Control of non-homeostatic feeding in sated mice using associative learning of contextual food cues

Feeding is a complex motivated behavior controlled by a distributed neural network that processes sensory information to generate adaptive behavioral responses. Accordingly, studies using appetitive Pavlovian conditioning confirm that environmental cues that are associated with food availability can induce feeding even in satiated subjects. However, in mice, appetitive conditioning generally requires intensive training and thus can impede molecular studies that often require large numbers of animals. To address this, we developed and validated a simple and rapid context-induced feeding (Ctx-IF) task in which cues associated with food availability can later lead to increased food consumption in sated mice. We show that the associated increase in food consumption is driven by both positive and negative reinforcement and that spaced training is more effective than massed training. Ctx-IF can be completed in ~1 week and provides an opportunity to study the molecular mechanisms and circuitry underlying non-homeostatic eating. We have used this paradigm to map brain regions that are activated during Ctx-IF with cFos immunohistochemistry and found that the insular cortex, and other regions, are activated following exposure to cues denoting the availability of food. Finally, we show that inhibition of the insular cortex using GABA agonists impairs performance of the task. Our findings provide a novel assay in mice for defining the functional neuroanatomy of appetitive conditioning and identify specific brain regions that are activated during the development of learned behaviors that impact food consumption.