in Watson et al. 39 and the present study were close to identical, the order of the non-cued and cued testing trials in Watson et al., was counterbalanced, with a break in between. In the present study, the non-cued trials were presented first, followed by the cued trials without a break. As the transfer tests were conducted in extinction with no rewards given immediately, the first-presented set of non-cued tests could have decreased the strengths of the R-O associations and led to decreased responding in the subsequent cued trials. Without a break, there could also be a fatigue effect on responding in the cued trials. This methodological artifact could have masked the specific transfer effects to some degree. Future studies presenting a randomized order of non-cued and cued trials could eliminate the order effect and better evaluate whether food-related specific transfer effects would be sensitive to satiation in different sleep conditions.

Although the PIT effects are conceptualized as a reflection of habitual control of food-seeking behavior in the present study, other theoretical accounts of the PIT effects have been proposed arguing that the PIT effects are goal-directed rather than habitually-controlled 44 . Recent theoretical developments also argue that overeating and other suboptimal behavior are goal-directed rather than habitually controlled 45 , 46 . Indeed, we found that the specific transfer effect observed in NSD was sensitive to outcome devaluation, consistent with the goal-directed account. The specific transfer effect being sensitive to outcome devaluation was also consistent with previous failed attempts to experimentally induce outcome-insensitive instrumental behavior 47 . These findings might suggest that future studies on the effect of sleep deprivation on the motivational aspect of eating behavior may not have to be theoretically limited to either the habit or goal-directed account, but to focus on the dissociable processes underlying cue reactivity and insensitivity to outcome devaluation. For instance, the double dissociation account of the PIT effects, i.e., conditioned cues and outcome devaluation could independently influence instrumental actions 48 , maps well onto the distinction between cue-elicited eating, which is normative, and eating despite satiation, which is more closely related to disordered eating and obesity. The use of the PIT paradigm allows for the examination of these two processes independently and interactively. The present study is the only study, to the best knowledge of the author, that has examined the effect of sleep deprivation on food-related PIT effects in human using a well-controlled within-individual experimental design.

Given the aforementioned order effect, the findings on the general transfer effect should be interpreted with caution. The absence of any interaction effects with sleep or satiation suggested that sleep conditions did not have any influence on the general transfer effect. The main effect of cue showed that the response vigor in the presence of CS-no reward was suppressed compared to CS-O4. It was a possibility that if the order effect was not present, an elevation of response vigor compared to that in CS- might be observed in the presence of CS-O4.

In the exploratory correlation analysis, specific transfer effects were not correlated with hunger, stress, liking of the food rewards, caloric consumption, response disinhibition, working memory, and delay discounting, suggesting that specific transfer was independent from homeostatic eating and hedonic eating, and was not affected by state levels of stress, response biases, or executive functioning. Quail et al. 38 found that trait levels of stress measured by the DASS were correlated with transfer effects. Although the DASS was employed in the present study, we screened out individuals with higher scores on the DASS and thus resulted in a restricted range of scores in the sample. Hence, correlations with DASS were not performed.

It should be noted that caloric consumption during satiation did not differ between TSD and NSD. Although this finding contradicted those from some previous studies showing that sleep deprivation led to increased caloric intake 15 , it was in line with other studies in which total caloric intake per day was not found to differ following sleep deprivation 49 . In particular, a recent study using a 5-night 6.5-hour partial sleep deprivation experimental design found a non-significant difference in total caloric consumption per day between the sleep deprivation and normal sleep duration conditions 19 . Duraccio et al. 19 found an increase in caloric consumption only during late evening, suggesting that the effects of sleep duration on caloric consumption might be modulated by the timing of food consumption. Similarly, a previous naturalistic, observational study found that habitual short sleep duration was associated with decreased consumption of calorie-dense food in breakfast 50 . The fact that the PIT and satiation procedures were conducted in the morning in the present study might explain why caloric consumption was not found to differ following sleep deprivation. The absence of a difference in caloric consumption in the present study also highlighted the complexity of the relationship between sleep deprivation and eating—timing of food consumption, food cues and availability, and food palatability, could modulate eating behavior following sleep deprivation. Examining sleep and caloric consumption alone may not capture important changes in the mechanisms underlying eating behaviors, such as cue-elicited mechanisms as shown in the present study.

The findings of this study should be interpreted considering the following. First, the current sample was overrepresented by young, healthy, normal-weight individuals who were likely not at risk of obesity and obesogenic eating behavior, the present findings may not be generalizable to less-healthy populations, who might be particularly susceptible to the effects of sleep deprivation. Moreover, 25 out of the 121 initially recruited participants dropped out from the experiment. Many of them withdrew from the study because they could not follow through with the normal sleep requirement during the washout period. Although these participants did not differ significantly on the demographic and clinical variables from those who completed the experiment, they could differ on other unmeasured variables. Second, the NSD condition was conducted in participants’ homes. Although sleep duration was validated by actigraphy, the differences in sleep environments across individuals and the commute to the laboratory in the morning to participate in the PIT tests could have introduced noise to the results. Third, the effect of satiation was tested with a within-individual design, meaning that the results of the post-satiation PIT test may be confounded by the effect of extinction that occurred during the first, pre-saturation PIT test. Similarly, the order effect of the non-cued and cued tests introduced noise to the interpretation of the specific and general transfer effects. Fourth, one-night TSD may not be representative of real-life patterns of chronic partial sleep deprivation. Future studies employing a multi-day partial sleep deprivation design will have greater ecological validities. Finally, measures of alertness and sleepiness were not obtained which would have allowed further evaluations of the relationship between these sleep-related constructs and transfer effects. Measures of the desire to eat the food rewards were not obtained either which was distinct from food liking. They could be compared to the findings of PIT effects to shed light on the convergence between subjective food-seeking motivation and PIT effects.

Despite the aforementioned limitations, this study constituted the first empirical test of the effects of one-night TSD on food-related PIT effects. Although the findings did not support the hypothesis, they suggested that sleep deprivation could alter the PIT effects. Future studies are needed to further examine the influence of sleep parameters on motivational processes underlying eating.

Transparency and openness

The study design, hypotheses were pre-registered with the Open Science Foundation prior to data collection ( https://osf.io/rk35y/?view_only=838f8c3125af4a2fac5600895775f21c ). The analysis plan was not pre-registered. The de-identified data and analytic codes were uploaded to an open repository 51 . All research materials are available for access upon request for research and review purposes. An ethical approval was obtained from the University Human Ethics Committee (Ref# EA1905006) prior to participant recruitment, and all recruited participants provided electronic informed consent prior to the commencement of the study procedures. Participants were compensated $50USD for participating in the experiment.

Participants

Generally healthy adults (aged 18–65) who were free of sleep disorders, psychiatric disorders, eating disorders, and mood disorders and had normal habitual sleep patterns (7–9 h of sleep per night from 22:00 to 08:00) were recruited to the study to minimize any confounding impact of mental or physical illness on eating and to ensure that the participants could follow the experimental sleep manipulation procedures. Specific inclusion criteria were (a) aged 18–65 years and (b) having habitual sleep of seven to nine hours per night between 22:00 and 08:00, confirmed by seven days of actigraphy. Exclusion criteria were (a) having elevated sleep disturbances, indicated by a score of 7 or above on the Chinese version of the Pittsburgh Sleep Quality Index (PSQI) 52 , 53 ; (b) having evening chronotypes indicated by a score < 42 on the morningness–eveningness questionnaire (MEQ) 54 ; (c) having any self-reported current or history of psychiatric disorders including eating disorders, obsessive-compulsive disorders, substance abuse, bipolar disorder, or severe mental illnesses; (d) having any eating disorder symptoms indicated by a score > 16.5 on the Eating Disorders Diagnostic Scale (EDDS) 55 ; (e) having severe depressive or anxiety symptoms indicated by a score > 12 on the depressive subscale or a score > 8 on the anxiety subscale of the Depressive Anxiety and Stress Scales (DASS) 56 ; (f) having any medical conditions that could be worsened by one-night TSD; g) currently taking any psychiatric medications; (h) currently trying to lose weight; and, (i) having allergies or aversion to the experimental food stimuli. Additionally, women who scored > 30 on the shortened premenstrual assessment scale 57 , indicative of premenstrual syndrome, were scheduled for the study during the follicular phase of their menstrual cycles to minimize the confounding effects of hormonal changes on appetite 58 .

Screening and confirmation of habitual sleep duration

Participants completed an online survey consisting of the screening measures. Those who met the initial inclusion criteria were provided with an actigraph (ActiGraphTM model GT9X Link) to confirm their habitual sleep patterns. Those who met all the inclusion and exclusion criteria were invited to enroll in the experiment.

Experimental sleep manipulation

A within-individual randomized crossover design with a minimum three-day washout period was employed (see Fig. 5 ). Participants were randomized to begin with either the TSD or the normal sleep duration (NSD) condition, followed by at least 3 days of washout prior to the other sleep condition. During TSD, participants were settled in an approximately 350 square-feet room decorated like a living room and ensured to be awake until the next morning by research assistants. They could use their laptops and engage in any activity that did not involve vigorous physical activity. Participants were not allowed to consume any alcohol, nicotine-containing substances, caffeine, or to nap during the study procedures, although they could eat self-prepared snacks in the laboratory until 04:00. The NSD condition was conducted in the participants’ homes. Participants were instructed to sleep for seven to nine hours between 22:00 and 07:00, with the duration of sleep confirmed by actigraphy. All participants wore an actigraph throughout the study to cross-validate the sleep manipulation, especially during the washout period and in the NSD condition.

Experimental Procedures. Note. PIT—Pavlovian-instrumental transfer paradigm. 3B—3-back working memory test. GNG—Go no go test. DD—delay discounting task. The sequence of the two conditions was randomized.

Pavlovian-instrumental transfer (PIT) paradigm

As shown in Fig. 5 , the PIT employed in the present study consisted of the instrumental and Pavlovian training phases and the two testing phases with satiation in between. The PIT training phases were completed prior to sleep manipulation to ensure that any differences in the PIT tests were not due to inadequate or unsuccessful learning associated with sleep loss. The PIT testing phases were completed in the following morning at the same time in both conditions to control for circadian influences. Participants also completed ratings of hunger, stress, and food liking at the beginning of the PIT testing phase, and completed tests of working memory, response inhibition, and delay discounting, at the end of the PIT testing phase. The PIT paradigm was programmed using the PsychoPy v3.2 software package and administered to each participant individually on a desktop computer with a 24-inch monitor while the participant sat in a quiet room. Participants were told that they would be presented with food pictures and actual food rewards and should follow the on-screen instructions to obtain the food rewards (instructions included in Supplementary 1 ). They were asked to not consume any food within four hours prior to the PIT training and testing phases. Participants were instructed to use their dominant hand to press the keys and pick up the food items throughout the PIT. The food items were placed to the side of the computer monitor corresponding to the participants’ dominant hand.

The classic PIT paradigm is widely used in animal and human studies to model the influence of Pavlovian cues, i.e., conditioned cues that become associated with an outcome via classical conditioning (stimulus-outcome [S-O] association), on instrumental responses learned via operant conditioning (response-outcome [R-O] association) 37 . The transfer effects derived from the PIT paradigm refer to the phenomenon that the presence of Pavlovian cues enhances the probability and/or the vigor of instrumental responses. There are two types of transfer effects. The specific transfer effect refers to the increase in the probability and/or the frequency of performing an instrumental response by the presence of the conditioned cue signaling the outcome associated with the instrumental response. The general transfer effect refers to the increase in the vigor of an instrumental response by the presence of cues signaling an outcome that is not associated with the instrumental response but generates a similar motivational state.

The procedures of the PIT paradigm used in the present study is presented in Table 4 . The number of key presses (R1 and R2) in response to the presence of CS- (no cue), CS-same (cue predicting the reward associated with the key), CS-different (cue predicting reward associated with the other key), CS-other-reward (cue predicting food rewards not associated with any key), and CS-no-reward (cue predicting the absence of reward) were recorded. There were 10 non-cued trials (CS-) and 50 cued trials (10 presentations of each of the five Pavlovian cues (C1-C5). The 10 non-cued trials were presented first, followed by 50 cued trials presented in a random order. Specific transfer was measured by biased responding in the presence of CS-same compared to CS-no-reward, i.e., the percentage of R1 or the rate of R1 responses in the presence of C1 (conditioned cue associated with O1) compared to C5 (control cue associated with no reward) for high-palatability food and the percentage of R2 or the rate of R2 in the presence of C3 (conditioned cue associated with O2) compared to C5 for low-palatability food. General transfer was measured by the number of key presses (combined R1 and R2) in the presence of C3 and C4 (CS-other-reward) compared to CS- and CS-no-reward.

The PIT paradigm used in the present study was largely similar to that in Watson et al. 39 and Quail et al. 38 with two exceptions. In Watson et al. 39 , there was only one CS-other-reward, which was associated with a non-hyperpalatable food reward. In the present study, there were two CS-other-reward, one hyperpalatable and one non-hyperpalatable, to control for food palatability. Because hyperpalatable food is suggested to have properties similar to addictive drugs and might induce greater motivational changes 59 , food palatability was counterbalanced in the instrumental and Pavlovian training phases. And hence, there were a total of four food rewards, two hyperpalatable (O1 and O3) and two non-hyperpalatable (O2 and O4) in the present study. Additionally, both food rewards were satiated during outcome devaluation. Rather than satiating on one food reward and examining the biased selective responding for one food reward, both food rewards were satiated during outcome devaluation and the rate of responding for the hyperpalatable food and that for the non-hyperpalatable food were examined to evaluate transfer effects associated with the two foods with different palatability.

Satiation was conducted to devalue the food rewards. Participants completed the PIT testing phase twice, once before and once after satiation. During satiation, each participant was given an excess of the two food rewards used in the instrumental training phase to consume in the next 20 min. They were told they could eat as much as they wanted and that they could use their phones during the task. Caloric consumption was recorded in each sleep condition.

Selection of food rewards

Two hyperpalatable foods and two low-palatability foods were chosen for each participant after controlling for their subjective liking of the food. The definition of hyperpalatable food was adopted from Fazzino et al. 60 , characterized by food composed of high (1) fat and sodium, (2) fat and sugar, or (3) carbohydrates and sodium. During screening, participants were asked to rate their liking of 18 food items (9 hyperpalatable, 9 non-hyperpalatable) on a 10-point Likert scale (1 = dislike it; 10 = love it). The two hyperpalatable and two non-hyperpalatable food items that received similar liking ratings (with a score of at least 3, indicating “don’t mind eating it”) for a participant were then used as the food rewards for that participant, to control for the impact of liking of the food on the PIT performance. Because the study adopted a within-subject design, each participant had to complete the PIT twice. Hence, two sets of food rewards were used to prevent carryover effects.

Screening measures

The depression and anxiety subscale of the DASS-21 was used in the present study for screening 56 . The validated Chinese version of the DASS-21 was used 61 . The depression and anxiety subscales each consisted of seven items assessing depressive and anxiety symptoms on a 4-point scale from 0 to 3; higher scores indicated greater severity of symptoms. Those who scored above mild levels of depressive (> 13) and anxiety symptoms (> 9) were excluded from the study. The DASS-21 had excellent internal consistency in the screening sample (Cronbach’s α = 0.93).

The EDDS was used to screen out individuals with significant levels of eating disorders symptoms 55 . The EDDS-DSM-5 version was used, and it was translated from English to Chinese using the forward and backward translation method. It consisted of 23 items assessing cognitive and behavioral symptoms of BED, BN, and AN. The composite score > 16.5, computed from summing items 1–17, was a validated cutoff for indicating the presence of disordered eating psychopathology 55 . Hence, those who scored 16.5 or above were excluded from the study. The EDDS had acceptable internal consistency in the screening sample (Cronbach’s α = 0.75).

The MEQ 54 was used to screen out participants who had evening chronotypes and could have difficulty following through with the experimental sleep manipulation procedures. It was translated from English to Chinese using the forward and backward translation method. Based on the overall score, respondents were categorized into five chronotypes: definite evening (scored 16–30), moderate evening (31–41), intermediate (42–58), moderate morning (59–69), and definite morning (70–86). Participants who were classified as having either definite evening or moderate evening chronotypes were excluded from the study. The MEQ had good internal consistency in the screening sample (Cronbach’s α = 0.82).

The shortened PAF consisted of 10 items assessing premenstrual symptoms of female participants’ last menstruation cycle 57 . The total score indicated different levels of symptom severity: absent (10), mild (11–29), moderate (30–44), and severe/extreme (45–60). Participants who scored above 30 were scheduled for the study during the follicular phase of their menstrual cycles to minimize the confounding effects of hormonal changes on appetite 58 . The PAF had good internal consistency in the screening sample (Cronbach’s α = 0.88).

The PSQI, a widely used self-report measure of sleep quality consisting of 19 questions, was used to screen out individuals who had sleep disturbances. Seven component scores were derived from the PSQI which formed an overall sleep quality index in the range of 0–21; higher scores indicate worse sleep quality 52 . The validated Chinese version of the PSQI was used and the cutoff score of 7 was used to screen out individuals with potential sleep disorders 53 . The PSQI had acceptable internal consistency in the screening sample (Cronbach’s α = 0.65).

Measures of covariates

Rating of hunger, stress, and food liking.

Participants rated their hunger and stress on visual analog scales (0–100) at the beginning of the PIT testing phases. They then reported their subjective liking of the food items on offer on visual analog scales (0–100) at the end of the PIT.

Working memory

The 3-back task 62 was used to assess working memory. Participants were instructed to monitor a series of letters displayed in the center of the computer screen. They were told to press the y key when the letter presented was the same as the one displayed three trials prior and to press the u key when it was not the same. Participants were provided with feedback to indicate the correctness of their response: the letter turned green when the response was correct and red when it was incorrect. Participants completed two blocks of 20 trials. The percentage of correct trials was used as an index of working memory.

Response disinhibition

The go-no-go task 63 was used to assess motor disinhibition. Participants learned which numbers were a “go” signal and which were a “no go” from feedback. Four blocks of 25 trials (5 go trials) were conducted. The percentage of trials with hit and correct rejection responses was used as an index of inhibition.

Delay discounting

The delay discounting task with hypothetical monetary rewards and the standard double-limit algorithm was used 64 . The discounting rate, k , was calculated and used as an index of a preference for immediate vs. remote rewards.

Data analysis

The study’s primary hypothesis involved multiple within-subject factors with multiple levels, including sleep, cues, and satiation. Hence, repeated-measures ANOVAs were used to examine the primary hypotheses. Because interactions were involved, type III sum of squares were used. Examination of multivariate outliers, normality, and sphericity were conducted to evaluate if the data met the assumptions required for repeated-measures ANOVAs and if statistical adjustments were needed. The Greenhouse-Geisser correction was applied when the sphericity assumption was violated. Bonferroni-adjusted p-values were reported for post-hoc comparisons. All analyses were conducted using RStudio version 2023.03.1.

Specific transfer effects

The effects of sleep and satiation on specific transfer effects were first examined using a 3-way (sleep × cue × satiation) repeated-measures ANOVA with the percentage of instrumental responses for the hyperpalatable food as the dependent variable. Sleep had two levels: TSD vs NSD. Cue had four levels: CS-, CS-no-reward, CS-same, and CS-different. Satiation had two levels: before vs after satiation. Given that an increased percentage of instrumental responses could be attributable to suppressed responding in the presence of the cue predicting no reward rather than elevated responding in the presence of the cue predicting specific food reward, two additional 3-way (sleep × cue × satiation) repeated-measures ANOVAs were conducted, one with the rate of responding for the hyperpalatable food as the dependent variable, and the other with the rate of responding for the non-hyperpalatable food as the dependent variable.

General transfer effects

The general transfer effect was determined as the change in response vigor, computed by summing the number of both key presses by trial, in the presence of CS-other-reward compared to CS- and CS-no-reward. The effects of sleep and satiation on the general transfer effect were examined using a 3-way (sleep × cue × satiation) repeated-measures ANOVA.

Exploratory correlation analysis

To explore whether transfer effects were related to hunger, stress, liking of each food reward, response disinhibition, working memory, delay discounting, and caloric consumption, their correlations with transfer effects were examined. An index of specific transfer effect was formed by subtracting the rate of responding in the presence of CS-same by that in the presence of CS-. An index of general transfer effect was formed by subtracting the combined response vigor in the presence of CS-other-reward compared to CS-. Because participants completed these measures multiple times, both within-individual correlations and between-individual correlations were examined. The Bonferroni-adjusted significance level was used (0.05/10 = 0.005).

Sample size calculation

A medium effect size (partial eta-square = 0.07) was observed for the effect of one-night TSD on behavioral responsiveness to the devalued outcome in Chen et al. 34 . Although a different test, i.e., the PIT, was used in the present study, a medium effect size was the closest estimate of the expected effect size. Hence, a priori power analysis was conducted to estimate the minimum N needed for a medium effect for the primary hypothesis (partial η2 = 06). Given that multiple within-individual factors were involved in the analysis, the power analysis was conducted using the most conservative tests for complex within-individual factors, repeated-measures MANOVA. Using G*Power for a within-subject factor in repeated-measures MANOVA, N = 62 is required for β = 0.90 with α = 0.05 and ρ = 0.3. Considering that some participants might not follow through with all the experimental procedures or successfully learn the S-O and R-O associations in the PIT paradigm, the target N was set at 100.

Data availability

Data and analysis codes are available on https://doi.org/10.25442/hku.21904359 .

Chattu, V. K. et al. The global problem of insufficient sleep and its serious public health implications. Healthcare 7 , 1 (2018).

Article   PubMed   PubMed Central   Google Scholar  

Hirshkowitz, M. et al. National Sleep Foundation’s sleep time duration recommendations: Methodology and results summary. Sleep Health 1 , 40–43 (2015).

Article   PubMed   Google Scholar  

Krueger, P. M. & Friedman, E. M. Sleep duration in the United States: A cross-sectional population-based study. Am. J. Epidemiol. 169 , 1052–1063 (2009).

Reis, C. et al. Sleep duration, lifestyles and chronic diseases: A cross-sectional population-based study. Sleep Sci. 11 , 217 (2018).

Seidell, J. C. & Halberstadt, J. The global burden of obesity and the challenges of prevention. Ann. Nutr. Metab. 66 , 7–12 (2015).

Article   CAS   PubMed   Google Scholar  

Cappuccio, F. et al. Meta-analysis of short sleep duration and obesity in children and adults. Sleep 31 , 619–626 (2008).

Owens, J. et al. Insufficient sleep in adolescents and young adults: An update on causes and consequences. Pediatrics 134 , e921–e932 (2014).

Patel, S. R. & Hu, F. B. Short sleep duration and weight gain: A systematic review. Obesity 16 , 643–653 (2008).

Chaput, J.-P. Sleep patterns, diet quality and energy balance. Physiol. Behav. 134 , 86–91 (2014).

Benedict, C. et al. Acute sleep deprivation enhances the brain’s response to hedonic food stimuli: An fMRI study. J. Clin. Endocrinol. Metab. 97 , 443 (2012).

Article   Google Scholar  

Calvin, A. D. et al. Effects of experimental sleep restriction on caloric intake and activity energy expenditure. Chest 144 , 79–86 (2013).

Chapman, C. D. et al. Acute sleep deprivation increases food purchasing in men. Obesity 21 , E555–E560 (2013).

Demos, K. E. et al. The effects of experimental manipulation of sleep duration on neural response to food cues. Sleep 40 , 11 (2017).

St-Onge, M. P. et al. Sleep restriction leads to increased activation of brain regions sensitive to food stimuli. Am. J. Clin. Nutr. 95 , 818–824 (2012).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Al-Khatib, H. K., Harding, S. V., Darzi, J. & Pot, G. K. The effects of partial sleep deprivation on energy balance: A systematic review and meta-analysis. Eur. J. Clin. Nutr. 71 , 614–624 (2017).

Dzaja, A. et al. Sleep enhances nocturnal plasma ghrelin levels in healthy subjects. Am. J. Physiol. Endocrinol. Metab. 286 , 963–967 (2004).

Schmid, S. M., Hallschmid, M., Jauch-Chara, K., Born, J. & Schultes, B. A single night of sleep deprivation increases ghrelin levels and feelings of hunger in normal-weight healthy men. J. Sleep Res. 17 , 331–334 (2008).

Felső, R. et al. Total sleep deprivation decreases saliva ghrelin levels in adolescents. J Sleep Res. 32 , 13746. https://doi.org/10.1111/JSR.13746 (2022).

Duraccio, K. M. R. et al. Losing sleep by staying up late leads adolescents to consume more carbohydrates and a higher glycemic load. Sleep 45 , 3 (2022).

Greer, S. M., Goldstein, A. N. & Walker, M. P. The impact of sleep deprivation on food desire in the human brain. Nat. Commun. 4 , 1–7 (2013).

Hogenkamp, P. S. et al. Acute sleep deprivation increases portion size and affects food choice in young men. Psychoneuroendocrinology 38 , 1668–1674 (2013).

St-Onge, M. P., Wolfe, S., Sy, M., Shechter, A. & Hirsch, J. Sleep restriction increases the neuronal response to unhealthy food in normal-weight individuals. Int. J. Obes. 38 , 411–416 (2014).

Lowe, M. R. & Butryn, M. L. Hedonic hunger: A new dimension of appetite?. Physiol. Behav. 91 , 432–439 (2007).

Gardner, B. A review and analysis of the use of ‘habit’ in understanding, predicting and influencing health-related behaviour. Health Psychol. Rev. 9 , 277–295 (2015).

Finlayson, G., King, N. & Blundell, J. E. Liking vs wanting food: Importance for human appetite control and weight regulation. Neurosci. Biobehav. Rev. 31 , 987–1002 (2007).

Pierce-Messick, Z. & Corbit, L. H. Problematic eating as an issue of habitual control. Prog. Neuropsychopharmacol. Biol. Psychiatry 110 , 110294 (2021).

Feig, E. H., Piers, A. D., Kral, T. V. E. & Lowe, M. R. Eating in the absence of hunger is related to loss-of-control eating, hedonic hunger, and short-term weight gain in normal-weight women. Appetite 123 , 317–324 (2018).

Horstmann, A. et al. Slave to habit? Obesity is associated with decreased behavioural sensitivity to reward devaluation. Appetite 87 , 175–183 (2015).

Dickinson, A. Actions and habits: The development of behavioural autonomy. Philos. Trans. R. Soc. Lond. B Biol. Sci. 308 , 67–78 (1985).

Article   ADS   Google Scholar  

Perez, O. D. & Dickinson, A. A theory of actions and habits: The interaction of rate correlation and contiguity systems in free-operant behavior. Psychol. Rev. 127 , 945–971 (2020).

Corbit, L. H. Understanding the balance between goal-directed and habitual behavioral control. Curr. Opin. Behav. Sci. 20 , 161–168 (2018).

Jackson, M. L. et al. Deconstructing and reconstructing cognitive performance in sleep deprivation. Sleep Med. Rev. 17 , 215–225 (2014).

Yannakoulia, M. et al. Sleep quality is associated with weight loss maintenance status: The MedWeight study. Sleep Med. 34 , 242–245 (2017).

Chen, J. et al. Sleep deprivation promotes habitual control over goal-directed control: Behavioral and neuroimaging evidence. J. Neurosci. 37 , 11979–11992 (2017).

Boddez, Y., Buabang, E. K., Zenses, A.-K. & Descheemaeker, M. Commentary: Sleep deprivation promotes habitual control over goal-directed control: Behavioral and neuroimaging evidence. Front. Behav. Neurosci. 12 , 1–3 (2018).

Stojek, M. K., Fischer, S. & MacKillop, J. Stress, cues, and eating behavior. Using drug addiction paradigms to understand motivation for food. Appetite 92 , 252–260 (2015).

Cartoni, E., Balleine, B. & Baldassarre, G. Appetitive Pavlovian-instrumental transfer: A review. Neurosci. Biobehav. Rev. 71 , 829–848 (2016).

Quail, S. L., Morris, R. W. & Balleine, B. W. Stress associated changes in Pavlovian-instrumental transfer in humans. Q. J. Exp. Psychol. 70 , 675–685 (2017).

Watson, P., Wiers, R. W., Hommel, B. & De Wit, S. Working for food you don’t desire. Cues interfere with goal-directed food-seeking. Appetite 79 , 139–148 (2014).

Watson, P., Wiers, R. W., Hommel, B., Gerdes, V. E. A. & de Wit, S. Stimulus control over action for food in obese versus healthy-weight individuals. Front. Psychol. 8 , 1–13 (2017).

Article   CAS   Google Scholar  

Kavaliotis, E. et al. The relationship between sleep and appetitive conditioning: A systematic review and meta-analysis. Neurosci. Biobehav. Rev. 144 , 105001 (2023).

Hanlon, E. C., Andrzejewski, M. E., Harder, B. K., Kelley, A. E. & Benca, R. M. The effect of REM sleep deprivation on motivation for food reward. Behav. Brain Res. 163 , 58–69 (2005).

Everitt, B. J. & Robbins, T. W. Drug addiction: Updating actions to habits to compulsions ten years on. Annu. Rev. Psychol. https://doi.org/10.1146/annurev-psych-122414-03345767,23-50 (2016).

Mahlberg, J. et al. Human appetitive Pavlovian-to-instrumental transfer: A goal-directed account. Psychol. Res. 85 , 449–463 (2021).

Buabang, E. K., Boddez, Y., Wolf, O. T. & Moors, A. The Role of goal-directed and habitual processes in food consumption under stress after outcome devaluation with taste aversion. Behav. Neurosci. https://doi.org/10.1037/bne0000439 (2022).

Moors, A., Boddez, Y. & De Houwer, J. The power of goal-directed processes in the causation of emotional and other actions. Emotion Rev. 9 , 310–318 (2017).

de Wit, S. et al. Shifting the balance between goals and habits: Five failures in experimental habit induction. J. Exp Psychol. Gen. 147 , 1043–1065 (2018).

Corbit, L. H. & Balleine, B. W. Instrumental and Pavlovian incentive processes have dissociable effects on components of a heterogeneous instrumental chain. J. Exp. Psychol. Anim. Behav. Process. 29 , 99–106 (2003).

Nedeltcheva, A. V. et al. Sleep curtailment is accompanied by increased intake of calories from snacks. Am. J. Clin. Nutr. 89 , 126–133 (2009).

Chan, W. S. Daily associations between objective sleep and consumption of highly palatable food in free-living conditions. Obes. Sci. Pract. 4 , 379–386 (2018).

Chan, W. S. Effects of sleep deprivation on habitual control of food-seeking: A randomized crossover experiment—data and codes. In HKU Data Repository. Dataset (2013).

Buysse, D. J., Reynolds, C. F., Monk, T. H., Berman, S. R. & Kupfer, D. J. The Pittsburgh sleep quality index: A new instrument for psychiatric practice and research. Psychiatry Res. 28 , 193–213 (1989).

Tsai, P. S. et al. Psychometric evaluation of the Chinese version of the Pittsburgh Sleep Quality Index (CPSQI) in primary insomnia and control subjects. Qual. Life Res. 14 , 1943–1952 (2005).

Horne, J. A. & Ostberg, O. A self assessment questionnaire to determine Morningness Eveningness in human circadian rhythms. Int. J. Chronobiol. 4 , 97–110 (1976).

CAS   PubMed   Google Scholar  

Stice, E., Telch, C. F. & Rizvi, S. L. Development and validation of the eating disorder diagnostic scale: A brief self-report measure of anorexia, bulimia, and binge-eating disorder. Psychol. Assess. 12 , 123–131 (2000).

Lovibond, S. H. & Lovibond, P. F. Manual for the Depression, Anxiety and Stress Scale-21 items (DASS-21). In Manual for the Depression Anxiety & Stress Scales (1995).

Allen, S., McBride, C. & Pirie, P. The shortened premenstrual assessment form. J. Reprod. Med. 36 , 769–772 (1991).

Reed, S. C., Levin, F. R. & Evans, S. M. Changes in mood, cognitive performance and appetite in the late luteal and follicular phases of the menstrual cycle in women with and without PMDD (premenstrual dysphoric disorder). Horm. Behav. 54 , 185–193 (2008).

Gearhardt, A. N. & Schulte, E. M. Is food addictive? A review of the science. Annu. Rev. Nutr. 41 , 387–410 (2021).

Fazzino, T. L., Rohde, K. & Sullivan, D. K. Hyper-Palatable foods: Development of a quantitative definition and application to the US food system database. Obesity 27 , 1761–1768 (2019).

Moussa, M. T., Lovibond, P. F. & Laube, R. Psychometric Properties of a Chinese Version of the 21-Item Depression Anxiety Stress Scales (DASS21 ) (2001).

Owen, A. M., McMillan, K. M., Laird, A. R. & Bullmore, E. N-back working memory paradigm: A meta-analysis of normative functional neuroimaging studies. Hum. Brain Mapp. 25 , 46–59 (2005).

Wessel, J. R. Prepotent motor activity and inhibitory control demands in different variants of the go/no-go paradigm. Psychophysiology 55 , 1–14 (2018).

Johnson, M. W. & Bickel, W. K. Within-subject comparison of real and hypothetical money rewards in delay discounting. J. Exp. Anal. Behav. 77 , 129–146 (2002).

Download references

Acknowledgements

Albe SY Ng, Anna KW Lee, and Krisya Louie are acknowledged for their contributions to data collection. Wing Yee Cheng contributed to table and figure creation.

This work was supported by The University of Hong Kong Seed Fund for Basic Research [#201904159003].

Author information

Authors and affiliations.

Room 664, 6/F, Department of Psychology, The Jockey Club Tower, Centennial Campus, The University of Hong Kong, Pokfulam Road, Hong Kong, Hong Kong SAR

Wai Sze Chan

You can also search for this author in PubMed   Google Scholar

Contributions

WSC contributes to the conceptualization, methodology, software, formal analysis, resources, writing – original draft, writing – review & editing, and funding acquisition.

Corresponding author

Correspondence to Wai Sze Chan .

Ethics declarations

Competing interests.

The author declares no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Supplementary information., rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Chan, W.S. Effects of sleep deprivation on food-related Pavlovian-instrumental transfer: a randomized crossover experiment. Sci Rep 14 , 10029 (2024). https://doi.org/10.1038/s41598-024-60223-2

Download citation

Received : 11 August 2023

Accepted : 19 April 2024

Published : 01 May 2024

DOI : https://doi.org/10.1038/s41598-024-60223-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

• Search in PubMed
• Search in NLM Catalog
• Add to Search

Effect of food deprivation during early development on cognition and neurogenesis in the rat

Affiliation.

• 1 Department of Neurology, Columbia Presbyterian Hospital, New York, NY, USA.
• PMID: 15256180
• DOI: 10.1016/j.yebeh.2004.03.008

Food deprivation has been recognized as having pronounced beneficial effects in adult animals, increasing longevity, reducing seizure susceptibility, and enhancing resistance to neurotoxins. It is not known whether food deprivation in developing animals is neuroprotective or harmful. To evaluate the effects of food deprivation on brain development, we evaluated visual-spatial learning and memory and neurogenesis in the dentate gyrus of the hippocampus in food-deprived (FD) and well-fed (WF) rats. To induce food deprivation, pups were removed from their dams for 12 hours per day from Postnatal Day (P) 2 to P19. FD and WF rat pups were then subjected to status epilepticus (SE) induced by lithium-pilocarpine at P20. After SE, neurogenesis was measured, while in another group of P38 rats, learning and memory were evaluated using the Morris water maze. Food deprivation was found to reduce neurogenesis when assessed after the period of food deprivation. Although SE reduced neurogenesis in the WF animals, it had little effect additional to food deprivation on neurogenesis in the FD rats. Compared with the WF group, FD rats had a mild impairment in memory in the water maze testing after SE. Our study demonstrates that food deprivation during the neonatal period in rats is associated with a decrease in neurogenesis and mild impairment of visual-spatial memory. Although SE decreased neurogenesis in the WF group, in FD animals, SE did not reduce neurogenesis more than what was seen with food deprivation alone. Our results suggest that although food deprivation during early development reduces dentate gyrus neurogenesis, the reduced neurogenesis is not a major factor in cognitive impairment after SE in FD rats.

PubMed Disclaimer

Similar articles

• Effect of long-lasting serotonin depletion on environmental enrichment-induced neurogenesis in adult rat hippocampus and spatial learning. Ueda S, Sakakibara S, Yoshimoto K. Ueda S, et al. Neuroscience. 2005;135(2):395-402. doi: 10.1016/j.neuroscience.2005.06.065. Neuroscience. 2005. PMID: 16125851
• Spontaneous recurrent seizure following status epilepticus enhances dentate gyrus neurogenesis. Cha BH, Akman C, Silveira DC, Liu X, Holmes GL. Cha BH, et al. Brain Dev. 2004 Sep;26(6):394-7. doi: 10.1016/j.braindev.2003.12.006. Brain Dev. 2004. PMID: 15275703
• An enriched environment improves cognitive performance after early-life status epilepticus accompanied by an increase in phosphorylation of extracellular signal-regulated kinase 2. Wang CA, Lai MC, Lui CC, Yang SN, Tiao MM, Hsieh CS, Lin HH, Huang LT. Wang CA, et al. Epilepsy Behav. 2007 Nov;11(3):303-9. doi: 10.1016/j.yebeh.2007.06.016. Epub 2007 Sep 7. Epilepsy Behav. 2007. PMID: 17826356
• [Effect of various environmental factors on brain development]. Dmitrieva NI, Kassil' VG. Dmitrieva NI, et al. Arkh Anat Gistol Embriol. 1982 Sep;83(9):84-90. Arkh Anat Gistol Embriol. 1982. PMID: 6758735 Review. Russian. No abstract available.
• Nutrition and brain development. Winick M. Winick M. Curr Concepts Nutr. 1984;13:71-86. Curr Concepts Nutr. 1984. PMID: 6391836 Review. No abstract available.
• Differentiating kinds of systemic stressors with relation to psychotic-like experiences in late childhood and early adolescence: the stimulation, discrepancy, and deprivation model of psychosis. Vargas T, Damme KSF, Osborne KJ, Mittal VA. Vargas T, et al. Clin Psychol Sci. 2022 Mar;10(2):291-309. doi: 10.1177/21677026211016415. Epub 2021 May 27. Clin Psychol Sci. 2022. PMID: 35402089 Free PMC article.
• Neighborhood deprivation, prefrontal morphology and neurocognition in late childhood to early adolescence. Vargas T, Damme KSF, Mittal VA. Vargas T, et al. Neuroimage. 2020 Oct 15;220:117086. doi: 10.1016/j.neuroimage.2020.117086. Epub 2020 Jun 25. Neuroimage. 2020. PMID: 32593800 Free PMC article.
• Influence of catch up growth on spatial learning and memory in a mouse model of intrauterine growth restriction. Duran Fernandez-Feijoo C, Carrasco Carrasco C, Villalmazo Francisco N, Cebrià Romero J, Fernández Lorenzo JR, Jiménez-Chillaron JC, Camprubí Camprubí M. Duran Fernandez-Feijoo C, et al. PLoS One. 2017 May 24;12(5):e0177468. doi: 10.1371/journal.pone.0177468. eCollection 2017. PLoS One. 2017. PMID: 28542302 Free PMC article.
• Transgenerational loss and recovery of early learning ability in foraging predatory mites. Reichert MB, Christiansen IC, Seiter M, Schausberger P. Reichert MB, et al. Exp Appl Acarol. 2017 Mar;71(3):243-258. doi: 10.1007/s10493-017-0122-1. Epub 2017 Apr 13. Exp Appl Acarol. 2017. PMID: 28409405 Free PMC article.
• Effect of diet on the structure of animal personality. Han CS, Dingemanse NJ. Han CS, et al. Front Zool. 2015 Aug 24;12(Suppl 1):S5. doi: 10.1186/1742-9994-12-S1-S5. eCollection 2015. Front Zool. 2015. PMID: 28400852 Free PMC article. Review.

Publication types

• Search in MeSH

Related information

• PubChem Compound
• PubChem Compound (MeSH Keyword)
• PubChem Substance

Grants and funding

• NS027984/NS/NINDS NIH HHS/United States
• NS041595/NS/NINDS NIH HHS/United States
• NS044296/NS/NINDS NIH HHS/United States

LinkOut - more resources

Full text sources.

• Elsevier Science
• Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

How Sleep Deprivation Affects Your Metabolic Health

By Nicole Molumphy

Do you ever get those late-night cravings or feel like your hunger is never satiated? Research shows that consistent short sleep duration (less than 7 hours per night) can influence metabolic health, affect the function of appetite hormones, increase food cravings, and lead to a 38 percent increase in obesity in adults.

Consistent sleep deprivation can lead to altered functioning of the appetite hormones ghrelin and leptin . Ghrelin increases our appetite and is released by cells in the stomach lining. When your stomach is “growling,” that is the ghrelin hormone talking. Conversely, the hormone leptin, made by our fat cells, lowers our appetite. Studies have revealed that sleep deprivation can lead to increased ghrelin and decreased leptin, resulting in an overall experience of constantly being hungry.

“There are so many health conditions associated with poor sleep,” says Rob Oh, MD, family physician at the Veterans Affairs Health Care System in Palo Alto. “With chronic sleep deprivation, your metabolism becomes dysregulated, leading to cravings for processed foods. Also, you’re less likely to exercise, you feel more stressed, and you’re more likely to think poorly. The combination of these factors can lead to metabolic health problems like obesity and type 2 diabetes.”

Sleep Deprivation Increases Stress Hormones

Our cortisol levels are typically lowest near midnight and then increase towards waking hours, ultimately peaking around 9 a.m. Studies demonstrate that reoccurring poor sleep is associated with an altered cortisol secretion pattern .

For example, delaying your bedtime could lead to high cortisol levels in the middle of the day, rather than just in the morning. Sustained high levels of cortisol can lead to an increased amount of insulin in the blood , which promotes the accumulation of belly fat and has the potential to lead to prediabetes, type 2 diabetes, and other metabolic disorders.  An increase in cortisol levels during the day may induce prolonged feelings of stress, increased food cravings, and further insomnia—promoting a recurrent, cyclical pattern.

“Inadequate sleep disrupts hormone levels, which dysregulates one’s metabolism and makes individuals hungrier,” says Dr. Oh. ” When we are hungrier, we eat more, which leads to weight gain and potentially metabolic disease.”

If that’s not bad enough, cravings for ultra-processed foods, sugars, and alcohol become more of a tease with sleep deprivation. A possible mechanism for this added hunger is an increased activation of the endocannabinoid system , which is found throughout the body and controls several biological systems, including sleep, mood, and appetite.

Sleep Deprivation and Increased Risk of Diabetes

Research shows that a lack of sleep may also result in insulin resistance , a driving factor in prediabetes and type 2 diabetes . Insulin is a hormone made by the pancreas that regulates blood glucose levels. Insulin resistance occurs when cells in the liver, fat, and muscles do not respond well to insulin, and in turn, glucose is not taken up into the blood.

The exact mechanism behind the causal relationship between insufficient sleep and insulin resistance is still being studied, however, some studies note that increased inflammatory markers, like C-reactive protein and hormone dysregulation, such as cortisol may be involved.

Currently, prediabetes affects one in three adults in the US. While many mechanisms contribute to the development of prediabetes and type 2 diabetes, some of the more prominent ones include impaired cellular insulin sensitivity, modified gut microbiota, and overly-sensitive sympathetic nervous system (fight-or-flight) activation. When the sympathetic nervous system is activated, it signals the liver to release more glucose into the bloodstream, leading to higher blood sugar levels.

What Can We Do About It?

Unfortunately, research finds that “weekend recovery sleep” is not enough to bring your metabolism back into balance after a lack of sufficient sleep throughout the week. Instead, Dr. Oh recommends taking a holistic approach and examining not only your sleep habits but also your nutritional choices and workout routines to support the restoration of metabolic health.

For those with pre-diabetes and type 2 diabetes, a diet focused on whole foods and low refined carbohydrates is critical for weight loss and glucose control.

Specifically, Dr. Oh recommends a low carbohydrate diet, such as consuming 75 to 100 grams of carbohydrates per day, to control and balance blood sugar levels. He also recommends quick, 20-minute, high intensity workouts for time-efficient and effective exercises that promote metabolic health. Regarding supplements, Dr. Oh recommends magnesium to promote muscle recovery and relaxation before bedtime.

“Optimal sleep is so intertwined with athletic and exercise performance, brain health, and metabolic health,” says Dr. Oh. “Sleep really is a keystone area where many of us can do better and hence reducing our risk for metabolic diseases and lengthening our healthspan.”

You might also like

• Lifestyle Medicine Blog
• In the News
• Research Publications
• Movement & Exercise
• Healthful Nutrition
• Restorative Sleep
• Stress Management
• Social Engagement
• Cognitive Enhancement
• Gratitude & Reflection

Email (required) *

Example: Yes, I would like to receive emails from Lifestyle Medicine. (You can unsubscribe anytime)

Ohio State nav bar

The Ohio State University

• BuckeyeLink
• Find People
• Search Ohio State
• Becoming a Principal Investigator
• Find Funding
• Research Administration Tools
• Student Involvement
• Knowledge Enterprise Resources
• Research Commons
• Report Emergency or Animal Concerns
• Institutional Animal Care and Use Committee
• Animal Care and Use Policies and Guidelines
• Animal Study Team Requirements
• Per Diem Rates and Fees
• Lab Animal Handling and Technique Training
• Laboratory Animal Medicine Training
• Animal Facilities and Access
• Animal Transfer
• Animal Ordering or Imports
• Surgery and Technical Services
• Quality Assurance Assessment Program
• Accreditation
• Recombinant DNA, Biohazard and Animal Incident Reporting
• Institutional Biosafety Committee
• IBC Registration Process
• Biosafety Study Team Requirements
• Biosafety Protocols and Amendments
• Biosafety Policies and Guidelines
• Report Human Subjects Concerns
• Institutional Review Board
• IRB Member Rosters and Meeting Dates
• IRB Submission Types
• Human Research Study Team Requirements
• Human Subject Research Policies and Guidance
• IRB Time to Approval
• Informed Consent, Assent and Parental Permission
• HIPAA and Human Subjects Research
• Exempt Research
• Event Reporting
• Collaborative and Multi-Site Research
• Research vs. Quality Assurance/Improvement
• Quality Improvement Program
• Information for Research Participants
• Community-Engaged Research
• Clinical Trials Registration
• Information for IRB Members
• Responsible Conduct of Research
• Disclosure Process
• Tips for Disclosing
• What Not to Disclose
• Common Disclosure Scenarios
• Policies and Documents
• Foreign Talent Programs
• Conflict Approval Committee
• Controlled Substances
• Dual Use Research
• University Activities
• Export Control Regulations
• International Research Engagements
• Export Control Training
• Export Control Terms and Definitions
• Human Gene Transfer
• LabArchives - Electronic Lab Notebook
• Research Diving
• Avoiding Plagiarism
• Research Security Governance Board
• Roles and Responsibilities
• Proposal Development and Submission Guidelines
• Types of Sponsors
• Grants, Gifts and Contracts
• Working with Business and Industry Sponsors
• Working with Federal Agencies
• Other Support/Current and Pending Support
• Clinical Trials
• Uniform Guidance
• Institutional Information
• Direct Cost Categories
• Employment Benefits (Fringe) Rates
• Subcontracts for Budget Preparation
• Participant Support Costs
• Facilities and Administrative Costs
• Central Cost Share Request
• Submission Checklist
• Sponsored Program Officers
• Bayh-Dole Certification
• Award Acceptance and Initiation
• Award Management Roles and Responsibilities
• Award Terms and Conditions
• Cost Sharing Requirements
• Cost Transfers
• Personnel Appointments
• Payments to Research Subjects
• Absence or Transfer of Principal Investigator
• Patents and Copyrights
• Travel on Sponsored Projects
• Ordering in Workday
• Procurement Policies and Procedures
• Purchasing Basics
• Bid Opportunities
• Consultant Agreement Procedures
• Subcontracting on Sponsored Awards
• Procurement and Receipt of Radioactive Material
• Equipment and Property
• Software License Agreements
• Special Purchasing Needs
• Physical Facilities
• International Purchasing
• Award Closeout
• Corporate Partnerships at Ohio State
• Invention Disclosure and Assessment
• Licensing Your Technology
• Forming a Startup Company
• Peter Mohler
• Cynthia Carnes
• Animal Care and Use Program
• Research Compliance
• Responsible Research Practices
• Secure Research
• Sponsored Programs
• Byrd Polar and Climate Research Center
• Campus Chemical Instrument Center
• Campus Microscopy and Imaging Facility
• Center for Emergent Materials
• Center of Microbiome Science
• Center for RNA Biology
• Center for Quantum Information Science and Engineering
• Chronic Brain Injury Program
• Foods for Health
• Gene Therapy Institute
• Global Water Institute
• Infectious Diseases Institute
• Institute for Materials Research
• Institute for Optical Science
• Translational Data Analytics Institute
• Facts and Figures
• 150 Years of Excellence
• Awards and Honors
• Associate Deans for Research
• Submit an Event
• Research and Innovation Showcase 2022
• Research and Innovation Showcase 2021
• Research and Innovation Showcase 2020
• Research and Innovation Showcase 2019

Shape the Future of Food for Earth and Beyond

Witness the convergence of deep space exploration and food tech innovation as we celebrate Phase 3 of the Deep Space Food Challenge.   Co-hosted by NASA Centennial Challenges, Methuselah Foundation, and The Ohio State University, the Deep Space Food Symposium will celebrate the Deep Space Food Challenge, a multi-year international effort to develop sustainable food systems for long duration habitation in space, on the Moon, and on Mars. 

PROGRAM HIGHLIGHTS

Exclusive Tours* Get a hands-on look at the cutting-edge research and technology shaping the future of food production at the Ohio State University, including The George Washington Carver Science Park’s Starlab Ground Lab US and the Controlled Environment Agriculture Research Complex.

Please note that the tours will be in two sessions: 12:00pm-2:00pm or 2:00pm-4:00pm

Deep Knowledge Sharing* Join the Teams, Judges, and Industry SME’s for in-depth technical discussions regarding new innovative approaches to space food systems from DSFC Phase 3 teams, including Interstellar Lab, Solar Foods, SATED, Nolux, Kernel Deltech, and Enigma of the Cosmos.

VIP Dinner: The Future of Food @ FYR Restaurant* Join Chef Tyler Florence and FYR’s own Executive Chef Sebastian La Rocca and indulge in an evening where culinary excellence meets innovation, blending the finest of Ohio's local cuisine with a cutting-edge vision for the future of space food.

*Please note that OSU Facility Tours, Knowledge Sharing Session, and The Future of Food @ FYR are by invitation only. Apply for attendance here.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List

The effect of personal relative deprivation on food choice: An experimental approach

Sofie van rongen.

1 Consumption and Healthy Lifestyles Group, Wageningen University & Research, Wageningen, Gelderland, The Netherlands

Michel Handgraaf

2 Urban Economics Group, Wageningen University & Research, Wageningen, Gelderland, The Netherlands

Maaike Benoist

3 Human Nutrition and Health Group, Wageningen University & Research, Wageningen, Gelderland, The Netherlands

Emely de Vet

Associated data.

All data sets are available from the repository DANS, doi: 10.17026/dans-z8q-vcgd .

Growing evidence suggests that relative disadvantage is more relevant than absolute socioeconomic factors in explaining disparities in healthfulness of diet. In a series of pre-registered experiments, we tested whether personal relative deprivation (PRD), i.e. the sense that one is unfairly deprived of a deserved outcome relative to others, results in choosing more palatable, rewarding foods. Study 1 ( N = 102) demonstrated the feasibility and effectiveness of a game for inducing real-time experiences of PRD. Study 2 ( N = 287) showed no main effect of PRD condition on hypothetical food choices, but an interaction between chronic PRD and condition revealed that those in the PRD condition chose more rewarding foods when feeling chronically deprived. In Study 3 ( N = 260) the hypothesized main effect was found on real, non-hypothetical food choices: those in the PRD condition chose more rewarding foods, controlling for sensitivity to palatable food. Our results provide preliminary indications that the experience of being relatively deprived, rather than the objective amount or resources, may result in a higher preference for high-caloric and palatable foods. It may be suggested that efforts to reduce societal disparities in healthfulness of diet may need to focus on perceptions of injustice beyond objective inequalities.

Introduction

The association between socioeconomic status (SES) and diet quality is globally well established [ 1 , 2 ]. People living on a low income have unhealthier diets [ 3 ] and higher rates of diet-related diseases such as obesity [ 4 , 5 ] than people of higher SES. Dominant explanations for socioeconomic disparities in diet and obesity have focused on physical and economic food access. Unhealthy food outlets are more prevalent in low SES neighbourhoods [ 6 ], and it has been claimed that unhealthy foods are cheaper than healthy alternatives [ 7 ]. Importantly, having a low income or a low educational status in an absolute sense does not fully explain socioeconomic inequalities in diet. Inequality also comprises a relative aspect, i.e. earning less than others or being less educated than others. Relative disadvantage may even be more relevant than absolute factors like income. This is illustrated with epidemiological evidence showing a positive correlation between societal income inequality and obesity prevalence [ 8 , 9 ], even after controlling for absolute socioeconomic measures [ 10 , 11 ]. Additionally, growing evidence shows that subjective, relative wealth is indeed more predictive of health than objective, absolute socioeconomic indicators [ 12 ]. The relative deprivation hypothesis proposes that making upward comparisons has negative psychological consequences, leading to health compromising behaviour. Focusing on dietary behaviour specifically, a correlational study showed that the Yitzhaki index, a demographic measure of relative deprivation (income), was associated with self-reported behaviours such as less healthful food choices [ 13 ]. However, evidence for a causal relation between relative deprivation and diet quality at the proximate, individual level is lacking. In a series of experimental studies, we aimed to address this gap by experimentally inducing relative deprivation and investigating how it affects food choice behaviour.

A common conceptualization of subjective, individual-level relative deprivation is personal relative deprivation (PRD), which relates to feelings of frustration and resentment in response to the idea of being deprived of a deserved and desired outcome, stemming from upward comparisons with similar others [ 14 – 16 ]. Human concern for justice is a key prerequisite for the experience of relative deprivation [ 17 ]; a threat to one’s personal deservingness produces perceptions of injustice and unfairness [ 14 , 15 , 18 ]. PRD has been associated with various adverse outcomes, including depression [ 19 ], physical and mental health issues [ 20 ], but also gambling and other risk behaviours [ 21 , 22 ]. As a psychological mechanism, feelings of PRD have been theorized to result in a greater desire for immediate small rewards The reasoning for this, drawing on theories of justice motivation [ 23 ] and delay discounting [ 24 ], is that people who experience feelings of not being treated in the same way as others prefer immediate small rewards because of the need to feel that their deservingness concerns are being addressed [ 21 ]. If people lose their trust in a just world, then they might be more attracted to immediate gratification at the expense of longer-term, larger gains [ 23 , 25 ].

We posit that, in the current food context of easy food access and abundant choice, people experiencing PRD may similarly have a preference for foods that are immediately rewarding rather than beneficial for health, as a way to combat these negative experiences and restore the sense of personal deservingness. Although food may not be as rewarding as monetary rewards in response to deprivation of resources, recent research on resource scarcity-induced eating indicates that the human motivational system for food and money overlap [ 26 – 28 ]. Furthermore, the experience of PRD may be a precursor to a higher general sensitivity to reward, and it has previously been found that individual sensitivity to reward predicts a preference for high-calorie food [ 29 , 30 ]. Moreover, the motivational component of food reward is essentially driven by the brain’s appetitive system, largely the dopamine pathway, which facilitates ‘wanting’–the motivation to pursue a stimulus [ 31 ]. This pathway also mediates the motivation for risk behaviours like gambling [ 32 , 33 ]. Hence, PRD may evoke a pleasure-oriented preference for selecting high-fat or high-sugar, palatable foods, as these are, based on previous encounters, (implicitly) associated with immediate reward [ 34 ].

The present research links to a growing number of experimental studies that have demonstrated an effect of a subjective relative socioeconomic manipulation on high-caloric food preference and intake (see for reviews [ 35 , 36 ]. For example, Briers and Laporte [ 37 ] showed that a manipulation of relative income resulted in a higher selection of high versus low caloric dishes. Cheon and Hong [ 38 ] demonstrated that low subjective socioeconomic status (SSS) resulted in a greater preference for high-caloric foods (over fruits and vegetables). SSS was induced with a popular, much-used manipulation, i.e. an adapted version of the MacArthur Ladder of SSS [ 39 – 42 ]. Participants were asked to compare themselves with those who were at either the very top (low SSS condition) or the bottom (high SSS condition) of the ladder by indicating where they stood relative to these people, and to write about a hypothetical interaction with one these individuals.

Although relative income and SSS income involve a subjective evaluation of relative standing, they lack an important emotional component of relative deprivation. PRD is a more specific and emotionally laden concept of relative status that comprises both cognitive and affective responses to unfair outcomes, and has been shown to be a better predictor of health than relative SSS [ 43 ]. Whether PRD affects food preference as a proxy for diet quality remains to be tested, and with this research we aimed to answer this question.

One previous study provided first evidence for a causal effect of PRD on food selection, showing that induced feelings of PRD resulted in the selection of larger meal portions in a computerized portion selection task [ 44 ]. However, in that experiment, the PRD manipulation involved reading a hypothetical scenario unrelated to the food (receiving a smaller versus equal work bonus relative to colleagues) and the portion size selection was hypothetical in nature. The present study expanded on these first results in two main ways. First, it aimed to test the impact of real-time experience of PRD in resources on both hypothetical and actual food choices. Second, the resources earned were linked to food choices, as earnings served as resources to be spent on foods. This contributes to external validity because in actual life/natural environments eating is almost always a choice and foods are usually obtained with one’s resources (e.g. while grocery shopping), rather than by self-serving from free buffets, as commonly applied in experimental studies focusing on SES (although self-serving is a valuable measure in other regards, i.e. actual assessment of quantity consumed and portion control). In Study 1, we tested the feasibility of a self-developed card game for manipulating real-time experiences of PRD. In Study 2, we tested the effect of PRD on food preference in a hypothetical online food shopping setting. Following a pre-test, the available foods were categorized into rewarding and neutral foods. Study 3 was a conceptual replication of Study 2 in a lab-in-the-field setting, where a diverse community sample made real (non-hypothetical) food choices using the points earned. It was hypothesized that PRD would result in a higher selection of high-caloric and palatable (immediately rewarding) foods.

Study 1: Testing the PRD manipulation task

The aim of Study 1 was to examine whether experiences of PRD can be effectively induced using a self-developed card game. In conformity with a pilot study in which participants were relatively disadvantaged in a Monopoly game [ 45 ], our PRD manipulation involved playing a computer card game in which participants actively experienced earning fewer (PRD condition) versus equal (control condition) resources relative to a fictitious player.

Participants and procedure

On the basis of a power calculation (power of .90, medium effect size (Cohen’s d ) of .5), 172 participants aged between 18 and 70 years with fluency in English were recruited online via Prolific, an online participant recruitment platform ( www.prolific.com ). Four participants were excluded and substituted with new participants because they completed the study in a substantially short amount of time (i.e. below 6 minutes). Another exclusion criterion was incorrect answers on both the attention check items (none excluded). The analytic sample consisted of 102 (59%) females, 15 different nationalities (51% British), an average age of 34.83 ( SD = 11.39), and a range of educational backgrounds (24% no education/high school degree, 27% college/associate degree, 47% academic degree). All participants provided written informed consent at the start of the study. After the game, participants completed demographic measures and a “bogus” end task in which they allocated all of their earnings (30 points) amongst four food products (two healthy (radishes and carrots) and two unhealthy (waffles and pralines)), based on how much they would want the product. An analysis of this data showed that conditions did not differ in the amount of points allocated to healthy or unhealthy foods (see S3 Appendix for the specific results). Yet, as this task was under-developed at this stage and the main goal of this study was to test the PRD manipulation (for which a purpose for the earnings needed to be included), we consider these non-significant results unproblematic with respect to the ultimate outcome of interest (food choice). The Social Science Ethics Committee of Wageningen University approved the study.

PRD manipulation: Card game

The card game was designed to induce subjective experiences of relative deprivation, which specifically entail upward social comparisons leading to perceptions of being worse off and unfairly treated as well as feelings of resentment, dissatisfaction, and anger [ 14 , 21 ]. The card game was developed in Qualtrics, an online survey tool [ 46 ]. It was explained that the card game served as a way to earn points that were necessary for completing a subsequent task. The participants were led to believe that they were playing the game against a previous participant (a fictitious opponent) of the same gender and age, so as to induce an idea of a ‘similar other’. In each of 10 rounds, two different playing cards were presented (e.g. four of diamonds and nine of clubs). All participants were shown the exact same cards. Participants had to guess whether the number of a third card, supposedly drawn at random by the computer, would be between or not between the numbers on these cards.

After a practice round, participants were randomly assigned to either the PRD or the control condition by Qualtrics. Before the actual game started, participants read an additional instruction page about the point earnings, which differed for the two conditions. Participants in the PRD condition learned that they would receive fewer points than their (fictitious) opponent for each correct answer (i.e. 5 versus 10 points) and fewer bonus points than their (fictitious) opponent if they, rather than their opponent, had most points at the end of the 10 rounds (i.e. 25 versus 50 bonus points). The participants in the control condition learned that would earn the same as their opponent for each correct answer (i.e. 5 points) and an additional 25 bonus points for having most points after 10 rounds. After each round, participants were shown a bar chart depicting the interim score of points earned by themselves and their opponent. For example, after round 1, participants in the control condition earned 5 points, the same as their opponent who also earned 5 points, whereas participants in the PRD condition earned 5 points while their opponent earned 10 points. Unbeknownst to the participant, the ‘game’ was completely pre-programmed, such that all participants (and the fictitious opponent) had a total of six correct answers (in round 1, 3, 4, 5, 8, and 10) and so each participant ended with total score of (6x5 =) 30 points. As the earning of bonus points was based on having the most points at the end of the game, the distribution of bonus points also differed between conditions (i.e. 50 points for the opponent of participants in the PRD condition and no bonus points for participants and their opponent in the control condition). Fig 1 presents the final score screens of the PRD and the control condition.

The points earned did not objectively differ between conditions, conditions differed in the idea that the opponent earned much more (PRD condition) or just the same (control condition) for the same number of correct answers.

Assessment of experienced PRD

A 7-item scale was developed based on Callan, Shead, and Olson’s revised PRD scale [ 21 ] that asked about the perception of being deprived and unfairly treated (e.g. “I felt worse off when I compared myself with my opponent”), as well as feelings of resentment, dissatisfaction, and frustration (e.g. “I was frustrated when I saw how many points I earned compared to my opponent”). The answer scale ranged from 1 (strongly disagree) to 7 (strongly agree). To detect the number of components of the self-report instrument, a principal component analysis was conducted with orthogonal rotation (varimax). Examination of the scree plot and eigenvalues over 1 suggested the presence of one component that accounted for 71.24% of the variance. The scale was reliable, Cronbach’s alpha = .93. A mean score was calculated, with higher scores representing higher experiences of PRD. See Table A in S1 Appendix for the items.

Assessment of game experience

It was additionally explored whether participants’ liking of the game (i.e. how much fun the card game was; if they would like to play this game again) and their involvement (i.e. if they cared about the points earned; if they did their best; if they cared about the points their opponent earned). The answer scale ranged from 1 (strongly disagree) to 7 (strongly agree). An average score of the liking items was calculated (Cronbach’s alpha = .89), and the involvement items were analysed separately, as this scale was not reliable (Cronbach’s alpha = .56)

A t-test showed that the PRD condition ( M = 5.33, SD = 1.33) scored significantly higher than the control condition on the experienced PRD scale ( M = 2.37, SD = 1.03), t (170) = 16.29, p < .001, 95% CI [2.61–3.33], d = 2.49. Conditions did not differ significantly in game liking, t (170) = 1.60 p = .11. Conditions did not differ either in caring about their points, t (170) = 1.21 p = .23, or their opponent’s points, t (170) = 1.22 p = .23, or doing their best, t (170) = 0.45 p = .65. On average, participants rated a positive liking experience (M = 5.33, SD = 1.37) and reported that they cared about their points (M = 5.58, SD = 1.46) and their opponent’s points (M = 4.88, SD = 1.80) and that they did their best (M = 5.33, SD = 1.37). See Table B in S1 Appendix for means and SDs of items per condition.

This manipulation was considered successful, as conditions differed significantly in their PRD experience during the card game. Furthermore, both conditions reported liking the game and caring about earning points (i.e. the main goal of the game). This is relevant, because it ensures that potential differences in food choice are not the result of plausible side-effect game experiences. Moreover, these observations indicate that the outcome was indeed desired, an important precondition of experiencing relative deprivation [ 16 , 19 ].

Study 2: Effect of PRD on hypothetical food choice

The aim of Study 2 was to test the effect of PRD on (hypothetical) food choice. The points earned in the card game served as resources to select (purchase) food products in an online food shopping task. It was hypothesized that the PRD condition would select more palatable, snack-type (i.e. rewarding) food products than the control condition. Hypotheses and methods were preregistered on Open Science Framework prior to data collection ( https://osf.io/vk6mq ).

Participants

Participants aged 18–70 years who were fluent in English were recruited via Prolific and were compensated with £1.30 upon completion of the study (average completion time was 13 minutes). Other inclusion criteria were no prior participation in Study 1 and no allergy for gluten, dairy/lactose, eggs, nuts, and wheat/grain. Based on a power calculation using G*Power (alpha of .05, power of .90, effect size (Cohen’s f ) of .2) and an additional 10% of oversampling allowing for exclusion, the sample size for recruitment was 292. A small to medium effect size was expected, given that food choice is a multifactorially determined behaviour and plausibly also/most affected by individual differences in liking of the specific products [ 47 , 48 ]. Six participants were excluded from analysis based on a priori exclusion criteria (i.e. incorrect answers on both attention check items (N = 2) and an exceptional completion time of below 6 minutes ( N = 1) or higher than 35 minutes ( N = 3), based on Study 1, suggesting inadequate performance). As a result, the sample for analysis consisted of 287 participants (58.2% male, 41.1% female, 0.01% other) with a mean age of 30.57 years ( SD = 10.52, range 18–68). Of the 44 participating nationalities, the most frequent were British (19.2%), American (11.5%), and Polish (11.1%).

After providing written informed consent, participants completed the Revised PRD Scale [ 21 ] and the Power of Food Scale [ 49 ] (see description of these control measures below). Next, they were randomly assigned to either the PRD or the control condition and played the card game described in Study 1 (PRD manipulation) in which they earned points that served as resources to be spent in the succeeding online food shopping task. Participants were then asked about their level of hunger, weight in kg, and height in cm (conversion information was provided, i.e. 1 inch = 2.54 cm and 1 pound = 0.4536 kg), dietary concerns, and demographic information including gender, age, nationality, and last completed education. Lastly, participants were thanked and debriefed. The study was approved by The Social Science Ethics Committee of Wageningen University.

Online food shopping task

Participants were asked to choose from eight products presented on screen, using their total earnings of 30 points. Each product picture portrayed one serving for immediate consumption. Four typical high-sugar/fat snack-type food products (i.e. chocolate cookie, chocolate bar, two types of crisps) were deemed as ‘rewarding’ foods and the other four products (i.e. unsalted peanuts, muesli bar, rice waffles, pear) were deemed as ‘neutral and healthy’. This classification was determined based on a pilot study in which participants rated the palatability and healthiness of food products (see S2 Appendix for details). As each product ‘cost’ 10 points, participants had to select three products (different products or more of the same product). This choice was obviously hypothetical, i.e. participants did not receive chosen products. To approach a sense of ‘wanting’, i.e. the anticipation of pleasure due to learned associations with reward [ 31 ], participants were instructed to imagine that the foods were immediately available to them and to base their food choice on what they desired the most at that moment. The dependent variable was the number of rewarding food choices, ranging from 0 to 3.

Control variables

Chronic PRD . Chronic tendencies for PRD were measured, because it was reasoned that these may in themselves influence food choices [ 44 ] and interfere with the state of experiences of PRD. The 5-item revised Personal Relative Deprivation Scale [ 21 ] was used, which assesses the extent to which participants feel subjectively worse off compared with others (e.g. “I feel deprived when I think about what I have compared to what other people like me have”) on a 6-point Likert scale ranging from 1 (strongly disagree) to 6 (strongly agree). A mean score was calculated for the 5 items. The scale had a Cronbach’s alpha of .76.

Sensitivity to palatable food . The Power of Food Scale (Cronbach’s alpha = .93) measures appetite for palatable foods and the psychological influence that the food environment has on the individual [ 49 ]. The scale has 15 items (e.g. “I find myself thinking about food even when I’m not physically hungry”) answered on a 5-point scale ranging from 1 (I don’t agree at all) to 5 (I strongly agree). A mean score was calculated. Sensitivity to palatable food was included because this trait-level factor may obviously have considerable influence on rewarding food choices in this experiment [ 50 , 51 ].

Hunger . Participants were asked how hungry they were on a 7-point scale ranging from 1 (not at all) to 7 (very much), as this has been shown to be a primary motive for eating [ 52 ] and may also influence food choices.

BMI . Body mass index was calculated by dividing self-reported weight (kg) by the square of the person’s height (m 2 ). Seven participants had an unrealistic BMI value (> 271) because of a low value for height (< 1 metre); these were coded as missing values. BMI was included because it has been reported that people with overweight or obesity tend to choose smaller immediate rewards [ 53 , 54 ], which may translate into more rewarding food choices.

Dietary concern . Dietary restraint was measured with the 6-item Concern for Dieting subscale of the Revised Restraint Scale, which assesses attitude towards dieting [ 55 ]. Items were answered on a 4 to 5-point scale. The scale had a Cronbach’s alpha of .72. Dietary concern was included because dieters’ food choice may be predominantly based on weight-control strategies such as eating fewer calories [ 56 ].

Descriptives, correlations, and comparability between conditions

Following the preregistration, an analysis of significant correlations between the control variables and the dependent variable–rewarding food choice–resulted in the identification of sensitivity to palatable food and hunger as covariates. T-tests revealed that conditions did not differ on chronic PRD, sensitivity to palatable food, hunger, age, BMI, and dietary concern, t < 1.11, p > .269. Conditions did not differ either on gender, χ 2 (2, N = 287) = 2.16, p = 0.34, nationality, χ 2 (43, N = 287) = 45.02, p = 0.39, or education level χ 2 (8, N = 287) = 8.79, p = 0.36, suggesting that randomization was successful. See Table C in S1 Appendix for the correlations between variables and the means and standard deviations (SDs) per experimental condition.

Test of hypothesis

Neither of the identified covariates (i.e. hunger and sensitivity to palatable food) interacted with condition for rewarding food choice, meaning that the assumption of homogeneity of regression slopes was met. As the residuals of the mean number of rewarding food choices were non-normally distributed, bootstrapping (10,000 samples) was applied [ 57 ]. There was no significant main effect of experimental condition on the mean number of rewarding food choices, F (1, 283) = 1.08, p = .60, 95% CI[-0.29, 0.17], η p 2 = 0.001. Participants in the PRD condition ( M adj, = 1.73, SE = 0.08) did not differ from those in the control condition on the number of rewarding foods chosen ( M adj, = 1.67, SE = 0.08). The covariates hunger and sensitivity to palatable food were significantly related to rewarding food choice, p ’s < .05. Tests of hypotheses and exploratory analyses were also performed without serious outliers in BMI (outliers were identified according to a Z-score criterion of ± 3,i.e. BMI > 56)). Exclusion of these seven outliers did not change the results.

Exploratory analyses, preregistered

Exploratory analyses of two-way interactions between experimental condition and control variables on rewarding food choice were conducted, using the PROCESS macro for a bootstrapped test [ 58 ]. A significant disordinal interaction between condition (centered) and chronic PRD (centered) on rewarding food choice was found, F (1, 283) = 7.08, p = .008, R 2 - change = 0.02, also when the identified covariates were controlled for, F (1, 281) = 4.48, p = 0.04, R 2 - change = 0.01. See Fig 2 for a visualisation of the interaction effect. Simple effects including identified covariates demonstrated that conditions did not differ significantly on rewarding food choice for participants with a high level of chronic PRD (+1SD; 4.02), B = .30, t (281) = 1.71, p = .087, 95% CI [-0.64, 0.04], average level of chronic PRD (3.09), B = .06, t (281) = .53, p = .59, 95% CI [-.29, .18], or a low level of chronic PRD (-1SD; 2.16), B = -.17, t (281) = -1.19, p = .23, 95% CI [-0.11, 0.47]. Simple effects without covariates showed the same pattern of results as those with covariates, except that the simple effect of high level of chronic PRD was significant (B = -.38, t (283) = -2.16, p = .03, 95% CI [-0.72, -0.03]). This difference was in the hypothesized direction, indicating that the PRD condition chose more rewarding foods than the control condition under high levels of chronic PRD. No interaction effects were found between condition and any other control variables, p > .21.

No main effect of the PRD manipulation on food choice was found. However, the manipulation seemed to have a differential effect for different levels of chronic PRD. Simple effects bordered on significance, showing a pattern that those that were relatively deprived in the card game and also experienced higher chronic relative deprivation appeared to select a higher number of rewarding foods. It may be reasoned that individuals already feeling deprived were more sensitive to the manipulation with respect to its influence on food choice, and that those who did not experience PRD in daily life did not translate incidental feelings of PRD into hypothetical food selection. A limitation of this study was that food choices were hypothetical; therefore, the experiment was next conducted in a lab-in-the-field setting.

Study 3—Effect of PRD on non-hypothetical (real) food choices

The aim of Study 3 was to conceptually replicate the online study (Study 2) in a lab-in-the-field setting in which participants from a community sample made real, non-hypothetical food choices. It was hypothesized that participants in the PRD condition would choose more palatable (rewarding) snack-type food products than participants in the control condition (i.e. a main effect). Moreover, given the results of the online study, this effect could be expected to be observed only, or at least to a greater extent, in participants scoring high on chronic PRD. The aim, hypothesis, and methods were preregistered on Open Science Framework prior to data collection ( https://osf.io/fwy8r ).

A total of 308 women were recruited at a one-week summer fair ( Libelle Zomerweek ), where they could participate in our workshop “grocery shopping game”. Twenty-one rounds of the workshop were held across the fair (at 10.30, 11.30, 12.30, 14.30, 15.30, and 16.30 h) in maximum groups of 15 participants. Given these constraints, the maximum sample size that could be reached was 315 participants. Participants could enroll for the workshop at the central registration desk and, to fill the open spots, some were actively recruited by research assistants. Twenty-five participants were excluded from analyses because they had an allergy or intolerance for gluten, dairy/lactose, eggs, nuts, soja, or wheat/grain, or followed a vegan diet. Additionally, 19 participants were excluded because they did not adhere to instructions in one or multiple ways (i.e. taking more or fewer than three products, stopping before the food choice task, performing the task together with a friend). Four participants that correctly guessed the purpose of the study were also excluded. Hence, the analytic sample consisted of 260 participants, with an average age of 48.75 ( SD = 14.05 range 16–76) and various educational backgrounds (categorized into 18.8% low, 42.2% middle, 37.3% high, according to a classification of Statistics Netherlands ( www.cbs.nl ), and 0.8% indicated ‘other’). A post-hoc power calculation using G*Power indicated that with this sample size, a power of .80 could be reached for a small to medium effect size of the hypothesized main effect, Cohen’s f = .175.

Participation in this annual summer fair was considered a good opportunity for efficient recruitment of a large and diverse community sample. As this female sample differed from the mixed gender sample in Study 2, we checked whether the results of Study 2 differed for men and women: no two-way interaction between gender and condition was found, and no three-way interaction between gender, chronic PRD, and condition was found.

Procedure and measures

Participants were seated at a long table and were separated by shields so that they would complete the study individually. Each participant was provided with a participant number, a handout with the steps, and a tablet computer. At the start of the workshop, participants were collectively introduced to the study “about grocery shopping” and were instructed not to talk with one another. The same methodology (i.e. informed consent, manipulation, and measures) as in the online study was applied, with the following adaptations. First, for feasibility reasons, the procedure was shortened by excluding the nationality measure and by using the 4-item Present Food subscale of the Power of Food Scale [ 49 ] and one item from the Dietary Concern scale, i.e. “How conscious are you of what you are eating?” [ 55 ]. The 4-item Present Food subscale of the Power of Food Scale involves “reactions to palatable foods when they are physically present but have not yet been tasted”. This subscale (Cronbach’s alpha = .84 in the current study) was conceptually closest to food choice and correlated most strongly with food preference in the online study. The 1-item dietary concern was based on a principal component analysis with varimax rotation of the online study data, which suggested two components; one of the components containing this item (loading .92) was conceptually closest to food choice. Also, weight and height items were deleted to avoid participants feeling uncomfortable answering these items. Second, the food choice measure was realized by letting each participant collect products from a wooden crate positioned at her table (see Food choice task ). Third, in the instructions for the card game, it was already explained that each product cost 10 points, as it was arguably better to know beforehand the value of the points that could be earned. Fourth, a final question asked what participants thought the purpose of the study was. Participants were collectively debriefed via a quiz. The study was approved by The Social Science Ethics Committee of Wageningen University. Prior to this study, a non-preregistered pilot study was conducted to test the feasibility of the PRD manipulation and a different form of the food shopping task among a community sample ( N = 101, see S4 Appendix ).

Food choice task

Participants were instructed to spend their points earned in the card game on groceries. Each product cost 10 points, as indicated on Qualtrics and on the ‘price tag’ near each product. Participants chose products from a cloth-covered crate containing two shelves that presented the eight different products (four products per shelf, three pieces of each product). The placement of the products was identical for all participants (see also Fig 3 ). Participants were instructed to remove the cloth from the crate and to put their choice of products in a paper bag. On the Qualtrics instruction screen, it was stated that participants should not think for too long and choose something that they would want to eat now. In conformity with the online study, the rewarding foods were chocolate waffle, chocolate bar, and two types of crisps. The neutral, healthy foods were muesli bar, pear, rice waffles, and unsalted peanuts. All foods were served as single-portion packages (except the pear). After participants left, food choices were recorded by the researchers (who were blind to the condition assignment) by counting the products taken.

An analysis of significant correlations between the control variables and the dependent variable–rewarding food choice–resulted in the identification of age, education level, dietary concern, and sensitivity to palatable food (Present Food subscale) as covariates. T-tests showed that conditions did not differ on the control variables sensitivity to palatable food (Present Food subscale), hunger, age, and dietary concern, t(258) < 1.72, p > .09 and education, χ 2 (7, N = 258) = 11.58, p = .12. Table D in S1 Appendix presents the correlations between the variables under study and the means and standard deviations (SDs) per condition.

Test of hypotheses

None of the identified covariates (i.e. age, education level, dietary concern, and sensitivity to palatable food) interacted with condition for rewarding food choice, meaning that the assumption of homogeneity of regression slopes was met. A bootstrapped (10,000 samples) ANCOVA showed that there was a significant main effect of experimental condition on the mean number of rewarding food choices, F (1, 250) = 4.61, p = .031, 95% CI[0.03, 0.45], η p 2 = 0.02. Participants in the PRD condition ( M adj, = 1.56, SE = 0.16) chose more rewarding foods than those in the control condition ( M adj, = 1.33, SE = 0.17). The covariates education level, dietary concern, and sensitivity to palatable food were significantly related to rewarding food choice, p ’s < .01. Age was not significantly related to food choice, p = .14. This main effect was not significant without any covariates, F(1, 258) = 2.49, p = .11, 95% CI[-.0.04, 0.40], η p 2 = 0.01. Controlling only for sensitivity to palatable food resulted in a significant main effect, F(1, 257) = 3.97, p = .043, 95% CI[0.01, 0.44], ηp2 = 0.02, indicating that this was an important covariate in the testing of hypotheses.

To analyse whether condition affects food choice differently for scores on chronic PRD (in conformity with Study 2), it was tested whether chronic PRD (centered PRD scale) interacted with experimental condition (centered), using bootstrapping in PROCESS [ 58 ]. This interaction was not significant, F (1, 256) = 1.86, p = .17, R 2 -change = 0.01; nor was it significant when the identified covariates were controlled for F (1, 248) = 0.53, p = .47, R 2 -change = 0.00. Removing two participants with the highest chronic PRD scores (> 4.20) from the sample did not change these non-significant results of this two-way interaction, ANOVA model p = .07, ANCOVA model, p = .37.

Three-way interactions between experimental condition, chronic PRD, and each of the control variables were explored with PROCESS [ 58 ]. A significant interaction was found between centered condition, centered chronic PRD, and centered age, F (1, 252) = 4.32, p = .039, 95% CI[0.001, 0.048], R 2 -change = .02, also when the identified covariates were controlled for, F (1, 245) = 5.03, p = .026, R 2 -change = 0.02. This adjusted three-way interaction was further disentangled by bootstrapped conditional effects. No significant conditional effects of the two-way interaction between chronic PRD and condition on rewarding food choice were found for participants with higher age (+ 1SD; 62.8 years), B = .45, t (245) = 1.97, p = .050, 95% CI [0.001, 0.89], lower age (- 1 SD; 34.7 years), B = -.22, t (245) = -1.91, p = .23, 95% CI [-0.58, 0.15], or medium age (48.8 years) B = .11, t (245) = 0.79, p = .43, 95% CI [-0.001, 0.41]. As the answer items of the PRD scale ranged from 1 to 6, it appears that PRD scores were overall rather low; these low, medium, and high ranges need to be interpreted in relative terms. Other three-way interactions between condition, chronic PRD, and any control variable other than age were not significant, all p ’s > .15.

On average, participants in the PRD condition chose more rewarding foods than those in the control condition, particularly when sensitivity to palatable food was controlled for. An interaction between condition and chronic PRD, as observed in Study 2, bordered on significance for participants in the higher age group.

General discussion

Drawing on the relative deprivation hypothesis and the theory of justice motivation [ 23 ], the aim of this research was to demonstrate the effect of experiences of PRD on food choice behaviour. In two preregistered experimental studies, some preliminary evidence was found for an effect of induced PRD on hypothetical and real food choices in a grocery shopping setting. Across the studies, a difference in food choice between conditions bordered on significance and hence the findings need to be interpreted with caution. The effect of PRD on real, non-hypothetical food choice appeared significant, when trait-level sensitivity to palatable foods was controlled for. A numerical difference was repeatedly in the hypothesized direction: those who were unfairly relatively deprived of resources earned in a card game spent these resources on more palatable, energy-rich food products (rather than neutral tasting, healthier foods), although this effect may be truer for those with higher chronic PRD.

This is one of the first studies to demonstrate some causal effect of feelings of PRD on actual food choice behaviour. Building on previous studies showing that chronic-level PRD is associated with health outcomes [ 19 , 20 ] and state-level PRD with an inclination towards immediate rewards [ 21 ], our results show that PRD may result in more palatable and unhealthy food choices. These findings are relevant in light of the current obesogenic environment characterized by its abundance of unhealthy foods, where it appears that part of society is disproportionally affected by these temptations. A large body of evidence indicates that inequality and obesity and health are linked at the societal level. The (preliminary) results of this study suggest that this association may be partly due to the effect of relative deprivation (a downstream psychological consequence of inequality) and food choices at the individual level.

Specifically, we carefully manipulated relative deprivation on the basis of a theoretical conceptualisation of PRD, including both cognitive (‘being worse-off’, ‘being unfairly treated’) and affective responses (anger, resentment) to unequal/unfair outcomes, stemming from upward comparisons [ 15 , 16 ]. Notably, as all participants in this study received objectively the same number of resources, this research indicates the relevance of targeting the perception of inequality and unfairness beyond a person’s absolute, economic situation. Indeed, Inglis, Ball [ 59 ] showed that socioeconomic inequalities in the healthfulness of food choices were not reduced through varying food budgets available to women with low and high incomes. Decreasing economic inequality has been declared as the approach to improve a nation’s health [ 60 ], but, with respect to dietary behaviours, this may be insufficient when not accompanied by a reduction in experiences of injustice stemming from social comparisons.

Our findings have important implications for theorizing about food choice behaviour under states of deprivation in two main ways. First, the finding that those who were relatively deprived chose more snack-type foods that were assessed as relatively unhealthy and highly palatable may imply that the hedonic–and so immediately rewarding–properties of foods become more important under deprivation. Hence, an implicit assumption for this reasoning is that the act of making food choices under deprivation theoretically entails a subconscious motivational trade-off between immediate pleasure and long-term health goals; this accords with empirical evidence for the relation between PRD and a preference for immediate rewards [ 21 , 22 ] and with research on self-regulation of eating [ 61 ]. However, in previous studies on SSS and caloric preference/intake, such a trade-off categorization has not been made and findings have generally been explained with an evolutionary approach that assumes that people have a functional, adaptive motivation to compensate for resource scarcity directly with calorie-rich foods, as this promotes survival [ 37 , 38 , 62 ]. This functional explanation could apply to the present study as well, especially given a sub-analysis showing that PRD also affected actual food choice when calorie-rich peanuts, which were not perceived as particularly tasty, were additionally categorized as rewarding foods rather than as neutral, healthy foods. Hence, it remains questionable whether PRD leads to a particular preference for energy versus hedonic properties of foods, although these specific processes may be hard to disentangle as the reward system has evolved such that humans prefer energy-dense sweet and fat foods [ 63 , 64 ]. More generally though, individuals’ liking of foods evolves over the life course as a result of socio-cultural influences, including food exposure [ 65 , 66 ]. It remains an avenue for future research on socioeconomic disparities in diet quality to investigate how the food liking of people on various SES levels is shaped by various socio-economic/cultural factors, including experiences of deprivation and differential exposure to food availability.

Second, the finding that the effect of state-level PRD on rewarding food choice was moderated by higher chronic feelings of PRD may elucidate the challenge of recapitulating an intrusive level of PRD as experienced by individuals actually living under deprived conditions [ 35 ]. Particularly, a plausible explanation for this finding is that only those with chronic PRD actively engaged in the self-comparison that is needed for a true PRD effect. Moreover, these findings may contribute to insights into how food choice behaviour is formed by previous experiences of deprivation. In light of studies focused on emotions as a stimulus for the conditioning of food craving and selection [ 67 , 68 ], our findings may imply that state-related deprivation may lead to palatable food selection because of a learned response to feelings of PRD, reflecting a form of habitual reinforcement [ 69 ] or more generally, of poor emotion regulation. Notably, a series of studies on the influence of childhood SES on food intake in a laboratory setting, showed that for those raised in high SES environments, food intake varied according to physiological energy need, but for those raised in low SES environments, food intake was high independent of of current energy needs [ 70 ]. It was theorized that early exposure to low SES may become embedded in energy regulation systems in later life, plausibly due to learning experiences with deprivation. Further research is needed to further disentangle whether and how states of PRD and childhood or chronic experiences of PRD may interact in the influence on food preferences/cravings and dietary patterns.

This study has some particular strengths. First, the experiment with real, non-hypothetical food choices was conducted in a community sample with various educational backgrounds, rather than in a homogenous highly educated student sample as commonly used in previous studies on social status and eating [ 27 , 38 , 40 , 44 , 71 ]. Second, we employed a real-time induction of PRD experiences where participants were actually unfairly deprived of resources compared to a (fictitious) opponent player, and we believe that this method is more powerful than a person imagining him/herself in a hypothetical scenario. Moreover, the resources that were the source of relative deprivation were directly used in the grocery shopping task, and this enhances ecological validity.

Yet, there may be some limitations regarding the conceptualisation, operationalisation and measurement of personal relative deprivation that we employed. First, the applied conceptualisation of relative deprivation was rather comprehensive, not only including the act of upward social comparisons, but also perceptions of unfairness and negative emotions. According to theories of relative deprivation [ 14 – 17 ], a sense of unfairness is part of PRD because human justice concern is a precondition for the experience of relative deprivation–the sense that one does not get what he/she deserves relative to others. Interestingly, a series of experiments testing social comparison processes in PRD showed that feelings of unfairness mediated the effect of adverse social comparison on feelings of resentment and dissatisfaction with a financial outcome [ 72 ]. Yet, in the wider context of aiming to understand the link between societal inequality and unhealthy diets, it may appear questionable whether feelings of unfairness is a relevant additional process as compared to the mere upward social comparison process. Previous studies have shown that upward social comparisons without feelings of unfairness per se influenced unhealthy eating behaviour [ 35 , 36 ]. For example, people that perceived to have a lower relative income than their interaction partners consumed more calories [ 27 ]. Hence, it would be of interest for future research to test whether and how a mere upward social comparison process would produce similar effects on eating behaviour as compared to feelings of being unfairly disadvantaged.

Second, one may question the validity of our self-developed digital manipulation with respect to the social comparison process. Participants were led to believe that they played against a previous participant with the same age and gender, and they were explicitly presented with the relative earnings, but it remains questionable to what extent they truly compared themselves with a similar other, rather than a computer or a dissimilar other. Although the manipulation check demonstrated that the PRD manipulation indeed led to feelings of being worse-off, the check items included wording specifying the comparative nature of the feelings. Hence, it cannot be verified to what extent the manipulation distinguishes from general forms of negative affect that are not due to social comparison per se. However, the digital design of the PRD manipulation provided great benefits in terms of feasibility as well as reliability, as it allowed us to compose a uniform game that included the exact same amount of relative deprivation ‘portions’ across all participants in the PRD condition. Building on the aforementioned strength of inducing real-time deprivation of resources that were actually spent on food products, we encourage future experimental studies on effects of PRD to also involve real persons (e.g. research confederates) for a more optimal social comparison process.

This study has also some methodological limitations regarding the assessment of food choice. First, the food products offered during the grocery shopping task were limited in variety and may not represent the foods chosen by the participants in daily life. Nevertheless, we selected products with which most people are familiar, and in sub-analyses we observed that all products were substantially selected by the samples, allowing some variety in popularity. Second, we assessed food preference at only one juncture, and more long-term assessments of food purchases are necessary to generalize to dietary patterns. Third, all foods had the same ‘price’ in the food shopping task, whereas in real life these foods have different prices. Nevertheless, we counterbalanced for actual prices of products in the two categories of foods (i.e. rewarding versus neutral) so that these categories were equally expensive. Future studies on the effect of PRD on food choice may disentangle the relative importance of price and palatability of the presented foods. Fourth, presented foods were categorized into rewarding versus neutral foods based on perceived palatability and healthiness. Other, unmeasured food properties (e.g. convenience of consumption) may also have differed between these categories and may have confounded the outcome. To investigate the idea that PRD leads to a heightened sensitivity to rewarding foods, we suggest that future studies should focus on a more proxy measure of rewarding food preference, such as rewarding value of food [ 51 , 73 ] or food cue reactivity [ 74 ]. Fifth, in Study 3, the sample consisted of women only and so its findings cannot be generalized to men. Yet, studying a female population may be especially relevant, as the social status and BMI relationship has been more consistently found among women than among men [ 35 , 36 , 75 ]. Future research on PRD and dietary outcomes may need to identify any differential effects for gender. Sixth, the perceived palatability and healthfulness of the foods was not consistently evaluated in each experiment. Although the selection of foods was based on previous studies [ 76 – 78 ], the lack of these data in our series of studies precludes evaluation of the extent to which null effects were the result of invalid food categorization across our different samples and/or within individuals. This is of particular relevance for Study 3 in which we recruited a female sample and measured real food choice. Furthermore, apart from feasibility concerns (i.e. time constraints), the validity of these evaluations pre or post experiment may be questionable, given potential carryover effects.

Finally, in neither Study 2 (hypothetical food choice) nor Study 3 (actual food choice), participants consumed the foods selected, which may be considered a limitation given that the hedonic experience is directly associated with reward. However, according to Kringelbach, Stein and van Hartevelt [ 79 ], the food pleasure cycle consists of three stages: anticipation/wanting, consumption/liking, and learning (our food choice measurement most likely tapping in the first stage, building on Pavlovian associations). In both our studies we included well-known food products, and we instructed participants to base their choice on what they would like to eat at that moment. Based on neuroimaging studies [ 34 ], it was reasoned the (palatable) food products shown on screen (study 1), and especially the products that were actually available for immediate or later consumption (study 2) would elicit reward responding related to ‘wanting’, which is an important driver of reward seeking behaviours including food choice [ 31 , 80 ].

Although conclusions can only be drawn with reservation, this research found some initial evidence for a causal effect of experiences of relative deprivation on unhealthy food choices. Effect sizes of this multifactorial behavioural outcome were small, and more research is needed to establish more firmly a causal link between relative deprivation and unhealthy dietary behaviours. Nevertheless, this research provides initial support for the idea that targeting inequality and injustice, and their psychological consequences, may be part of the solution to reduce societal disparities in the healthfulness of dietary patterns.

Supporting information

S1 appendix, s2 appendix, s3 appendix, s4 appendix, acknowledgments.

We thank Sanne Raghoebar and Joyce Copier for assistance with running the experiment in Study 3.

Funding Statement

YES. The research was financially supported by a personal grant from the talent scheme of the Netherlands Organization for Scientific Research (NWO) awarded to prof. dr. EWML de Vet, grant number 452-14-014. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data Availability

Advertisement

The effect of food deprivation on human resolving power

• Brief Report
• Published: 03 May 2017
• Volume 25 , pages 455–462, ( 2018 )

Cite this article

• Noa Zitron-Emanuel 1 &
• Tzvi Ganel 1  

5777 Accesses

8 Citations

2 Altmetric

Explore all metrics

The feeling of hunger is an inseparable part of people's daily lives. It has been established that hunger, caused by food deprivation, influences people’s physiological and emotional state and their everyday behavior. Yet, it remains unclear whether and in which manner food deprivation affects the way people perceive their environment. In two experiments, we examined the effects of food deprivation on the perceptual resolution of food portion size. We calculated Just Noticeable Differences (JNDs) to measure sensitivity to detect the smallest difference between two stimuli of different sizes. Participants' resolution in both experiments was higher to detect changes in food size compared with baseline when they were hungry due to a short period of food deprivation. Food deprivation did not lead to any biases in the average perception of food size. The results show that food deprivation changes the way people perceive their environment. We discuss the possible role of attention in mediating the effect of food deprivation on the visual resolution of food size.

Similar content being viewed by others

Food deprivation disrupts normal holistic processing of domain-specific stimuli.

Does hunger promote the detection of foods? The effect of value on inattentional blindness

Perceptual salience influences food choices independently of health and taste preferences

Avoid common mistakes on your manuscript.

For several decades, psychologists have looked into possible effects of motivational factors on human perception and cognition. Findings were largely inconsistent; several studies yielded results suggesting that motivational factors influence perception, as demonstrated by Bruner and Goodman ( 1947 ), who found that poor children perceive the size of a coin as bigger compared with rich children. However, those studies were heavily criticized due to methodological issues as well as for the theoretical aspects of the drawn conclusions (Tajfel, 1957 ; Veltkamp, Aarts, & Custers, 2008 ).

Previous related literature has mainly focused on the effects of different descriptive variables, such as the participant’s gender, age, or body mass index (BMI), on average perceived size , whereas fewer studies have looked into the effects of motivational states, such as hunger and food-deprivation (Beasley, Hackett, Maxwell, & Stevenson, 2004 ). We focus on the effect of hunger, caused by mild food deprivation, on human perceptual resolution.

The feeling of hunger is an inseparable part of our daily lives. It has been widely established that hunger can have a strong influence on people’s physiological and emotional state and on their everyday behavior (Berridge, 2004 ; Loewenstein, 1996 ). Despite its major impact on our daily lives, it remains unclear whether and in which manner hunger affects the way people perceive their external environment. Lazarus, Yousem, and Arenberg ( 1953 ) reviewed several studies that looked at the relationship between hunger and perception and pointed out that some of these studies did not actually measure perception. Virtually, all previous studies on motivational effects (including hunger) on size perception have focused on the effect of motivational variables on biases in average perceived size (Witt & Dorsch, 2009 ). A recent study asked participants to estimate the size of a water glass, under fluid deprivation, and showed that deprived participants perceived a glass of water as bigger compared with nondeprived participants (Veltkamp et al., 2008 ). Yet, the literature in the field is inconclusive, probably due to the large variability across experimental methods and stimuli used. For example, Kral, Roe, and Rolls ( 2004 ) served participants with different meals (breakfast, lunch, and dinner) of varying portion sizes (100%, 150%, 200%) across 3 weeks. Participants were asked to rate the perceived size of the first and last dish of each meal—representing conditions of hunger and satiety. No significant differences were found between the conditions.

In a seminal study that looked at the effects of food deprivation on perception, Lazarus and his colleagues ( 1953 ) provided initial evidence that mild food deprivation may lead to a decreased threshold for shape identification of food-related stimulus. More recently, Radel and Clement-Guillotin ( 2012 ) showed that food-deprived participants were more likely to identify briefly presented food-related words compared with nondeprived participants. Note, however, that the last study focused on the effect of food deprivation on semantic word identification rather than on visual resolution of food-related stimuli.

The present study examined the effects of hunger on the perceptual resolution of physical food size. To this purpose, we used the classic method of constant stimuli tailored to the domain of food size perception. We calculated JNDs to measure the sensitivity to detect the smallest difference between two stimuli of differing sizes and calculated possible biases in food portion size perception.

Experiment 1

We examined the effects of mild food deprivation on participants' sensitivity to detect changes in the size of a food-related stimulus compared with neutral, control stimuli. We hypothesized that food-deprived participants would be more sensitive to differences in the size of a food stimulus compared with simple-shape stimulus and compared with nondeprived participants. In other words, it was predicted that JNDs for food related-stimuli would be smaller compared with baseline in the food-deprived group.

Participants

Seventy-two undergraduate female students (age range: 18-29 years, mean = 22.98, SD = 1.53) from Ben-Gurion University of the Negev, participated in the study for credit in a psychology course or received the equivalent of $6 for their participation. All participants had normal or corrected-to-normal vision; 39 participants were randomly allocated to the food-deprivation condition; the other 33 participants were allocated to the nondeprivation condition. Data from five participants (3 food-deprived, 2 nondeprived participants) were excluded from the analyses due to low (smaller than 70%) values of Goodness-of-Fit. Data from four additional participants (1 food-deprived, 3 nondeprived) self-reported as being clinically diagnosed with an eating disorder also were excluded. Participants gave an informed, written consent before their participation in the experiment, which was approved by the Ethics Committee of the Psychology Department at Ben-Gurion University of the Negev. The experiment lasted approximately 40 minutes.

Stimuli consisted of high-resolution color images of food and neutral stimuli or round contours of those stimuli as control (Fig.  1 ). We used images of a chocolate chip cookie (food stimulus), a tennis ball (neutral stimulus), and a circle contour (simple-shaped stimulus). Images were edited using Photoshop CS by cutting the images into a nearly perfect circle. The resulting stimulus set included a base, standard stimulus, and 12 additional reference stimuli, smaller or bigger than the 95-mm standard stimulus in constant 1mm intervals. All stimuli were presented on a white background and were viewed from 60-cm distance on a 19” computer screen.

Stimuli used in Experiments 1 and 2 . Depicted are the standard (baseline) stimulus and one of the reference stimuli: ( a ) chocolate chip cookie standard image and chocolate chip cookie reference-image used in Experiment 1 ; ( b ) simple circle shape standard image and reference image; ( c ) chocolate pie standard image and chocolate pie reference image used in Experiment 2 ; ( d ) tennis ball standard image and tennis ball reference image used in Experiment 2

Design and procedure

Participants who were assigned to the food-deprivation condition were asked not to eat at least 3 hours before the experiment. Those assigned to the nondeprived condition were asked to eat during the hour before the experiment.

Participants were asked to choose the smaller image of the two images presented on the screen by pressing the appropriate keyboard key. The experiment began with two practice trials that were discarded from further analyses and continued with four experimental blocks: block A consisted of a pair of food stimuli—a chocolate chip cookie standard image and a chocolate chip cookie reference image; block B consisted of a pair of simple-shaped stimuli—a circle-shaped standard image and a circle-shaped reference image. These two blocks were designed to measure JNDs or perceptual resolution (Fig.  1 ). Block C consisted of a food stimulus and a simple-shaped stimulus—a chocolate chip cookie standard image and a circle-shaped reference image; and block D consisted of a neutral stimulus and a simple-shaped stimulus—a tennis ball standard image and a circle-shaped reference image. These two blocks were designed to measure PSEs to look for potential effects of perceptual bias. All blocks were randomly ordered across participants.

Participants were presented with a pair of images in each trial—the standard image and 1 of the 12 reference images. The images were presented at opposite heights and sides to avoid participants of using height cues. Relative locations were counterbalanced across trials (left-up, left-down, right-up, right-down). A 500-msec fixation dot appeared between trials. Within each block, each of the 12 standard-reference combinations was randomly repeated 16 times, resulting in 192 trials per block, and 768 trials overall.

Following the experimental session, participants were asked to complete the EAT-26 (Eating Attitudes Test) questionnaire as a screening tool for eating disorders symptoms. Higher scores indicate greater pathology, and a score greater than 20 is recommended as the clinical cutoff (Garner, 2004 ).

Data analysis

Blocks A and B were used to calculate JNDs for the food and control stimuli. We first calculated for each of the reference stimuli the proportion of the number of trials in which it was perceived as smaller than the standard stimulus compared with the total number of trials. We then fitted the data on a sigmoid function (perceptual function), which plotted those proportions against the values of the stimuli, ranging from 8.9 cm to 10.1 cm. Finally, the JND was calculated by dividing the stimulus range between 25% and 75% correct discrimination (this range is termed – “the area of uncertainty,” an area in which the stimulus size is not perceived as distinctively different from the standard stimulus) by two (Namdar, Avidan, & Ganel, 2015 ).

Blocks C and D were used to calculate PSEs for the chocolate chip cookie and the tennis ball. The same calculation as described above was performed for each participant, only by identifying the stimulus value that corresponds to 50% correct discrimination (at this stimulus value the participant perceived the reference stimulus as equal to the standard, which represents its perceived size).

Note that the data from blocks C and D can theoretically be used to calculate JNDs, but the interpretation and the meaning of these measures is quite problematic and not effective. The design in blocks C and D does not measure subjects’ resolution in detecting the slightest difference between two almost-identical food stimuli (a within-category comparison), but rather subjects’ ability to tell the difference between the diameter of a food stimulus and the diameter of a different object (a between-category comparison).

To validate the fit of the participant’s pattern to the sigmoid function, we computed Goodness of Fit (GOF) scores for each block. Participants, who showed GOF < 0.7 in one of the blocks, were excluded from further analysis. The total averages (SD) of GOF scores were 0.95 (0.05), 0.94 (0.05), 0.93 (0.05), and 0.94 (0.04) in blocks A, B, C, and D, respectively.

Results and discussion

JND scores were submitted to a 2-factor mixed effects analysis of variance (ANOVA). It consisted of a within-subject factor of stimulus type (food stimulus vs. baseline) and one between-subjects factor of group (food deprived vs. nondeprived). The results are presented in Fig.  2 . JNDs for the food-deprived group were smaller for the food stimulus compared with the control stimulus. The nondeprived group did not show such an increase in performance for the food-related stimulus.

JND scores for the food stimulus and the baseline stimulus, for food-deprived, and for nondeprived participants in Experiment 1 . Perceptual sensitivity for food stimuli in the food deprived group was significantly higher than baseline. No statistical difference was found for the nondeprived group. Error bars represent 95% confidence intervals for mixed-design ANOVAs

The EAT-26 questionnaire consisted of 26 items; Cronbach's alpha is 0.833 , which indicates a high level of reliability and internal consistency. The average measure ICC for PSE was 0.749 with a 95% confidence interval from 0.585 to 0.848 (F(62,62) = 3.983, p < 0.0001) and the average measure ICC for JND was 0.699 with a 95% confidence interval from 0.503 to 0.818 (F(62,62) = 3.324, p < 0.0001). This states that 74.9% of the variance in the PSE scores and 69.9% of the variance in the JND scores could be accounted for, indicating an acceptable level of intra-subject reliability. The main effects of stimulus type [F(1,61) = 0.369, p > 0.05, η p 2 = 0.006] and group [F(1,61) = 0.042, p > 0.05, η p 2 = 0.001] were not significant, suggesting that there were no statistical differences between JNDs for food and for baseline stimuli, nor between the deprived and nondeprived groups.

More importantly, the interaction between stimulus type and group was significant [F(1,61) = 4.416, p < 0.05, η p 2 = 0.068], indicating that food-deprived, compared with nondeprived participants, showed a different pattern of JND scores to food compared with baseline, control-stimuli. Specific comparisons revealed that JNDs for food stimuli were significantly lower compared with baseline in the food-deprived group [t(34) = 2.95, p < 0.01] but not in the nondeprived group [t(27) = 0.777, p > 0.05]. Although there was a trend indicating smaller JNDs for food stimuli and larger JNDs for baseline stimuli, specific comparison did not show significant effects either for the baseline [t(61) = 0.874, p > 0.05] or for the food stimuli [t(61) = 1.083, p > 0.05] when the food-deprived group was compared with the nondeprived group (Fig.  2 ).

We also performed a Bayesian repeated measures ANOVA, aimed to select the best-fitting ANOVA model to the data, based on the Bayes Factor (BF) statistic. BFs were calculated using the JASP software package (JASP Team, 2016 ). In contrast to the classical ANOVA reported above, the Bayesian analysis failed to support any of the substantial models. Specifically, the null model performed better than any other model. Specifically, the best-fitting model included a main effect of stimulus type; however, even this model did not provide a better fit than the null model [BF 10 = 0.318]. We hypothesize that these nonsignificant effects in the specific comparisons and in the Bayesian analysis are due to the relatively low statistical power inherent to the between-subject design used in Experiment 1 . Experiment 2 was designed accordingly to extend and replicate the results of the current experiment using a more powerful experimental design.

As for the JND scores, we also calculated PSEs for food and neutral stimuli for the two groups. The main effects of stimulus type [F(1,61) = 0.182, p > 0.05, η p 2 = 0.003] and group [F(1,61) = 0.033, p > 0.05, η p 2 = 0.011] were not significant. The interaction between stimulus type and group also was not significant [F(1,61) = 0.673, p > 0.05, η p 2 = 0.001], indicating that food-deprived participants in general did not differ from nondeprived participants in terms of their average perceptual biases for food portion size. The Pearson correlations between JND scores for food stimuli and the EAT-26 scores [r = 0.004, p > 0.05] and between PSE scores for food-stimuli and the EAT-26 scores [r = 0.012, p > 0.05] were not statistically significant.

These results show that mild amount of food depravation leads to an increased spatial resolution for food-related stimuli. Results show that JNDs were smaller for food-related stimuli compared with baseline in the food-deprived group but not in the nondeprived group. This evidence supports the hypothesis that JNDs for the size of food stimuli are affected by objective physiological factors.

We propose that the results of Experiment 1 suggest that hunger, triggered by mild food depravation, has a significant effect on human resolving power for food-related stimuli. However, although the relationship between food deprivation and hunger has been well established (Benau, Orloff, Janke, Serpell, & Timko, 2014 ; Loewenstein, 1996 ), the design of Experiment 1 did not support a direct link between food deprivation and hunger, because the subjective feeling of hunger was not directly assessed.

Before embracing the idea that food deprivation led to increased sensitivity for food-related stimuli, a different interpretation also should be considered. In particular, the food stimuli in Experiment 1 were richer in texture and in fine visual details and complexity than the simple-shape stimulus used as baseline. Therefore, it could be argued that food deprivation led to increased sensitivity in attending to the fine details of a stimulus, rather than to specific increased sensitivity to food-related stimuli. To answer this potential concern, a control stimulus with rich textures and high resolution was used in Experiment 2 instead of the simple-shaped circle used in Experiment 1 .

Finally, we note that the between-group experimental design used in Experiment 1 could have led to incidental between-subjects differences, which could decrease the statistical power of the experiment. Such statistical noise could have accounted for the nonsignificant pattern of results revealed in the specific comparisons analysis across the two experimental groups as well as the results of the Bayesian analysis. Experiment 2 therefore was designed as a within-subject experiment to increase statistical power. In addition, as in most previous studies in the field of food perception, only female subjects were included in Experiment 1 . Research in the field typically focused on females probably due to their elevated vulnerability to food-related disorders, such as Bulimia Nervosa (Ptacek et al., 2016 ). In Experiment 2 , we included male and female participants to generalize the results of Experiment 1 across gender. This also allowed us to compare females and males in respect to their sensitivity to food stimuli.

Experiment 2 was designed to extend and generalize the results of Experiment 1 along several aspects. First, we established the role of hunger in mediating resolution for food size by including a direct measure of perceived hunger. Second, to address the issue of incidental between-group differences and to extend the generality of the results across gender, Experiment 2 was designed as a within-subject experiment, and similar numbers of male and of female subjects participated in the experiment. Finally, a different type of food and neutral stimuli were used in Experiment 2 to extend and to generalize the findings across different situations.

Experiment 2

Each of the participants was included in both the food-deprived and nondeprived conditions, in a counterbalanced order and in separate days. Furthermore, Experiment 2 was designed to address the issue of potential differences in visual details between the food-related stimulus and the control stimulus and to generalize the results of Experiment 1 across male and female subjects.

The design of Experiment 2 was similar to that used in Experiment 1 , with several changes described hereinafter.

Eighty-eight undergraduate students (age range: 18-31 years, mean = 23.36, SD = 2.35; 45 females and 43 males), who did not participate in Experiment 1 , took part in the experiment. All participants underwent the two experimental conditions: food-deprivation and nondeprivation. Condition order was counterbalanced between subjects. Data from eight participants were excluded from the analyses due to low (<70%) values of Goodness-of-Fit. Two participants had very noisy data, with JNDs higher than 3 SD than average, and their data were excluded. Data from one participant who self-reported to be clinically diagnosed with an eating disorder also were excluded.

The experiment consisted of an 11-cm diameter standard stimulus and 12 references stimuli, varying between 10.4 cm and 11.6 cm diameters. We used images of a chocolate pie and a tennis ball (Fig.  1 ).

This experiment included two sessions, separated by 1-6 days, one for each experimental condition described in Experiment 1 . Prior to the experimental session, participants were asked to rate how hungry they feel, on a subjective “hunger-scale” (rated on a 7 points scale: 1 = "replete," not hungry at all, 7 = "starving," extremely hungry).

Data were fitted on a sigmoid function (perceptual function), which plotted those proportions against the values of the stimuli, ranging from 10.4 cm to 11.6 cm. The total average (SD) of GOF scores were 0.94 (0.06) and 0.93 (0.06), for the food-deprived and nondeprived conditions, respectively.

An initial analysis that included gender as a between-subject factor did not reveal a main effect of gender or any interactions related to gender. We therefore collapsed the results across females and males in subsequent analyses. JND scores were submitted to a two-way repeated measures analysis of variance (ANOVA) with the within-subject factors of stimulus type (food vs. baseline) and condition (food deprived vs. nondeprived). As shown in Fig.  3 , the results provide a close replication of the results of Experiment 1 ; higher spatial sensitivity was found for the food related stimulus compared with baseline in the food-deprived condition. Such differences were not observed in the nondeprived condition.

JND scores for food-related and control stimuli, in the food-deprived and the nondeprived conditions in Experiment 2 . As in Experiment 1 , under food deprivation, perceptual resolution was higher for the food-related compared with the control stimulus. Resolution did not statistically differ for the two stimulus types in the nondeprived condition. Error bars represent 95% confidence intervals for repeated measures ANOVAs

A main effect of stimulus type [F(1,76) = 7.225, p < 0.01, η p 2 = 0.087] was found, with overall larger JNDs for the baseline compared with the food stimulus. The main effect of condition [F(1,76) = 0.131, p > 0.05, η p 2 = 0.002] was not significant. More importantly, a significant interaction was found between stimulus type and condition [F(1,76) = 15.062, p < 0.01, η p 2 η 2 p= 0.165], indicating that food deprivation led to a different pattern of sensitivity for the food-related compared with the control stimuli. Specific comparisons showed that JNDs were significantly lower for the food-related stimulus compared with the baseline stimulus in the food-deprived condition [t(76) = 4.099, p < 0.0001, d = 0.52]. JNDs for the food-related stimulus were not statistically different than baseline in the nondeprived condition [t(76) = 0.061, p > 0.05, d = 0.24]. Interestingly, JNDs were significantly lower for the food stimulus in the food-deprived compared to the nondeprived condition [t(76) = 2.067, p < 0.05, d = 0.25]. Conversely, JNDs were significantly higher for the baseline stimulus in the food-deprived compared with the nondeprived condition [t(76) = 2.593, p < 0.05, d = 0.25]. In the Bayesian ANOVA, the best-fitting model included condition, stimulus type, and their interaction as predictors [BF 10 = 16.13]. Moreover, BF 10 was 1.28 for a model with only two main effects. Accordingly, the model that included the interaction term was 12.6 times more likely compared with the model that included only the two main effects.

These findings extend and replicate the results found in Experiment 1 and show again that perceptual resolution for the size of food-related stimuli is modulated by food deprivation. The specific comparisons showed that while food deprivation led to a significant increase in perceptual resolution for food-related stimuli, it also led to a significant decrease in perceptual resolution for the neutral, food-unrelated stimulus. These findings suggest that attentional processes may have modulated the effect of food deprivation on perceptual resolution in a similar manner to the way attention has been shown to affect performance under spatial attention tasks (Montagna, Pestilli, & Carrasco, 2009 ). In particular, it can be argued that under food deprivation, participants allocate increased attentional resources to relevant, food related stimuli but also allocate less attentional resources (compared with the nondeprived condition) to food-unrelated, less attended neutral stimuli. We further elaborate about the possible role of attention in the General discussion section.

As expected, subjective hunger ratings were significantly higher in the food-deprived condition compared with the nondeprived condition, (5.18 and 2.27, respectively; t(76) = 26.161, p < 0.0001, d = 3.71). These results establish the relationship between hunger and food deprivation and confirm the experimental manipulation. We note, however, that such a relationship cannot confirm causality.

General discussion

The purpose of the current study was to examine the effect of mild hunger caused by food-deprivation on spatial resolution. The results show that motivational states can lead to an increased spatial resolution for domain-relevant stimuli. More particularly, participants' perception was more sensitive to differences in food size, when they were hungry following a mild period of food deprivation. The fact that mild food deprivation increases human resolution power for images of food-related stimuli presented on a computer screen provides a strong indication for the generality of the effect.

How food deprivation affects visual resolution: a possible role for attention

The current findings extend previous research in the field of size perception, concerned mainly with perceptual biases (difference between perceived food size and real food size) and attentional biases (Veltkamp et al., 2008 ). Previous research has shown that visual attentional processes could be modulated by motivational states that dictate goals and incentives. For example, Mogg, Bradley, Hyare, and Lee ( 1998 ) used the dot probe task to investigate whether hunger is associated with attentional and preattentive biases for food stimuli. Results showed a significant increase in selective attention to food-related words for food-deprived compared with nondeprived participants. These findings indicate attentional vigilance for stimuli that are relevant to the subject’s psychological state (Piech, Pastorino, & Zald, 2010 ). Similar findings were found among restrained eaters (Hollitt, Kemps, Tiggemann, Smeets, & Mills, 2010 ; Neimeijer, de Jong, & Roefs, 2013 ). Therefore, stimuli of relevant significance to the motivational state can increase arousal and potentially attract more attention compared with neutral stimuli (Balcetis & Dunning, 2006 ; Engelmann & Pessoa, 2007 ; Gu, Liu, Kyritsis, & Bondy, 2009 ; Piech et al., 2010 ).

We note, however, that unlike the vast majority of studies in the field of attention, that traditionally examined the allocation of spatial attention in situations in which attended and unattended stimuli are simultaneously presented within the same visual framework, the present study applied a different experimental design. More particularly, in the present set of experiments, the food-related and the food-unrelated stimuli were presented at different trials and at different time points. At a particular point in time, therefore, selective attention could always be fully allocated to the target stimulus. However, it is still possible that when participants were food deprived, they engaged larger attentional resources in trials in which food-related stimuli were presented compared with neutral trials. It therefore is intriguing to speculate that under the experimental settings of the present experiments, the greater amount of attention allocated to food stimuli led to an increase in spatial resolution. In a similar vein, studies of spatial selective attention showed that selective attention can lead to an increase in perceptual resolution to attended compared with unattended stimuli (Carrasco, Williams, & Yeshurun, 2002 ; Piech et al., 2010 ). Such an increase in spatial resolution has been suggested to result from the modulation of spatial frequency tuning neural populations across the visual field (Carrasco et al., 2002 ). Although spatial selective attention was not directly examined in the current study, it is possible that attentional mechanisms mediated the effects of food deprivation on visual resolution. As in studies of spatial attention that showed that the processing of attended stimuli is enhanced compared with unattended stimuli (Montagna et al., 2009 ), the results of Experiment 2 showed that food deprivation led to an increase in resolution for food related stimuli but also to a decrease in resolution for food-unrelated stimuli, which could be therefore regarded as “less attended” compared with food-related stimuli. Therefore, it is possible that the effects of food deprivation on performance in the present study were mediated by the allocation of attention (for a discussion, see Firestone & Scholl, 2015 ). However, due to the nature of the current experimental design, which did not directly manipulate attention, the possible relevance of selective attention to the present findings should be clarified in future research.

Unlike the effect food deprivation had on visual resolution in the present investigation, food deprivation did not lead to any distortions or biases in the average perception of food size. The scarce research in the field showed conflicting findings. This finding is congruent to several earlier studies; Carter and Schooler ( 1949 ) found no significant differences between rich and poor children's perception of coin size. Conversely, several studies have reported overestimation of stimuli size under certain motivational states and, more specifically, overestimation of food size following food-deprivation or under dietary states (Bruner & Goodman, 1947 ; Neimeijer et al., 2013 ; Piech et al., 2010 ; Veltkamp et al., 2008 ). Accurate perception can be critical for one's survival. This may explain why motivational states or factors would not always lead to perceptual biases or distortions; rather they may more consistently sharpen people’s perceptual sensitivity to increase processing efficiency (Tajfel, 1957 ).

Conclusions

The results of the present study show that food deprivation leads to an increased resolution power for food stimuli. In addition, the present study provides a basic psychophysical method to assess differences in size perception under different motivational states. By systematically investigating the effect of food deprivation and hunger on food size—perception in laboratory settings, it would be possible to gain further insights on how perception in normal subjects and in clinical populations suffering from eating disorders is modulated by motivational factors. Understanding such possible differences might help to shed a new light on the mechanisms that underlie normal perceptual processing of food-related stimuli.

Balcetis, E., & Dunning, D. (2006). See what you want to see: Motivational influences on visual perception. Journal of Personality and Social Psychology, 91 (4), 612–625. doi: 10.1037/0022-3514.91.4.612

Article   PubMed   Google Scholar  

Beasley, L. J., Hackett, A. F., Maxwell, S. M., & Stevenson, L. (2004). The effect of a dietary preload on estimation of usual food portion size by photograph in female volunteers. Journal of Human Nutrition and Dietetics, 17 (3), 219–225. doi: 10.1111/j.1365-277X.2004.00515.x

Benau, E. M., Orloff, N. C., Janke, E. A., Serpell, L., & Timko, C. A. (2014). A systematic review of the effects of experimental fasting on cognition. Appetite, 77, 52–61. doi: 10.1016/j.appet.2014.02.014

Berridge, K. C. (2004). Motivation concepts in behavioral neuroscience. Physiology and Behavior, 81 (2), 179–209. doi: 10.1016/j.physbeh.2004.02.004

Bruner, J. S., & Goodman, C. C. (1947). Value and need as organizing factors in perception. Journal of Abnormal Psychology, 42 (1), 33–44. doi: 10.1037/h0058484

Carrasco, M., Williams, P. E., & Yeshurun, Y. (2002). Covert attention increases spatial resolution with or without masks: Support for signal enhancement. Journal of Vision, 2 (6), 467–479. doi: 10.1167/2.6.4

Carter, L. F., & Schooler, K. (1949). Value, need, and other factors in perception. Psychological Review, 56 (4), 200–207. doi: 10.1037/h0060745

Engelmann, J. B., & Pessoa, L. (2007). Motivation sharpens exogenous spatial attention. Emotion, 7 (3), 668–674. doi: 10.1037/1528-3542.7.3.668

Firestone, C., & Scholl, B. J. (2015). Cognition does not affect perception: Evaluating the evidence for “top-down” effects. Behavioral and Brain Sciences, 4629 (2016), 1–77. doi: 10.1017/S0140525X15000965

Google Scholar  

Garner, D. M. (2004). Eating Attitudes Test (EAT-26): Scoring and interpretation. Http://Www.Eat-26.Com/Scoring.Php . http://doi.org/ Retrieved from: http://www.eat-26.com/downloads.php

Gu, J., Liu, Y., Kyritsis, A. P., & Bondy, M. L. (2009). Molecular epidemiology of primary brain tumors. Neurotherapeutics, 6 (3), 427. doi: 10.1016/j.nurt.2009.05.001

Article   PubMed   PubMed Central   Google Scholar  

Hollitt, S., Kemps, E., Tiggemann, M., Smeets, E., & Mills, J. S. (2010). Components of attentional bias for food cues among restrained eaters. Appetite, 54 (2), 309–313. doi: 10.1016/j.appet.2009.12.005

JASP Team. (2016). JASP (Version 0.8.0.0)[Computer software].

Kral, T., Roe, L., & Rolls, B. (2004). Combined effects of energy density and portion size on energy intake in women. The American Journal of Clinical Nutrition, 79 (6), 962–968. Retrieved from https://ajcn.nutrition.org/content/79/6/962.full

Lazarus, R. S., Yousem, H., & Arenberg, D. (1953). Hunger and perception. Journal of Personality, 21 (3), 312–328. doi: 10.1111/j.1467-6494.1953.tb01774.x

Loewenstein, G. (1996). Out of control : Visceral influences on behavior. Organizational Behavior and Human Decision Processes, 65 (3), 272–292. doi: 10.1006/obhd.1996.0028

Article   Google Scholar  

Mogg, K., Bradley, B. P., Hyare, H., & Lee, S. (1998). Selective attention to food-related stimuli in hunger: Are attentional biases specific to emotional and psychopathological states, or are they also found in normal drive states? Behaviour Research and Therapy, 36 (2), 227–237. doi: 10.1016/S0005-7967(97)00062-4

Montagna, B., Pestilli, F., & Carrasco, M. (2009). Attention trades off spatial acuity. Vision Research, 49 (7), 735–745.

Namdar, G., Avidan, G., & Ganel, T. (2015). Effects of configural processing on the perceptual spatial resolution for face features. Cortex, 72, 115–123. doi: 10.1016/j.cortex.2015.04.007

Neimeijer, R. A. M., de Jong, P. J., & Roefs, A. (2013). Temporal attention for visual food stimuli in restrained eaters. Appetite, 64, 5–11. doi: 10.1016/j.appet.2012.12.013

Piech, R. M., Pastorino, M. T., & Zald, D. H. (2010). All I saw was the cake. Hunger effects on attentional capture by visual food cues. Appetite, 54 (3), 579–582. doi: 10.1016/j.appet.2009.11.003

Ptacek, R., Stefano, G. B., Weissenberger, S., Akotia, D., Raboch, J., Papezova, H., … Goetz, M. (2016). Attention deficit hyperactivity disorder and disordered eating behaviors: Links, risks, and challenges faced. Neuropsychiatric Disease and Treatment, 12 , 571.

Radel, R., & Clement-Guillotin, C. (2012). Evidence of motivational influences in early visual perception: Hunger modulates conscious access. Psychological Science, 23 (3), 232–234. doi: 10.1177/0956797611427920

Tajfel, H. (1957). Value and the perceptual judgment of magnitude. Psychological Review, 64 (3), 192–204. doi: 10.1037/h0047878

Veltkamp, M., Aarts, H., & Custers, R. (2008). Perception in the service of goal pursuit: Motivation to attain goals enhances the perceived size of goal-instrumental objects. Social Cognition, 26 (6), 720–736. doi: 10.1521/soco.2008.26.6.720

Witt, J. K., & Dorsch, T. E. (2009). Kicking to bigger uprights: Field goal kicking performance influences perceived size. Perception, 38 (9), 1328–1340. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/19911630

Download references

Acknowledgments

The authors thank Yoav Kessler and Daniel Algom for their help and advice.

Author information

Authors and affiliations.

Department of Psychology, Ben-Gurion University of the Negev, Beer-Sheva, 84105, Israel

Noa Zitron-Emanuel & Tzvi Ganel

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Tzvi Ganel .

Rights and permissions

Reprints and permissions

About this article

Zitron-Emanuel, N., Ganel, T. The effect of food deprivation on human resolving power. Psychon Bull Rev 25 , 455–462 (2018). https://doi.org/10.3758/s13423-017-1296-6

Download citation

Published : 03 May 2017

Issue Date : February 2018

DOI : https://doi.org/10.3758/s13423-017-1296-6

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Food deprivation
• Psychophysics
• Visual perception
• Motivational effects
• Spatial resolution
• Find a journal
• Publish with us
• Track your research

Somalia: Unicef - Two-Thirds of Somali Children Face Extreme Food Deprivation

Mogadishu, Somalia -- In a stark revelation highlighting the deepening humanitarian crisis in Somalia, the United Nations Children's Fund (UNICEF) has reported that 63 percent of Somali children--equivalent to two out of every three--have experienced extreme food deprivation in their early years.

This alarming statistic underscores the severe and ongoing food insecurity crisis plaguing the region.

The report, which sheds light on the dire nutritional state of Somali children, also warns that up to 50 percent are at heightened risk of wasting, a severe form of malnutrition that can be fatal if left untreated. The situation is particularly grim for children aged between 6 and 23 months, a critical developmental period.

UNICEF's findings reveal that only 20 percent of children in this age group receive essential nutrients from eggs, fish, poultry, or meat as part of their diet. Even more concerning is that two-thirds of these young children do not consume any vegetables or fruits, key components of a balanced diet necessary for healthy growth and development.

Wafaa Saeed, UNICEF's representative in Somalia, expressed deep concern over the widespread food poverty among children, attributing the crisis to ongoing climate-related disasters and persistent conflicts. "The situation is worsened by ongoing climate-related crises and conflicts, which have left children vulnerable to both chronic and severe malnutrition," Saeed stated.

She emphasized the urgent need for collective action to improve the food system for young children, ensuring that families have access to a diverse range of locally available foods.

"Together with the government and other partners, we need to do more to improve the food system for young children and enable families to have access to a wide variety of locally available foods, especially fish, meat, fruits, and vegetables, which are currently limited in their diets," Saeed added.

Despite some progress in tackling food insecurity, Somalia remains deeply affected by recurring climatic challenges, including droughts and floods. These environmental stresses, coupled with ongoing conflict, insecurity, disease outbreaks, and widespread poverty, have exacerbated the humanitarian needs in the Horn of Africa.

According to estimates, around 4 million people in Somalia are currently facing crisis or emergency levels of food insecurity, with a staggering 1.7 million children at risk of acute malnutrition.

Among these children, approximately 430,000 are expected to suffer from severe malnutrition in 2024, a life-threatening condition requiring immediate intervention.

Sign up for free AllAfrica Newsletters

Get the latest in African news delivered straight to your inbox

By submitting above, you agree to our privacy policy .

Almost finished...

We need to confirm your email address.

To complete the process, please follow the instructions in the email we just sent you.

There was a problem processing your submission. Please try again later.

The UN Office further underscores the gravity of the situation for the Coordination of Humanitarian Affairs (OCHA), which has indicated that $1.6 billion is needed to fund the 2024 Humanitarian Response Plan (HNRP) in Somalia. However, as of August 2, only $507 million had been secured, leaving a significant funding gap that threatens to undermine relief efforts.

The crisis in Somalia is part of a broader global challenge, with UNICEF reporting that 181 million children worldwide are living in severe food poverty.

Of these, 65 percent are concentrated in 20 countries, including Somalia. In these countries, more than 80 percent of caregivers reported that their child had gone an entire day without eating, a sobering indicator of the widespread hunger affecting millions of children.

In addition to Somalia, other regions severely impacted by food poverty include South Asia, with 64 million affected children, and Sub-Saharan Africa, home to 59 million of these vulnerable young lives. These staggering figures highlight the urgent need for a coordinated global response to address child malnutrition and food insecurity.

Read the original article on Radio Dalsan .

• East Africa
• Sustainable Development
• Food and Agriculture

AllAfrica publishes around 500 reports a day from more than 100 news organizations and over 500 other institutions and individuals , representing a diversity of positions on every topic. We publish news and views ranging from vigorous opponents of governments to government publications and spokespersons. Publishers named above each report are responsible for their own content, which AllAfrica does not have the legal right to edit or correct.

Articles and commentaries that identify allAfrica.com as the publisher are produced or commissioned by AllAfrica . To address comments or complaints, please Contact us .

AllAfrica is a voice of, by and about Africa - aggregating, producing and distributing 500 news and information items daily from over 100 African news organizations and our own reporters to an African and global public. We operate from Cape Town, Dakar, Abuja, Johannesburg, Nairobi and Washington DC.

• Support our work
• Sign up for our newsletter
• For Advertisers

• © 2024 AllAfrica
• Privacy Policy

• Search Please fill out this field.
• Manage Your Subscription
• Give a Gift Subscription
• Newsletters
• Sweepstakes

This Popular Artificial Sweetener Is Linked to Heart Attacks and Strokes, Research Shows

The sugar substitute is typically used in flavored waters, ice cream, and other low-calorie treats.

Korin Miller has spent nearly two decades covering food, health, and nutrition for digital, print, and TV platforms. Her work has appeared in Women's Health, SELF, Prevention, The Washington Post, and more.

Food & Wine / Getty Images

There are a lot of artificial sweeteners available for people who want to have less sugar in their diet, but one, in particular, has been getting a lot of attention lately — and not for a good reason. It’s called erythritol, and it’s been most recently linked to blood clots. 

New research published in the American Heart Association journal Arteriosclerosis, Thrombosis, and Vascular Biology analyzed data from two small groups of people — 10 who had a beverage with 30 grams of the artificial sweetener erythritol and 10 who had a drink with 30 grams of sugar after fasting overnight. Their blood was drawn before having the drink and 30 minutes later. The researchers discovered that people who had the erythritol drink had more than double the risk of developing blood clots than those in the sugar group. (Blood clots can form in blood vessels, which can then break off and travel up to the heart or brain, causing a heart attack or stroke.)

This isn’t the first time the safety of erythritol has been questioned. A 2023 study published in Nature Medicine linked the sweetener to a higher risk of stroke, heart attack, and death.  

It’s important to point out that the research isn’t conclusive, given that these studies are small. But the findings are definitely raising eyebrows. Here’s what you need to know about erythritol, plus what it’s commonly found in. 

What is erythritol?

Erythritol is a form of carbohydrate called a sugar alcohol, explains Keri Gans, R.D., author of The Small Change Diet . “It is a common replacement for table sugar,” she says. 

Erythritol, which is approximately 70% as sweet as sugar, per the National Institutes of Health, “is naturally found in small amounts in some fruits and fermented foods,” says Scott Keatley, R.D., co-founder of Keatley Medical Nutrition Therapy . “Unlike regular sugar, erythritol provides a sweet taste without contributing significantly to calories or impacting blood sugar levels,” he adds. 

Erythritol is classified by the U.S. Food and Drug Administration (FDA) as generally recognized as safe or GRAS, meaning it’s considered safe when it’s used as intended. 

What is erythritol in?

Erythritol is in a wide range of goods as a sugar substitute. “It is especially popular in low-calorie, sugar-free, keto, and diabetic-friendly products,” Keatley says. “Its main appeal lies in its ability to mimic the sweetness of sugar without the associated calories, making it a popular ingredient for those managing weight or blood sugar levels.”

It’s usually found in sugar-free varieties of ice cream, candy, cookies, cakes, gum, and fruit spreads, Gans says. “It is also commonly used in stevia and monk-fruit sugar alternatives,” she adds. 

These are some of the most common products you’ll find erythritol, per Keatley:

• Bai flavored waters
• Lily's Sweets, including sugar-free chocolate bars and baking chips
• Swerve Sweets low-carb and keto-friendly baking mixes
• DC24 daily care chewing gum
• Truvia sweetener 
• Swerve sweetener
• Halo Top ice cream
• Quest nutrition bars

How can you know if your food contains erythritol?

This involves some reading. “The FDA does not require a package’s Nutrition Facts Label to identify the specific type of sugar alcohol it contains,” Gans says. “However, it is required that the type of sugar alcohol be on the list of ingredients. “

Erythritol should be listed under its name, Keatley says. “It is not typically listed under any other name, making it relatively easy to spot,” he adds. 

While the latest findings on erythritol aren’t great, experts recommend putting them into perspective. “Everything — even food — has risks and rewards,” Keatley says. “A low/no-calorie sweetener is a game-changer for people who are with or at risk for diabetes, obesity, and PCOS.”

But, he says, “we see with this and other research like it that our body does react to these substances and we need to consume them in small amounts and with less frequency.” Most studies on erythritol look at 30 grams of the sweetener — which is a lot at once, Keatley points out. “But if someone were to have a couple of beverages, a baked good, and some ice cream sweetened with erythritol in the same day, the accumulation may be over 30 grams,” he says. 

Keatley agrees that it’s best to limit how much of the sweetener you have. “I do not think you need to avoid erythritol completely, but you should limit daily consumption,” she says. 

If you rely on artificial sweeteners for sweet foods, Keatley recommends that you “eat around.” 

“Having a little aspartame, sucralose, saccharin, stevia, and allulose means that you're not using the same pathways and are reducing your risk of side effects,” he says.    

Related Articles

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

The Effects of Food Deprivation on Concentration and Perseverance

Related Papers

willy ocaya

Handbook of Behavior, Food and Nutrition

britta renner

Lifeisharsh101 Pine

This is my research of food deprivation

Physiology & Behavior

Allen Neuringer

Sabrina Hardenbergh

Laura Vanessa Zambrano Navarro

David Levitsky

Behavioural Processes

Telmo Peña-Correal

Physiology Behavior

Wayne Askew

Food and Nutrition Sciences

monica meza

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

RELATED PAPERS

Current Trends in diagnosis & Treatment

roopam bassi

Hollie Raynor

International Journal of Environmental Research and Public Health

Marshall D McCue

Journal of Biological Chemistry

Yosefa Avraham

Endocrinology

Nancy Levin , Liza Soriano

Clinical Medicine Insights: Case Reports

MICHELLE KATHERINE TAPIA MINA

Barbara Laraia

African Journal of Food, Agriculture, Nutrition and Development

Victor Temple

Physiology &amp; Behavior

Robin Kanarek

Psychonomic Bulletin & Review

American Journal of Clinical Nutrition

Enrique Jacoby

Journal of the American Dietetic Association

Janet Polivy

Catherine Ross , Terrence D Hill

Behavioral Interventions

Brian Gaunt

Marwan El Ghoch

Current opinion in clinical nutrition and metabolic care

Karl Friedl

Anastasi Kosmadopoulos

Alison Strack

The American journal of clinical …

Jennifer Perloff , L. Sagor

•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024