Cognition, attention and food: the problem with scientific studies

The imperfect interpretation of research findings on disordered eating in real life

Photo by Jessica Spengler on Flickr.

For the past two years, I’ve researched the psychology of disordered eating, specifically the attention paid to food images by those with various forms of eating pathology. Basically, I show people pictures of delicious looking food, and a fancy eye tracking camera records what they look at, and for how long. We compare people with self-reported “disordered eating” symptoms to those without and see if there are any group differences through analysis of the tracking data. The idea is that this may help inform us of underlying cognitive biases that contribute to disordered eating, and it’s a huge and unwarranted leap to say that a predisposition to looking at food images directly leads to problematic eating behaviours, and thus, weight gain.

So where does scientific theory meet reality and vice versa? One of the principles of scientific experimentation is that correlation does not equal causation--just because two variables are tied together in some way doesn’t mean that one directly causes the other to occur, in either direction.

Before I started as a researcher, I participated in one of the studies I now coordinate. I knew as I was looking at the images that I fixated on the food more than anything else, and it was because of my culinary interests, not a desire to binge on the real life versions of the foods I was viewing. An eye tracker can tell you if someone is looking at food or not, but it can’t explain why. From an evolutionary perspective, isn’t it actually beneficial for us to display a bias towards food, given that we need it in order to survive? What happens though when food becomes too salient and this hypervigilance leads to negative outcomes like increase consumption and subsequent weight gain? Short of specifically asking the subject, you don’t really know what is driving their attention, and even asking them involves their personal biases.

Another issue with this type of study is what is known as “ecological validity.” As you might think, looking at pictures of food elicits a very different response than having it there in front of you. Even the most enticing food porn cannot convey the tastes, textures and aromas that make food and eating the involved, multi-sensory experience that it is.

It’s safe to say that increased attention to pictures of food would be related to increased attention to actual food, but you can’t draw conclusions without having the data to support it. Short of putting a portable eye tracker on someone’s head and following them around, you cannot confirm that attention to food pictures and attention to real food are correlated. Even if one predicts the other, you would still need to relate attention to food to actual food consumption in order for the research to have any significant real world implications. Otherwise, there are surely plenty of individuals who display preferential attention to food, but are perfectly content with their weight or dietary habits.

Conflicting study results in the literature surrounding cognition and attention to food don’t help to clarify matters. The first study I ran found that those with symptoms of binge eating displayed increased attention to food images compared to those without. A follow-up study I conducted looking at food addiction initially found the opposite: decreased attention to food images in those with food addiction versus controls, which might be interpreted as purposeful avoidance of such images in this group. 

As I collected data from more participants, the results changed, such that those with food addiction did display increased attention to food images, specifically those we classified as “unhealthy” and were high in salt, sugar, fat and calories. I think this lack of clarity largely has to do with the extensive number of factors that contribute to disordered eating, as well as the overlap, much of which is indiscernible, between many of these behaviours and diagnoses. For example: binge eating and food addiction are highly comorbid, meaning they frequently occur simultaneously. How do you pry apart which of the two influences increased or decreased attention to food? It’s a classic case of, “Which came first, the chicken or the egg?”

Given that food and eating is central to the human condition, it would be fundamental to our understanding of how we live to know why we eat the way we do, especially when it becomes detrimental to our well-being. Am I critical of my research? Absolutely. That’s one of the fundamental principles of science: never take your hypotheses, results, or the way by which you got to them for granted. You must constantly question, re-evaluate and make alterations to your study design and your interpretation of results, based on previous findings of your own and of others. Nothing is ever proven. Instead, there is simply evidence for or against a hypothesis or study question at any point in time. Examining the attention we pay to food is just a small piece of a much larger pie. Alone, it means very little, but combined with other behavioural, social, and psychological aspects, it may help to elucidate our extremely complex relationships with food.

Photo by Jessica Spengler on Flickr.