Skip to main content

Brain scans link high GI diet to food cravings

Thursday 27 June 2013

"Chips and white bread trigger cravings in the brain," The Daily Telegraph reports.

It's a weakness that many of us are guilty of – you pick up a tube of Pringles intending to have a few and before you know it you've scoffed half the tube. But could chip butties or crisps really be as addictive as heroin or cigarettes, as the Mail Online suggests?

The stories in the Telegraph and the Mail Online are based on a very small study that used brain scans to look at the effects of high and low glycaemic index (GI) meals on activity in the brain. GI is a measure of the effect different foods have on blood sugar levels. The higher the GI level, the quicker the food leads to a spike in blood sugar levels. Many foods with a high GI level tend to be both energy rich and nutritionally poor, such as crisps, chips and white bread.

Researchers found that four hours after eating a high GI meal, blood flow in areas of the brain associated with “reward and craving” behaviour was greater than after eating a low GI meal.

These results seem to imply that these types of foods cause physical cravings in the same way as cigarettes. However, this study is too small to draw any firm conclusions from as it only included 12 men.

healthy, balanced diet should contain a variety of foods including fruit and vegetables, carbohydrates and foods containing protein and fat.

Where did the story come from?

The study was carried out by researchers from Boston Children’s Hospital, the Beth Israel Deaconess Medical Center, Brigham and Women’s Hospital and Harvard Medical School in the US, and from Ulm University, Germany. It was funded by grants from various public and charitable bodies.

The study was published in the peer-reviewed American Journal of Clinical Nutrition.

This very small study, in a specific population sample, does not show that certain foods are addictive, as the Mail Online website claims.

What kind of research was this?

This was a randomised crossover trial that looked at the effects of high or low GI foods on brain activity in the hours after eating. The crossover design means that all people in the trial receive both interventions (in this case low and high GI foods) but they are allocated to receive them in a random order.

Researchers sometimes use this design when the trial has a very small population size. This is because any results given by a randomised controlled trial (RCT) with this number of participants would have very little ‘statistical weight’. The smaller the sample size in an RCT, the higher the chance that the results occurred by chance.

If the researchers had used an RCT design for this very small trial they would only have had six people in the high GI group and six in the low GI group. Using the crossover approach increased the number to 12 in each group. However, the results of small crossover trials should also be viewed with a great deal of caution as results could still be due to chance.  

The researchers say that although the taste of food influences eating behaviour, little is known about what happens in the brain when different types of food are consumed. They point out that certain regions of the brain are regulated by the brain chemical dopamine, and these play a key role in “reward and craving”. These brain regions appear to be involved in the response to food. The researchers suggest that low and high GI diets may have different effects on this system.

The glycaemic index of food was originally developed to find out which foods were best for people with diabetes, but more recently there has been interest in the role of GI in diet generally. One theory says that foods with a low GI, such as most fruit and vegetables are healthier and better for weight control.

What did the research involve?

The researchers recruited a random selection of overweight or obese men who were otherwise healthy, and who were aged 18 to 35. The researchers excluded from their trial those with major medical problems. They also excluded those who used medication that affected appetite or weight, those who used tobacco or recreational drugs, and those who were in a weight-loss programme or who had high levels of physical activity. The men’s height and weight were measured before the experiment. Researchers also collected other details such as the men’s ethnicity.

The men were randomised to receive one of two types of test meals in the form of milkshakes. Each had the same number of calories, taste and sweetness. Both test meals had similar ingredients – such as egg white or corn syrup – but in different proportions. The only difference between the two meals was the measure of their glycaemic index, with one type having a high, and one a low GI.

The men had their test meal after having fasted for 12 hours or more, and abstained from alcohol the previous evening. They consumed the test meal within five minutes.

Researchers took blood samples before and every 30 minutes after the test meal during the following five-hour period. They also asked the men whether they found the meal “tasty” and asked how hungry they felt. The men could choose responses ranging from “not hungry at all” to “extremely hungry”.

Four hours after the test meal, the men underwent a special type of MRI scan called a functional MRI, which looks at the blood flow through the brain to see which areas of the brain are active.
After an interval of 2–8 weeks, the men swapped over, so that those who had consumed the high GI meal now consumed the low GI meal and vice versa. Researchers carried out the same procedures.

They analysed their results using standard statistical techniques.

What were the basic results?

Twelve men took part in the study with an average age of 29 years and an average BMI of 32.9.

The researchers found that four hours after each meal, compared with those who had consumed the low GI meal, those who had consumed the high GI meal had lower blood sugar levels (average 4.7mmol/l compared with 5.3mmol/l) and reported greater hunger.
At the same time, those who had eaten the high GI meal had greater brain activity in a region of the brain associated with “reward and craving” behaviour (the right nucleus accumbens), which spread to other areas including those associated with the sense of smell.

How did the researchers interpret the results?

The researchers suggest that high GI food may foster food cravings, in particular for the same type of foods, and may cause cycles of overeating. They also point out that the region of the brain affected is “crucially involved” in substance abuse and dependence, raising a question about whether certain foods might be addictive. Limiting high GI foods could help obese individuals reduce cravings and the urge to overeat, they suggest.


This study involved only 12 men, so its findings should be viewed with considerable caution. 

The brain is a complex organ and functional MRI can only provide a very crude estimation of the way it works. It is unclear whether the increase in blood flow observed in certain parts of the brain after a high GI meal was directly associated with feelings of food cravings in the men. 

It is well known that certain carbohydrates produce a surge in blood sugar, followed by a crash and feelings of hunger some hours later. Low GI foods, such as wholegrains can keep people feeling fuller for longer. This study lends support to this popular belief. However, alone it provides no proof that high GI foods are addictive and lead to food cravings.

The advice that people trying to lose weight should limit their consumption of high GI food is valid. Try swapping high GI food such as chips and crisps for lower GI foods such as fruit, vegetables, beans and wholegrains. Read more about healthy food swaps that you can make.

Analysis by Bazian
Edited by NHS Website

Links to the headlines

Chips and white bread trigger cravings in the brain

The Daily Telegraph, 26 June 2013

Food really is addictive: Study finds brain activity similar to heroin users after eating certain processed foods

Mail Online, 26 June 2013

Links to the science

Lennerz BS, Alsop DC, Holsen LM, et al.

Effects of dietary glycemic index on brain regions related to reward and craving in men

American Journal of Clinical Nutrition. Published online June 26 2013