
(© Iryna - stock.adobe.com)
Full But Still Craving? Scientists Say Your Brain May Be To Blame
In A Nutshell
- Even after eating a specific food to fullness, participants’ brains showed no reduction in reward response when images of that food appeared on screen.
- Consciously, participants felt full and made choices reflecting that. Their neural activity told a completely different story.
- The brain’s habit-driven reward system appears to operate independently of the body’s satiety signals, meaning hunger status alone may not be enough to shut off cravings triggered by food images.
- This study does not prove that food cues directly cause overeating, but the findings raise serious questions about what constant exposure to food imagery does to an already full brain.
Scrolling through Instagram after dinner and suddenly craving the pizza in an ad is not a personal failing. Research suggests it may be baked into how the human brain works, and raises fresh questions about what constant exposure to food images is actually doing to us.
Research from the University of East Anglia and the University of Plymouth found that even after people ate a specific food to the point of fullness, the brain’s reward circuitry showed no meaningful drop in activity when images of that food appeared on screen. Participants said they were full. Their choices in a game-based task reflected that. Their brain activity, on the other hand, did not budge. The neural system that had learned to want the food simply did not register that the meal was over.
“Food cues may thus serve as an entry point for over-eating in otherwise healthy individuals,” the researchers wrote in the journal Appetite. For anyone who has ever reached for a snack ten minutes after finishing dinner because something on television looked good, that finding hits close to home.
Why the Brain Ignores a Full Stomach
The brain runs on more than one decision-making system. One is deliberate and goal-directed, weighing current conditions and adjusting behavior accordingly. Another is automatic and habit-driven, responding to familiar stimuli based on what has been rewarding in the past. Food images, through years of repeated exposure, become deeply wired into that second system.
The habit system is fast and largely invisible. It is also, based on these findings, indifferent to whether the body needs more food. The brain’s reward response to food images ran independently of whether participants had just eaten the food pictured. That failure of fullness to quiet a learned reward response is often associated with habitual rather than goal-directed behavior.
In practical terms, the conscious mind knows food is no longer needed. The habit system keeps treating food images as rewarding anyway. Anyone trying to resist a late-night snack is not simply making a choice. They may be fighting a neural response that satiety cannot reach.

Inside the Study: EEG Caps and Earned Snacks
Ninety students at the University of Plymouth took part, with 76 providing usable data after technical exclusions. All were between 18 and 29 years old with BMIs in the healthy to moderately overweight range. People with eating disorders, neurological conditions, or those currently dieting were excluded to keep the sample as unaffected as possible.
Participants arrived hungry and were fitted with EEG caps, electrode-covered headsets that track the brain’s electrical activity with millisecond precision. They rated their hunger and the personal appeal of eleven different foods. Researchers selected two with similar ratings for each participant, one sweet and one savory.
Participants then played a computerized decision-making game designed to earn them access to those foods. Successful choices ended with a photo of the target food; unsuccessful ones showed an empty plate or cupboard. The brain’s response to each image was recorded continuously. Midway through the session, participants ate one of their two foods, requesting more helpings until they no longer wanted any. Hunger and desirability ratings were collected again. Then they went back to the game.
The Brain Scans Tell a Different Story
Consciously, participants responded exactly as expected. Ratings of the food they had eaten dropped sharply after the meal. During the second half of the game, they steered choices toward the food they had not eaten, a clear behavioral sign the sated food had lost its appeal.
The brain, however, told a different story.
Researchers focused on a signal called the reward positivity, an electrical spike that appears roughly a quarter of a second after a person sees something rewarding. In established neuroscience, this signal reflects how rewarding the brain treats a stimulus in the moment. If satiety had registered neurally, the reward positivity triggered by images of the eaten food should have weakened after the meal. It showed no change at all.
No areas within the EEG measures the researchers examined showed a meaningful reduction in reward response to the sated food’s images. Using a statistical tool called Bayes factors, which quantifies the strength of evidence for competing conclusions, researchers found moderate to strong evidence across wide areas of the brain in favor of what they called devaluation insensitivity. The learned neural response to food images had not faded, even in areas most directly tied to reward processing.
“The absence of any effect of the devaluation procedure in the EEG immediately following image presentation, despite clear behavioural and subjective effects of devaluation,” the researchers wrote, “raises the possibility that the food’s motivational value is not well represented in this time window.”
To be clear, this study does not prove that food cues caused overeating in participants, nor does it establish that habits were directly responsible for the persistent brain signals observed. What it does show is a measurable disconnect between what the body signals after eating and how the brain continues to respond to food images in the short window following their appearance.

What This Means for a World Saturated in Food Images
Other research has shown that food cues can measurably influence behavior. Overweight children eat more after exposure to food cues. Obese children recognize more food-related advertisements than their non-obese peers. Some epidemiological studies also suggest overeating may be more learned than inherited, and that caloric intake is the strongest driver of BMI variation among European adults.
While this study did not test advertising or social media exposure directly, its findings point toward a possible mechanism behind why food imagery is so hard to ignore on a full stomach. The brain’s habit-driven reward response appears to run on its own track, one that operates separately from the body’s satiety signals. Every food photograph on social media, every fast food billboard on the commute home, every recipe video that autoplays before bed may be reaching a system that hunger status alone cannot shut off.
Participants in the study did adjust their choices after eating, which shows that conscious, goal-directed control can still influence behavior even when the underlying neural response has not faded. But the gap between what the brain signals and what the body actually needs raises harder questions about the food environment people navigate every day. Blaming individuals for caving to cravings after a full meal may be missing the point. The brain, it turns out, was never fully on board with stopping.
Disclaimer: This article is based on a single laboratory study conducted with a healthy young adult population and should not be interpreted as medical or dietary advice. The findings reflect early-stage research into how the brain processes food cues and do not establish direct cause-and-effect relationships between food imagery and overeating. Consult a qualified healthcare professional for guidance on nutrition, eating behavior, or weight management.
Paper Notes
Study Limitations
The researchers flag several important caveats. Before the meal, participants completed many trials in which food images were framed as the correct, desirable outcome. That framing may have trained a response to food images as inherently correct rather than inherently appetizing, and that association could have carried into the post-meal stage regardless of hunger. The researchers also acknowledge they cannot rule out that the persistent brain signal reflects motivational salience, a general sense of alertness or importance, rather than reward value specifically. The absence of an aversive outcome condition limits the ability to separate those interpretations. Additionally, the EEG measures used captured neural activity only within the first 700 milliseconds following image presentation and do not speak to later cognitive control processes that may also shape eating behavior. The two-step task used to distinguish habitual from goal-directed learning has also faced criticism in the scientific literature for limitations in measuring model-based learning precisely, and no significant correlations were found between the task’s learning-style measure and any of the devaluation outcomes.
Funding and Disclosures
The authors declare no known competing financial interests or personal relationships that could have influenced the work. No external funding source is identified in the paper. The study received ethics approval from the School of Psychology research ethics committee at the University of Plymouth, and all participants provided informed consent.
Publication Details
Authors: Thomas D. Sambrook (School of Psychology, University of East Anglia), Andy J. Wills, Ben Hardwick, and Jeremy Goslin (School of Psychology, University of Plymouth). | Title: “Devaluation insensitivity of event related potentials associated with food cues” | Journal: Appetite, Volume 218 (2026), Article 108390 | DOI: https://doi.org/10.1016/j.appet.2025.108390 | Received: August 1, 2024; Revised: November 13, 2025; Accepted: November 20, 2025; Available online: November 20, 2025 | Published open access under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).







