The Society for the Study of Ingestive Behaviour meets each year to discuss and highlight recent developments in our understanding of all aspects of controls of eating and drinking. The aim of this satellite was to focus on key issues in human food intake control: palatability, central and peripheral signalling, dietary restraint, macronutrients, energy density, heritability and the significance of animal models for research in man, and to bring together researchers approaching these issues from different perspectives. The presentations including live discussions are reflected in the following papers.

The traditional view of palatability was that it reflected some underlying nutritional deficit and was part of a homeostatically driven motivational system. However, this idea does not fit with the common observation that palatability can lead to short-term overconsumption. Here, we attempt to re-evaluate the basis of palatability, first by reviewing the role of salt-need both in the expression of liking for salty tastes, and paradoxically, in dissociating need from palatability, and second by examining the role of palatability in short-term control of appetite. Despite the clarity of this system in animals, however, most salt (NaCl) intake in man occurs in a need-free state. Similar conclusions can be drawn in relation to the palatability of food in general. Importantly, the neural systems underlying the hedonic system relating to palatability and homeostatic controls of eating are separate, involving distinct brain structures and neurochemicals. If palatability was a component of homeostatic control, reducing need-state should reduce palatability. However, this is not so, and if anything palatability exerts a stronger stimulatory effect on eating when sated, and over-consumption induced by palatability may contribute to obesity. Differential responsivity to palatability may be a component of the obese phenotype, perhaps through sensitisation of the neural structures related to hedonic aspects of eating. Together, these disparate data clearly indicate that palatability is not a simple reflection of need state, but acts to promote intake through a distinct hedonic system, which has inputs from a variety of other systems, including those regulating need. This conclusion leads to the possibility of novel therapies for obesity based on modulation of hedonic rather than homeostatic controls. Potential developments are discussed.

Dieting Makes You Fat, the title of a 1980s book on weight control, is a popularised paradox, conveying a conclusion that is consistent with personal experience and the reported failure of most dietary approaches in the treatment of obesity. Few studies have been designed specifically to test this association. Yet there are prospective data showing that baseline dieting or dietary restraint increases the risk of weight gain, especially in women. Metabolic adaptations and the disinhibited eating of restrained eaters have been the most commonly cited explanations for such weight gain. Dietary restraint has also been implicated in the development and persistence of binge eating. The present paper critically evaluates the evidence supporting this paradox and reaches a rather different conclusion.

Dieting has developed a negative reputation among many researchers and health care professionals. However, ‘dieting’ can refer to a variety of behavioural patterns that are associated with different effects on eating and body weight. The wisdom of dieting depends on what kind of dieting is involved, who is doing it, and why. Thus, depending on what one means by the term, dieting can be quite harmful, merely ineffective or actually beneficial. The present paper considers examples of all three. In particular, we argue that judgements about the desirability of dieting should consider the likely consequences to particular individuals of engaging in, or not engaging in, dieting behaviour.

The present review summarises the effects of different carbohydrate and fat structures on food intake and appetite and the differences in response at various levels of processing of macronutrients. Several physico-chemical properties of carbohydrate and fat molecules appear to influence the short-term satiating properties. However, long-term substantiation of these findings expressed in terms of food intake or body weight is not currently available. Such studies will be required to make clear recommendations regarding dietary composition to aid satiety.

The influence of protein and amino acid on the control of food intake and the specific control of protein and amino acid intakes remains incompletely understood. The most commonly accepted conclusions are: (1) the existence of an aversive response to diets deficient in or devoid of protein or deficient in at least one essential amino acid; (2) the existence of a mechanism that enables attainment of the minimum requirement for N and essential amino acids by increasing intake of a low-protein diet; (3) a decrease in the intake of a high-protein diet is associated with different processes, including the high satiating effect of protein. Ingested proteins are believed to generate pre- and post-absorptive signals that contribute to the control of gastric kinetics, pancreatic secretion and food intake. At the brain level, two major afferent pathways are involved in protein and amino acid monitoring: the indirect neuro-mediated (mainly vagus-mediated) pathway and the direct blood pathway. The neuro-mediated pathway transfers pre-absorptive and visceral information. This information is for the main part transferred through the vagus nerve that innervates part of the oro-sensory zone: the stomach, the duodenum and the liver. Other information is directly monitored in the blood. It is likely that the system responds precisely when protein and essential amino acid intake is inadequate, but in contrast allows a large range of adaptive capacities through amino acid degradation and substrate interconversion.

The effects of alcohol on food and energy intake in human subjects have been the subject of a number of controlled studies recently. Unlike the evidence for other macronutrients, there is minimal evidence for any compensatory reduction in food intake in response to energy ingested as alcohol. In contrast, all studies testing intake within 1 h of preload ingestion report a higher intake of food following alcohol relative to energy-matched controls, although this short-term stimulatory effect is not evident if the test meal is delayed beyond 1 h. This time-course suggests that short-term stimulation of appetite may be mediated by the pharmacological action of alcohol on the appetite control system, either through enhanced orosensory reward or impaired satiety. In the long term, energy ingested as alcohol is additive to energy from other sources, suggesting that moderate alcohol consumption results in long-term passive over-consumption alongside short-term active over-consumption of energy through appetite stimulation. Despite the consistency of enhanced energy intake after moderate alcohol, evidence of an association between alcohol in the diet and obesity remains contentious, although the most recent results suggest that alcohol intake correlates with BMI. Future research needs to address this issue and clarify the mechanisms underlying appetite stimulation by alcohol.

The effect of energy density (ED) on energy intake (EI) has been assessed in short-term and long-term experiments. In the short term, it was found that ED affects EI directly in situations when the subjects cannot estimate the ED of the food; then subjects mainly monitor the weight of the food ingested. In the long term, the effects of ED on EI are modulated. Average daily EI appears to be related to ED of the food and drinks when ED is determined by specific macronutrients, but not when ED is only determined by the weight of water. Thus, the short-term effect ED has on EI cannot be extrapolated to the long term, because a possible dominating effect of the weight of water determining ED undoes the relationship of ED with EI. Moreover, in the long-term portion sizes are used to compensate for correctly estimated ED, resulting in less variation in EI than ED alone would imply. Finally, dietary restraint compensates for the effect of a relatively high ED on daily EI, whereas dietary unrestraint compensates for the effect of relatively low ED on daily EI. We conclude that the short-term effect of ED on EI is modulated by the effect of water on ED, and compensated for by the effect of dietary restraint and adapted portion sizes.

The prevalence of obesity among adults and children has increased steadily over the last few years worldwide, reaching epidemic proportions. Particularly alarming is the link between obesity and the development of chronic disorders such as heart disease, type 2 diabetes, hypertension and some cancers (Bjorntorp, 1997). Environmental causes of obesity are thought to include a sedentary lifestyle and an abundance of highly palatable energy-dense foods (Hill et al. 2003). Genetic factors also contribute to susceptibility to obesity, although the genetic basis of most human obesities is thought to be polygenic (Comuzzie & Allison, 1998; Barsh et al. 2000). The present paper considers some of the animal models used to infer aspects of human obesity, with an emphasis upon their usefulness.

The worldwide increase in the incidence of obesity is a consequence of a positive energy balance, with energy intake exceeding expenditure. The signalling systems that underlie appetite control are complex, and the present review highlights our current understanding of key components of these systems. The pattern of eating in obesity ranges from over-eating associated with binge-eating disorder to the absence of binge-eating. The present review also examines evidence of defects in signalling that differentiate these sub-types. The signalling network underlying hunger, satiety and metabolic status includes the hormonal signals leptin and insulin from energy stores, and cholecystokinin, glucagon-like peptide-1, ghrelin and peptide YY3-36 from the gastrointestinal tract, as well as neuronal influences via the vagus nerve from the digestive tract. This information is routed to specific nuclei of the hypothalamus and brain stem, such as the arcuate nucleus and the solitary tract nucleus respectively, which in turn activate distinct neuronal networks. Of the numerous neuropeptides in the brain, neuropeptide Y, agouti gene-related peptide and orexin stimulate appetite, while melanocortins and α-melanocortin-stimulating hormone are involved in satiety. Of the many gastrointestinal peptides, ghrelin is the only appetite-stimulating hormone, whereas cholecystokinin, glucagon-like peptide-1 and peptide YY3-36 promote satiety. Adipose tissue provides signals about energy storage levels to the brain through leptin, adiponectin and resistin. Binge-eating has been related to a dysfunction in the ghrelin signalling system. Moreover, changes in gastric capacity are observed, and as gastric capacity is increased, so satiety signals arising from gastric and post-gastric cues are reduced. Understanding the host of neuropeptides and peptide hormones through which hunger and satiety operate should lead to novel therapeutic approaches for obesity; potential therapeutic strategies are highlighted.

Genes are known to have independent influences on the height and weight of individuals, their overall levels of nutrient intakes and their meal sizes and frequencies. Recent evidence suggests that genes exert multiple and subtle influences on the controls of food intake. There are significant genetic influences on the level and responsiveness of the individual to physiological factors, such as the preprandial stomach contents of nutrients and subjective hunger, and also to environmental and psychological factors such as social facilitation of eating, diurnal rhythms of intake, palatability, cognitive restraint and dietary density. The general model of intake regulation provides an integrated and comprehensive account of how these physiological and environmental factors might fit together to produce the control of intake and body weight.

Studies of monozygotic twins in the context of overfeeding and energy deficit experiments have shown that gene–environment interactions affect energy balance. From a clinical standpoint, this implies that some individuals are more susceptible to body-weight gain or loss than others because of genetic differences. This opens new perspectives in predictive medicine. In the future, health professionals should be able to count on early diagnosis of individuals at risk for developing long-term metabolic problems and obesity or for not responding adequately to clinical interventions. However, before predictive medicine is in a position to contribute significantly to prevention or treatment of patients, an enormous amount of work has to be done to identify all genetic and environmental factors of relevance, and their network of interactions.