Food and diet

Low-carb diets and memory

“Atkins-style low carbohydrate diets ‘can damage the memory,’” claims The Daily Telegraph . The newspaper says that in just one week slimmers could develop memory loss caused by following diets that avoid carbohydrate-rich foods such as bread and pasta.

The news story is based on a study testing 19 women’s performances in complex mental tests during a three-week diet. The women chose to follow either a low-carbohydrate Atkin’s-style diet, or a balanced reduced-calorie one. Women on the Atkin’s-style diet performed worse in some tests than those on the balanced diet. The newspaper says this is because their brains lacked glucose, which is restricted in Atkin's-style diets.

The study is very limited by its small sample size, which increases the possibility its findings occurring by chance. Also, the fact that the women chose which diet to follow means other factors may differ between the two groups and account for their different results.

Although short-term memory in the low-carbohydrate group was impaired on some tests, other cognitive measures did not differ between the two groups. The relevance of the different results between the groups is uncertain, particularly with the short follow-up, but does highlight the need for further research to see how dieting might affect short-term memory and awareness.

Where did the story come from?

This research was carried out by Kristen E D’Anci and colleagues of Tufts University, Medford, US, and the Jean Mayer USDA Human Nutrition Research Center on Aging. No sources of funding are reported. The study was published in the peer-reviewed medical journal Appetite .

What kind of scientific study was this?

This was a non-randomised controlled trial designed to investigate how low-carbohydrate diets affect cognition, the ability to perceive, reason and remember.

It has been found that calorie restriction can affect planning and functioning, due to preoccupying thoughts about food. Low-carbohydrate diets in particular can cause cognitive deficits due to, in theory, a lack of the energy from blood glucose, which is normally produced in the breakdown of carbohydrates.

Researchers recruited 19 women aged between 22 and 55 to take part in their three-week study on diet. All women were healthy, without depression or any other psychological condition, did not have heart disease or diabetes and were not taking medication.

The women in this study were given the choice of whether they wanted to follow a low-carbohydrate (LC) diet or a low-calorie macronutrient balanced (ADA) diet in line with the American Dietetic Association guidelines. The LC diet was chosen by nine women, while 10 women chose the ADA diet.The LC diet involved one-week of zero carbohydrates, a second week where they introduced 5 to 8g of carbohydrate per day, and in the third week they increased this to 10 to 16g of carbohydrate per day. The caloric content of the ADA diet was determined according to the individual’s current weight.

Before starting their diets, the women took part in a testing session that involved being weighed, a Profile of Mood States Questionnaire, a hunger questionnaire and five computer-based cognitive tests. These cognitive tests assessed visual and spatial memory, recollection of forward and reversed number sequences, vigilance, and positive and negative consequences of food preoccupation (which involved matching food and non-food word combinations).

The test sessions were repeated after 48 hours, one week, two weeks, and three weeks, into the diets. The women kept food diaries that were reviewed by a researcher to check compliance. The test results were then statistically compared between the two diet groups.

What were the results of the study?

The women in the two groups did not differ in body mass index (BMI) (28-30kg/m2) prior to dieting. After three weeks there was no significant difference in weight loss between the groups (LC group in total lost 1.88kg, ADA diet lost 1.76kg).

There were no differences between the groups on the hunger questionnaire or food preoccupation at the beginning of the study, but there were some differences on the tests of food preoccupation in later testing. Women in the ADA group showed a consistent and faster response in matching non-food pairs than food pairs which improved with testing at one, two and three weeks (indicating a practice effect). Women in the LC group showed no difference in matching food and non-food pairs and no improvement in time scores with each weekly testing.

Before dieting there were no differences in the performance of the number sequencing tests between the two diet groups. At week one, those women in the LC group recalled significantly fewer digits on the reversed number sequence test than those in the ADA group.

Prior to dieting women in the LC group scored better on the short-term visual/spatial memory test than the ADA group, but there was difference in testing at one week when women in the ADA group did significantly better than the LC group. There was no difference in long-term memory.

The vigilance tests were complex and involved no differences on some measures better response to targets in the LC groups compared to the ADA group, suggesting improved attention.

On the Profile of Mood States Questionnaire, women in the    ADA group showed greater confusion at the one- and two-week sessions.

What interpretations did the researchers draw from these results?

The researchers concluded that the two popular weight loss diets demonstrated no differences in weight loss over the three-week period. They noted that at one week (the period of no carbohydrate intake and low glycogen scores for the LC group) those on the LC diet showed impaired short-term memory in the visual/spatial and reversed number sequence tests than those in the ADA group.

What does the NHS Knowledge Service make of this study?

This research was designed to investigate the effects upon cognitive performance of low-carbohydrate diets compared to a more balanced low-calorie diet. However, the study does have important limitations:

  • With only 19 participants, the study was extremely small and any differences seen between the groups may have occurred by chance only.
  • As the women selected their own diet type rather than being randomly allocated one, there is the possibility of differences between the two groups may have actually accounted for some of the differences seen.
  • A number of complex tests were performed, and there were variable effects seen on different measures. It is uncertain how representative these tests may be of the women’s general daily functioning and memory, i.e. the relevance that they would have to everyday life.
  • It should be noted that the women on the LC diet did not perform worse than the ADA diet group on all measures, and that the women on the ADA diets demonstrated impaired attention and confusion on some measures.
  • The study was only short, and a longer study would have given a more definite indication of the effects of the diets over time. In particular, it would be important to see whether a zero carbohydrate diet maintained for a period of greater than one week would have had greater effect on the cognitive tests.

Despite the limited conclusions that can be made from this study, it highlights the need for further research to see whether low or no carbohydrate diets to put cognitive function at risk.

Sir Muir Gray adds...

The safest thing is to concentrate on taking more exercise: try walking an extra hour a day if you want to lose weight.


NHS Attribution