Lesions of both dorsal and ventral hippocampus were produced by multiple infusions of the excitotoxin AMPA. Meal patterns recorded before and after lesioning showed no change in total food intake, but a striking behavioral syndrome in which the lesioned rats took smaller meals 2-3 times as frequently and showed a similar change in drinking. In addition, lesioned rats alternated more frequently between feeding and drinking during a single bout of ingestive behavior. There were no group differences in the satiety sequence that followed a meal. In an open field test, lesioned rats showed enhanced locomotion in the periphery and reduced rearing. An olfactory habituation-dishabituation task showed that the lesioned rats investigated olfactory stimuli less but dishabituation to a changed stimulus was normal. The data are discussed in terms of changes in behavioral switching or a possible interoceptive agnosia following hippocampal damage.

HighlightsNon-rewarded or irrelevant prior associations are important for flexible responding.Associations of reward and non-reward in reversal learning are neurally dissociable.Disruption of prior irrelevant or rewarded associations cause pathological deficits.Experimental paradigms of cognitive flexibility can be improved to aid translation.

5-Hydroxytryptamine (5-HT)(2C) and 5-HT(1B) receptors are implicated in the inhibitory modulation of feeding behaviour. However, their respective, and possibly different, roles have not been clearly identified because of a lack of selective 5-HT(2C) receptor agonists. Here, using the putative, selective 5-HT(2C) receptor agonist VER23779, we show that its effects on feeding are fully reversed by pretreatment with a selective 5-HT(2C) receptor antagonist, but unaffected by pretreatment with either a 5-HT(1B) or a 5-HT(2A) receptor antagonist. In mice eating a palatable mash, feeding ends earlier, inactivity is increased but the behavioural satiety sequence is preserved. In a second-order schedule of reinforcement with an initial, non-food-reinforced appetitive phase, VER23779 produces a much greater relative reduction in appetitive responding than the 5-HT(1B) receptor agonist CP-94,253. Increased c-fos immunoreactivity patterns following VER23779 also differ from those described for CP-94,253, in particular showing strong activation of the basolateral amygdala. The different behavioural consequences of 5-HT(2C) and 5-HT(1B) receptor activation may relate to the patterns of c-fos immunoreactivity. In particular, the basolateral amygdala may have a role in maintaining response in the appetitive phase of the second-order schedule and also be susceptible to serotonergic modulation through activation of 5-HT(2C) receptors.

Reducing activity at the 5-HT(2C)R facilitates reversal learning in the mouse by reducing the influence of previously non-rewarded associations.

The hypophagic effect of d-fenfluramine in mice is unlikely to be mediated by the 5-HT1B receptor. Instead, the evidence suggests that an adaptive change in 5-HT2C receptor function occurs in 5-HT1B receptor KO mice and contributes to their reduced response to d-fenfluramine.