
(Photo by juliaap on Shutterstock)
Sugar May Limit Memory Recovery After a Junk Food Diet
In A Nutshell
- Rodents switched from a fat- and sugar-heavy diet to healthy chow scored better on memory tests than rodents kept on junk food.
- They still trailed animals that ate healthy food all along, though a single study drove that gap.
- Memory gains showed up with high-fat diets but not clearly with high-sugar or combined diets.
- Mood, anxiety, and activity tests showed no reliable change, and every result comes from rodents, not people.
Swapping fries and soda for salad raises a fair question: how much lost memory comes back? A review of rodent experiments, published in Nutritional Neuroscience, found that animals moved from a diet heavy in fat and sugar to plain healthy chow scored better on memory tests than animals kept on the junk diet. They still trailed animals that had eaten healthy food all along.
Researchers at the University of Technology Sydney pooled results from 27 studies in rats and mice that fed animals diets high in fat, sugar, or both before returning them to standard chow. Tests of anxiety, depression-like behavior, motivation, and activity showed no consistent change after the switch.
Two cautions apply. Rodents are not people, so none of this shows how human memory responds to a diet overhaul, even though earlier human studies link diets heavy in fat and sugar to weaker memory. And the memory shortfall appeared only within the study periods, so nothing here proves it is permanent.
Researchers Screened 3,291 Articles to Find 27 Rodent Studies
Searches of five scientific databases turned up about 6,300 records. After duplicates were removed, 3,291 articles were screened, and 27 studies made the cut. Each gave rodents constant access to a high-fat or high-sugar diet for at least two weeks, then switched them to standard chow before testing memory or behavior. Studies with genetically altered animals, on-and-off junk food access, food restriction, or stress were excluded.
Animal experiments let researchers control what the animals eat far more tightly than human diet studies can. Rats appeared in 13 studies and mice in 14, and 21 used only male animals, with diets ranging from lard-heavy chow to cafeteria-style spreads of bacon, pastries, and sugar water. In the memory studies, animals averaged about nine weeks on the unhealthy diet and about six weeks back on healthy chow. Using earlier rat-to-human age estimates, the authors say that could correspond to more than five years of poor eating followed by about three and a half years of healthier eating, though they call extrapolating to humans difficult and imprecise.
Memory Improved After Rodents Left a Junk Food Diet
Fourteen studies compared 385 animals switched to healthy chow with 402 that stayed on the junk diet. Switched animals scored reliably better. The clearest gains came from the novel object location test, in which a rodent that remembers where an object used to sit spends extra time investigating it after it has been moved. That task leans on the hippocampus, a brain region central to learning and memory that the authors describe as particularly sensitive to diet changes.
Diet type mattered. Gains showed up in studies using high-fat diets but not in those using high-sugar or combined fat-and-sugar diets. A result that misses the statistical cutoff does not prove that sugar blocks recovery, though the authors raise that as one possible interpretation. Longer time on the junk diet before the switch predicted bigger gains. The authors doubt that simply means longer exposure causes bigger problems to recover from, since earlier studies they cite found memory slips within a day or a week of starting a junk diet. Time spent on healthy chow and the animals’ sex did not change the results.
Why memory improved remains unsettled. Studies of brain inflammation, a possible explanation, gave mixed results, with one finding a drop in the hippocampus after the switch and another finding none there.
Memory Still Trailed Animals That Never Ate Junk Food
Sixteen studies compared 644 switched animals with 765 that ate healthy chow throughout. Switched animals scored lower on memory, but the authors urge caution. That gap was driven by a single study using a more demanding test, one that asks animals to link objects with the places where they appeared. Other memory tests showed no clear gap. In most studies, animals spent longer on the junk diet than on healthy chow afterward, which may explain the leftover shortfall.
Mood and Activity Tests Showed No Clear Change After the Switch
Switching diets produced no reliable change in anxiety-like behavior, depression-like behavior, or physical activity, whether animals were compared with those left on junk food or those that always ate healthy food. Motivation, measured by how hard an animal works for food rewards, was examined in only two studies, and only against animals that always ate healthy food. Too few studies existed to compare it with animals kept on the junk diet.
Those results are hard to read. No difference from junk-fed animals means no clear improvement, while no difference from always-healthy animals means no clear leftover harm, so these tests cannot show that recovery is impossible. The authors offer one possible explanation for anxiety: switching to blander food may raise stress while easing brain inflammation may lower anxiety, so the two effects could cancel out.
Rodent evidence favors switching over staying put and leaves open whether longer stretches of healthy eating would close the remaining memory gap. Prevention still looks like the safer bet, and the authors call for “strategies designed to reduce intake of diets high in sugar and saturated fat.”
Disclaimer: This article is for general information only and is not medical or nutritional advice. The findings come from animal research and may not apply to people. Anyone with questions about diet or memory should talk with a doctor or registered dietitian.
Paper Notes
Limitations
All 27 studies used rats or mice, so the findings may not carry over to people, and 21 of them tested only male animals. Some categories rested on few studies, particularly depression-like behavior (four to five studies) and motivation (two studies), and results varied moderately to widely between studies in most categories. The pooled analysis assumed a fixed correlation between measurements taken from the same animals rather than using measured data, a standard but acknowledged shortcoming. Publication bias could not be fully ruled out. Statistical tests found little evidence of it in most analyses, though the depression analysis against junk-fed animals showed a signal of small-study effects. The memory shortfall relative to healthy-chow animals was driven by one study using a single test, and most studies had longer junk-diet periods than recovery periods. The review could not compare specific sugar and fat sources, control diets ranged from grain-based chow to purified diets, and species differences could not be sorted out. Many studies did not fully report how animals were assigned to groups or whether testers knew each animal’s diet. The authors could not pin down the biological pathways behind memory recovery, and they note that rodent studies cannot fully model the psychological processes that shape diet-related mood changes. Across all measures, far more comparisons used healthy-chow controls (117) than junk-fed controls (68).
Funding and Disclosures
Funding came from an Australian Research Council Discovery Project grant (DP220103462), and Michael D. Kendig is supported by an Australian Research Council DECRA Fellowship (DE230101025). The authors reported no potential conflicts of interest.
Publication Details
Paper title: Cognitive and behavioural effects of high-fat, high-sugar diet reversal: a systematic review and meta-analysis of animal studies. Authors: Simone Rehn, Isabella Eikelis, Joanne M. Gladding, Laura A. Bradfield, and Michael D. Kendig, all of the School of Life Sciences, University of Technology Sydney, Ultimo, Australia. Bradfield also holds an affiliation with the School of Psychology, University of Sydney. Journal: Nutritional Neuroscience: An International Journal on Nutrition, Diet and Nervous System, volume 29, issue 11, pages 1412-1433, published online May 17, 2026, by Informa UK Limited, trading as Taylor & Francis Group. DOI: https://doi.org/10.1080/1028415X.2026.2664635. Corresponding author: Michael D. Kendig ([email protected]). The article is open access under a Creative Commons Attribution License (CC BY 4.0).







