Last month, a few weeks before the sugar celebration that is Halloween, a press release summarizing research from Connecticut College received widespread media coverage.
The study from Joseph Schroeder’s lab used rodents and compared behavioral reactions to addictive drugs and to food, concluding that the responses were the same. They also looked at brain responses, using a post-mortem measurement of neuronal activity, to drug delivery and to food consumption.
Schroeder’s lab examined whether Oreo cookies or drugs activated more neurons in the nucleus accumbens, an area that is routinely nicknamed “the brain’s pleasure center.” Drugs of addiction are known to alter how the brain functions, and many of them affect neuronal activity in the nucleus accumbens. Schroeder and his students reported that more accumbens neurons were activated when the rodents ate Oreo cookies than when they took drugs.
Putative scientific findings aside, the fact that this study received so much media attention both fascinates and frustrates me.
I find it frustrating that a press release was even written for this research before it was published in a peer-reviewed journal. It fascinates me how widespread the media coverage was. All the media attention might make it seem to the general public like this study was published.
But this is not true. The peer-review process required of scholarly journals has not yet vetted this study.
Science relies on the peer-review process to ensure that experiments were conducted in a way that the results are interpretable. The results Schroeder summarizes in the press release and on his conference abstract, which does not qualify as a peer-reviewed publication, should not really be discussed in the press at all. And if the results are discussed, they can only be discussed with an abundance of caution. The headlines I read about this research were not cautious at all.
The press release uses the adjective “significantly” to describe the difference in neuronal response to Oreos over drugs. Does this mean statistical significance? It is very important to know if the differences in neuronal activity Schroeder reports are statistically significant and if they are reproducible.
What about comparing brain responses to food and drugs in the same animals? There is wide variability, especially among humans, in who becomes addicted to drugs and who becomes obese from pathological overeating. There is strong genetic evidence for susceptibility to addiction in people, and it is possible that just a few members of the group who are more susceptible to addiction could drive the group result.
An additional implication of the media attention this work received, it might seem like this study was one of the first to discover that food can be addicting.
This is also not true. The idea of food being addicting is certainly not new.
Nora Volkow, a worldwide leading researcher on drug addiction and the head of the National Institute on Drug Abuse, has published numerous studies examining human brain responses to drugs and food and has penned several review articles summarizing other’s work. For example, Volkow’s group published a detailed examination (in a peer-reviewed journal) that compared human brain responses to drugs of addiction and food. The study was published in 2004, almost 10 years ago.
The idea of food being addicting certainly is controversial, to scientists and the public.
Overeating is frequently discussed in terms of personality traits or behavioral characteristics: someone is lazy or gluttonous or lacks will power. In truth, overeating – like most human behavior – is an extremely complex phenomenon. Sometimes it might indeed arise from behavior or sometimes maybe even from hormonal imbalances. External influences also contribute to overeating.
David Kessler, a pediatrician by training who oversaw the Food and Drug Administration in the 1990s, wrote the best selling The End of Overeating in 2010. In his book, Kessler summarizes published research on addiction and obesity. He also fleshes out in detail an external influence that leads to overeating and obesity: processed food that is carefully engineered to hijack our brains into making us overeat.
Kessler writes about how the salt, fat, and sugar content of processed foods are carefully balanced to make them highly palatable – and addicting. Oreos are a good example of such an engineered food.
In the press release, Schroeder discusses the implications his work has for food addiction.
"Our research supports the theory that high-fat/ high-sugar foods stimulate the brain in the same way that drugs do," Schroeder said. "It may explain why some people can't resist these foods despite the fact that they know they are bad for them."
So why was Schroeder’s research written up for the press before it was published? Why did that press release get so much attention? I think the answer is actually just one word: Oreos.
Oreos are famous. They are found in many American kitchens and in packed school lunches. Even if we do not eat them regularly now, at one point in time, most of us enjoyed dunking one into a glass of milk. Oreos are familiar. That familiarity is a simple reason why the press release could have been written prematurely, before the work was published, and why so many news outlets covered the press release.
The finding of food activating the nucleus accumbens more than cocaine or morphine, if it holds up after peer-review, is intriguing and would add to the mounting evidence of the addictiveness of highly processed foods.
However, I would much rather read about Schroder’s work after it has gone through the peer-review process that is so critical to science.