Feeding behavior, obesity, and neuroeconomics

Abstract

For the past 50 years, the most prevalent theoretical models for regulation of food intake have been based in the physiological concept of energy homeostasis. However, several authors have noted that the simplest form of homeostasis, stability, does not accurately reflect the actual state of affairs and most notably the recent upward trend in body mass index observed in the majority of affluent nations. The present review argues that processes of natural selection have more likely made us first and foremost behavioral opportunists that are adapted to uncertain environments, and that physiological homeostasis is subservient to that reality. Examples are presented from a variety of laboratory studies indicating that food intake is a function of the effort and/or time required to procure that food, and that economic decision-making is central to understanding how much and when organisms eat. The discipline of behavioral economics has developed concepts that are useful for this enterprise, and some of these are presented. Lastly, we present demonstrations in which genetic or physiologic investigations using environmental complexity will lead to more realistic ideas about how to understand and treat idiopathic human obesity. The fact is that humans are eating more and gaining weight in favorable food environments in exactly the way predicted from some of these models, and this has implications for the appropriate way to treat obesity.

FIGURE 1

Optimal energy reserves (kJ, top panel), optimal intake rates (Watts, middle panel), and optimal proportion of time spent foraging (bottom panel) as a function of gross energy gain while foraging (Watts). Data redrawn from Figures 7,8 and 9 of Houston & McNamara [19] for a closed economy with the specific model parameters, basal metabolic rate = 2.5W, foraging metabolic rate=7.5W, plus 0.0005 W/kJ body reserves.

FIGURE 2

Results from Logan's [36] closed economy study in which rats were lever pressing on various FR schedules for water reinforcement. A free 12 g food ration was given daily. The top panel shows how the demand for water increased with unit reinforcer size. The middle panel shows the demand for water decreased as a function of increasing unit cost, but became inelastic at higher FRs. The lower panel shows that increasing the relative effort expended to press the lever had a relatively small effect on demand.

FIGURE 3

Median daily response output (left panel) and food demand (45 mg pellets earned per day; right panel) in rats studies under 8 different unit price conditions, where the Unit price = (fixed ratio × lever weight)/(pellets per reinforcement × probability). The left panel shows 3 of the 8 conditions, the standard (1 pellet per reinforcement, probability 1.0, lever force 0.265N), a lower cost (2 pellets per reinforcement, probability 1.0, force 0.265N), and the highest cost studied (1 pellet per reinforcement, probability 1.0, force 0.53N). On the right, food demand is plotted against consummatory cost; all 8 groups fell on or near the single function that is shown. Intake is well conserved at relatively low costs, but drops precipitously as cost becomes high. Data redrawn from Figures 2 and 3 of Hursh et al [39].

FIGURE 4

Effect of various consummatory costs (panels A-C) and procurement costs (panels D-F) on mean number of meals per day (A,D), mean meal size (B,E) and total lever presses emitted (C,F). Derived from data in Collier, Hirsch & Hamlin [35].

FIGURE 5

Mean number of meals per day decreases and mean meal size increases as procurement fixed ratio (PFR) increases in lean and obese (ob/ob) mice. The consummatory cost for each pellet was 5 presses at low PFR and 10 at the higher PFRs. Redrawn from [73].