What Properties Make a Food "Addictive"?

Although the concept of food addiction remains controversial, there's no doubt that specific foods can provoke addiction-like behaviors in susceptible people.  Yet not all foods have this effect, suggesting that it's related to specific food properties.  A new study aims to identify the properties that make a food "addictive".

Introduction


Common sense tells us that one of the reasons we eat too much is that we're often exposed to food we find extremely seductive.  In common parlance, we refer to these foods as "addictive", which may not be too far off the mark.  A growing number of researchers propose that some of us really are addicted to certain foods, via a process that resembles drug addiction.  In other words, certain foods can exert such a strong pull on our motivational systems that they cause us to make decisions that are self-destructive*.

To define food addiction, Kelly Brownell and Ashley Gearhardt designed a tool called the Yale Food Addiction Scale (YFAS) (1).  It's a questionnaire that's based on established diagnostic criteria for other types of addiction, and the authors describe it as "a sound tool for identifying eating patterns that are similar to behaviors seen in classic areas of addiction".  Brownell and Gearhardt's research shows that many people exhibit behaviors toward food that resemble drug addiction, and that this addiction-like behavior is more common among binge eaters and people with obesity (2).  They argue that food addiction is a real clinical entity and should be treated as such.

Yet, people virtually never get addicted to celery and lentils.  They get addicted to junk foods, the same foods that exert a strong pull on most of us-- even if we aren't quite addicted.  What is it about junk foods that has this effect on us?

The study

Gearhardt and colleagues recruited 120 volunteers and had them complete a questionnaire that asked them to rank the "addictiveness" of 35 different foods representing a broad spectrum of nutritional values (3).  Some foods were unprocessed and low in calorie density, while others were highly processed "junk foods".

The results

As expected, highly processed foods were rated as most addictive.  The top seven most addictive foods were:

1. Chocolate
2. Ice cream
3. French fries
4. Pizza
5. Cookies
6. Chips
7. Cake

Also as expected, unprocessed foods low in calorie density were rated as least addictive.  Here are the bottom seven foods:

1. Bananas
2. Plain carrots
3. Plain brown rice
4. Water
5. Cucumbers
6. Broccoli
7. Beans

When they analyzed the nutritional qualities that were most closely associated with addictive ratings, they found that fat content and glycemic load were strongly associated with addictive potential.  

Interpretation

Since food was such an important part of the survival and successful reproduction of our ancestors, the motivational systems in our brains are highly attuned to it.  The brain receives information from sensors in the mouth and small intestine that measure the nutritional value of what we eat, and it sets our level of motivation to eat each food according to its nutritional value (mostly its calorie content).  This process involves dopamine release in the nucleus accumbens, just like everything good in our lives, including sex, hitting a home run, and getting a raise (4).  

Yet too much dopamine makes us too motivated, and this is the essence of addiction.  When a stimulus increases brain dopamine too much-- such as crack cocaine flooding the nucleus accumbens with dopamine-- it causes an unnaturally strong motivational state that can lead to destructive decisions.

Gerhardt and colleagues found that the most "addictive" foods are processed foods high in fat and with a high glycemic load.  Consistent with findings from other areas of research, this suggests that we're particularly drawn to foods that deliver concentrated calories that are easily and rapidly digested.  Gearhardt explains that this scenario parallels key properties of drugs of abuse:

The current study provides preliminary evidence that not all foods are equally implicated in addictive-like eating behavior, and highly processed foods, which may share characteristics with drugs of abuse (e.g. high dose, rapid rate of absorption) appear to be particularly associated with “food addiction.”

To someone who is familiar with food reward circuitry, this is no surprise.  The brain systems that mediate learning and motivation respond to specific food properties**, and when we eat foods that deliver unnaturally concentrated combinations of those properties, the brain produces an unnaturally strong motivational state.  In some people, that state can be so exaggerated that it resembles drug addiction.  In the rest of us non-addicted folk, it just makes us eat too much.


* I think if we were really honest with ourselves, we would recognize that most people in affluent nations like the US make self-destructive choices about food, due to the motivational pull of certain calorie-rich foods.  We know many of these foods are unhealthy, yet we (over)eat them anyway, and they end up making us sick and limiting our mobility and enjoyment of life.  This satisfies the core diagnostic criteria of addiction: repeatedly engaging in a rewarding behavior despite serious, known harmful consequences.  Frankly, I suspect the main reason we're resistant to classifying this behavior as addiction is that we don't want to admit we have a major, widespread problem with our eating behavior.

** These include carbohydrate, fat, protein, salt, sweet taste, glutamate (umami) taste, and calorie density.