Gearhardt and colleagues [1] propose that certain foods are addictive and that this has important implications for efforts to combat obesity. Their argument is that shifting blame for overeating from the individual to the availability of ‘hyperpalatable’ addictive foods leads to the result that ‘corporate responsibility, public heath approaches, environmental change and global efforts all warrant strong consideration in reducing obesity and diet-related disease’ (p. 1208). Undoubtedly, the rising tide of obesity world-wide is a major challenge of our time, and Gearhardt et al.'s [1] analysis adds to the many calls for action to be commensurate with the size of the problem. Their contribution is to draw parallels with interventions including taxation and curbs on marketing, which have been successful in reducing tobacco use, in particular.

Claiming that foods such as popular products of fast-food restaurants can be addictive is provocative, and might help to provoke policy changes, but can obesity really be understood and treated as food addiction? Of course, this depends partly on how addiction is defined. That seeing and eating liked foods activates brain circuitry also activated by addictive drugs is unsurprising if these drugs are ‘highjacking’ pathways that evolved to regulate dietary behaviour. This issue is, perhaps, one concerning the relative potency of foods and drugs consumed by people (and differences in food-related brain activity associated with or predictive of obesity may be consequences rather than causes of the obese state or eating behaviours that precede it). It would seem reasonable to label binge eating, as occurs in bulimia nervosa and binge eating disorder, as food addiction (e.g. there is compulsion, marked change in affect and long-term harm, and physiologically based satiety tolerance enables escalation of food ‘dose’); however, the route to obesity is generally a rather modest average daily excess of energy intake over energy expenditure.

What, then, are the characteristics of food that increase consumption (of calories)? Gearhardt et al. [1] identify palatability as a key factor. They label certain foods as ‘hyperpalatable’ and contrast these with some ‘traditional’ foods. All else being equal, we eat more of what we like, but it is far from clear that the foods listed by Gearhardt et al. [1] vary as much in palatability as they imply—for this author, an orange (‘traditional’) is as palatable as sweetened breakfast cereal (‘hyperpalatable’). In Italy, at least, pizza is very much a traditional food. It may also be highly liked, although the prevalence of obesity in Italy is among the lowest in Europe. The difficulty is that there is a lack of objective evidence for increases in food palatability paralleling increases in obesity rates [2], and there also seems to be rather little variation between people and foods in the palatability of what is typically eaten [2]. Note that palatability (the hedonic or liking response to oro-sensory stimulation during eating) is modified by individual eating experiences and is not a fixed property of individual foods.

Rather than palatability, a key characteristic of the foods identified by Gearhardt et al. [1] that poses a risk of over-consumption is their relatively high energy density. Two inter-related features of energy-dense foods would appear to be important here. They are more attractive; that is, they have greater incentive salience, because for the same portion (per unit volume or weight) their capacity to satiate (satisfy hunger) is greater than foods lower in energy density. However, on a calorie-for-calorie basis they are less satiating, and therefore they are selected in larger (calorie) portions than energy-dilute foods, irrespective of palatability [3] (palatability was found to be unrelated to energy density [3]). Reducing consumption of energy-dense foods could involve the approaches advocated by Gearhardt et al. [1], as changes in cost, accessibility, serving size and even prevalence of eating while engaged in other activities may all contribute significantly to the risk of obesity. Additionally, it might be possible to formulate foods to increase their satiating capacity (calorie-for-calorie), and to enhance the attractiveness of lower-calorie, less energy-dense foods.

Declaration of interest

Peter Rogers has received funding for research from UK food and pharmaceutical companies, UK research councils and the European Commission.