Do Certain Words Entice Us to Eat High-Calorie Foods?

Do Certain Words Entice Us to Eat High-Calorie Foods?

Researchers identify how food word cues influenced by both stress and genetics can be associated

with increased food desire and intake



November 3, 2015



Johns Hopkins University: Ekaterina Pesheva (Child and Adolescent Psychiatry):, 410-502-9433

Sarah Weber (Global Obesity Research Center):, 410-502-3332

The Obesity Society: Mollie Turner,


LOS ANGELES, CA: New research shows that brain responses to written food words differ between lean individuals and those with obesity, and suggests that both stress and genetics could influence excess eating.

The pair of studies led by Susan Carnell, PhD, member of The Obesity Society (TOS) and Assistant Professor of Psychiatry and Behavioral Sciences at Johns Hopkins University School of Medicine, reinforces the need to better understand how the external food environment interacts with our biology, and may aid the development of behavioral interventions to help individuals with obesity or those at high risk for the disease. The findings will be unveiled today during an oral presentation, and a poster presentation on Wednesday, Nov. 4, at The Obesity Society Annual Meeting at ObesityWeekSM 2015 in Los Angeles, CA.

In recent years, obesity researchers have greatly enhanced our understanding of “food cues,” which are internal or external environmental factors that influence the desire to eat. They come in many forms including emotions, images, smells, tastes and even food words. Food words could be considered a relatively minimal food cue compared with images or smells; however, because they are ubiquitous in advertising and other contexts they have significant potential to impact eating behavior.

In one study, the research team found that individuals with obesity were more likely to consume energy-dense foods (foods high in calories per unit of weight) compared to those of normal weight after experiencing stress. As seen in brain imaging scans, neural responses to high-calorie compared with low-calorie food words was also increased for individuals with obesity under both stressed and non-stressed conditions. To conduct the study, seventeen participants with obesity and 12 at normal weight underwent a functional magnetic resonance imaging (fMRI) scan during which they viewed words describing high-calorie foods, low-calorie foods and non-foods, and rated how much they wanted to eat each food item.

“Our study found that individuals with obesity had a stronger response to words associated with high-calorie foods - such as chocolate spread and chicken wings - in a widespread neural circuit spanning multiple areas of the brain,” said Dr. Carnell. “When we subjected individuals to a combined social and physiological stressor, both individuals with obesity and those of normal weight showed slightly altered responses to high-calorie food words, but only those with obesity ate more at a subsequent meal. This suggests that people with obesity show a consistently different response to mere words describing foods than lean individuals. This could contribute to excess intake of energy-dense foods in both stressful and non-stressful environments.”

In the second study, the research team identified an association between higher genetic obesity risk in teenagers based on several known obesity-associated genetic variants and subjective responses to food words. In addition, one specific genetic variant, MC4R, was associated with greater intake of high-calorie foods during a laboratory test meal, while another, FTO, was associated with lower scores on a questionnaire measuring self-regulation of food intake.

A genetic variant is a genetic difference that makes one individual or population different from another.

“We all have tiny differences in our genome that affect how we interact with the surrounding environment,” says Carnell. “While some of the genetic variants we see may have helped people maintain a healthy body weight in the past, they could now be working against us, making certain populations more susceptible to obesity and diabetes.”

To conduct the study, the research team genotyped 35 adolescents between 14 and 19 years old with varying familial risk for obesity. Subjective appetite responses to food and non-food words were measured using a computer paradigm, and food consumption was measured in a laboratory meal that followed. Participants also filled out a questionnaire measuring habitual self-regulation of intake.

“While we know that certain genetic variants are tied to obesity, our study provides additional insight into how these particular obesity-associated genetic variants may be working – by increasing appetite and food intake,” said Leora Benson, MS, research coordinator for the study. “The fact that many of these genetic variants act through eating behavior is exciting because behavior can be changed,” adds Carnell. “This research tells us that there may be ways we can help prevent individuals with these variants from developing obesity.”

The new understandings gleaned from these two studies could help the clinical research community identify behavioral treatment strategies based on reducing the impact of food cues, particularly for those at high risk for obesity.

“It may be possible to train our brains to react differently to certain food cues,” said Martin Binks, PhD, FTOS, Secretary Treasurer of and spokesperson for The Obesity Society. “This research is a step toward better understanding how food words – relatively minimal food cues – may influence food consumption and how other common experiences like stress may interact with associated food cues to influence eating behavior. These types of studies may eventually lead to more effective behavioral strategies.”

The Obesity Society calls for more research into neurohormonal regulation of eating behavior. Both abstracts are available below.

# # #

This press release can be published in full or in part with attribution to The Obesity Society.



Susan Carnell, PhD, Johns Hopkins University School of Medicine


Neural Responses to Food Words in Obese and Lean Individuals Under Stressed and Non-Stressed Conditions


Obese individuals may show heightened responses to environmental food cues, and stress has been associated with greater intake and weight. We aimed to study neural responses to minimal food cues (written words) under normal conditions and following a stressor in obese vs. lean individuals.


We recruited 12 lean and 17 obese participants. On two separate days, participants underwent an fMRI scan during which they viewed words representing high energy-density [ED] foods, low-ED foods, and non-foods and rated how much they wanted to eat each food item. On one day, the scan was preceded by a Socially-Evaluated Cold Pressor Test [SECPT] (stress); on the other (counter-balanced) by a warm water task (control). A multi-item ad libitum meal followed each scan.


Wanting scores were highest for the high-ED foods under both conditions (p<0.001), and obese, but not lean, individuals consumed more calories during the stress (vs. control) condition (p=0.018). Imaging analyses for the non-stress condition revealed that obese (vs. lean) individuals showed increased responses to food (vs. non-food) cues and high-ED (vs. low-ED) food words in multiple cortical and sub-cortical brain regions including the caudate, dlPFC, cingulate, sensorimotor cortex, SMA and brainstem. In comparisons of the stress vs. control condition, both obese and lean individuals showed increased food vs. non-food and high-ED vs. low-ED responses in distinct but overlapping neural circuits.


Our results suggest that obese vs. lean adults show heightened neural responses to minimal food vs. non-food and high-ED vs. low-ED food word cues under normal, non-stressed conditions. Both obese and lean adults showed some evidence for increased neural food cue responses following a stress manipulation, but only obese individuals consumed more at a subsequent ad libitum meal. Heightened neurobehavioral responses to food cues in both stressed and non-stressed conditions could contribute to excessive intake and weight gain.


Carnell S. Johns Hopkins University School of Medicine. Poster abstract presentation at: The Obesity Society Annual Meeting at ObesityWeekSM 2015; November 2-6, 2015; Los Angeles, CA.



Leora Benson, MS, Johns Hopkins University School of Medicine


Genetic obesity risk and appetite in adolescents


Common obesity-associated genetic variants have been identified via genome-wide association studies, and many of the genes involved are expressed in the brain. However, relatively little is understood about the behavioral mechanisms by which these variants influence body weight.


We genotyped weight-associated SNPs in FTO (rs9939609), MC4R (rs17782313), NEGR1 (rs2815752) and TMEM18 (rs6548238) in 35 adolescents between 14 and 19 y old (20 F; 15 M), with a mean BMI percentile of 64 ± 28 (range: 12-99), recruited for varying familial (i.e. both genetic and environmental) obesity risk based on maternal weight. We assessed subjective appetitive responses to food and non-food words using a novel computer paradigm, as well as ad libitum intake in a laboratory meal. Participants also completed the Satter Eating Competence Inventory, a questionnaire measuring habitual self-regulation of intake.


Higher genetic risk scores based on the number of risk alleles in all four variants were associated with greater ‘wanting’ responses to food words compared with non-food words in the computer paradigm (r=0.37; p=0.030). Analyses of individual variants revealed that the FTO risk allele was associated with lower scores on questionnaire measures of intake self-regulation (p=0.044); and the MC4R risk allele was associated with greater ad libitum intake of high energy-density foods (p=0.036).


Our results suggest that common obesity-associated genetic variants influence weight via multiple appetitive endophenotypes, including a heightened subjective desire to eat when exposed to simple food word cues. This genetic effect on appetite was evident even in this small sample. Targeted behavioral and environmental interventions may help to limit the impact of increased appetitive responses and diminished self-regulation of intake in those at raised familial/genetic obesity risk.


Benson L. Johns Hopkins University School of Medicine. Poster abstract presentation at: The Obesity Society Annual Meeting at ObesityWeekSM 2015; November 2-6, 2015; Los Angeles, CA.

About The Obesity Society

The Obesity Society (TOS) is the leading professional society dedicated to better understanding, preventing and treating obesity. Through research, education and advocacy, TOS is committed to improving the lives of those affected by the disease. For more information visit: Connect with us on social media: Facebook, Twitter and LinkedIn. Find TOS disclosures here.

About ObesityWeek 2015

ObesityWeek is the premier, international event focused on the basic science, clinical application, prevention and treatment of obesity. TOS and the American Society for Metabolic and Bariatric Surgery (ASMBS) host the world’s pre-eminent conference on obesity, ObesityWeek 2015, Nov. 2-6, at the Los Angeles Convention Center in Los Angeles, California. For the third year, both organizations hold their respective annual scientific meetings under one roof to unveil exciting new research, discuss emerging treatment and prevention options, and network and present. Connect and share with ObesityWeek on Twitter and Facebook, or by using #OW2015.