The nucleus accumbens septi is the major target of mesolimbic dopamine neurons originating in the ventral tegmental area of the mesencephalon. Studies involving experimental manipulation of dopamine transmission by drugs and by lesions, as well as in vivo monitoring of extracellular dopamine concentrations, have provided evidence that the dopamine transmission of the nucleus accumbens plays an important role in behaviour motivated by conventional (e.g., food, sex) and drug reinforcers. Motivated behaviour is distinguished into an appetitive (preparatory/anticipatory) phase consisting of flexible response patterns intended to search and approach the reward itself, and a consummatory phase, consisting of fixed response patterns (eating, drinking, copulating, etc.) finalized to the utilization of the biological resources of the reward (caloric, metabolic, genetic, etc.). While some studies reported a stimulation of dopamine transmission in the nucleus accumbens in relation to appetitive as well as consummatory behaviour, other studies reported a relationship exclusively with consummatory behaviour. Therefore, the precise relationship between dopamine transmission in the nucleus accumbens and specific phases of motivated behaviour is debated. On the basis of topographical, histochemical and connectional evidence, the nucleus accumbens has been subdivided into two compartments, a medioventral "shell" and a laterodorsal "core". This heterogeneity may be relevant to the current debate over the role of nucleus accumbens dopamine in behaviour. Thus, one might hypothesize that, depending on the specific compartment of the nucleus accumbens where dopamine transmission is monitored, a different relationship with specific stimuli which motivate behaviour is obtained. In order to verify this possibility we monitored by microdialysis the changes in dopamine transmission in the nucleus accumbens shell and core during appetitive and consummatory phases of behaviour motivated by food. As food we utilized a palatable snack food (Fonzies) whose consumption has been shown in previous studies from our laboratory to release dopamine in the nucleus accumbens shell and in the medial prefrontal cortex. Unpredicted consumption of Fonzies preferentially stimulated dopamine transmission in the shell as compared to the core. Appetitive food stimuli (perforated Fonzies-filled boxes) phasically stimulated dopamine transmission in the core but not in the shell and sensitized the dopamine response to feeding in the core but inhibited that in the shell. These clear-cut differences between nucleus accumbens shell and core suggest that phasic dopamine transmission in each compartment of the nucleus accumbens subserves different roles in motivated behaviour.