In this article, the effect of chronic injections of amphetamine on feeding and behavioral activation was analyzed. Rats were given milk either through an intraoral cannula or in a standard drinking tube, and the level of their behavioral activation was monitored before, during, and after access to the milk. Cannula- and bottle-fed rats given saline showed similar patterns of intake and activity. Bottle-fed rats given amphetamine (2 mg/kg) showed substantially greater suppression of intake than did cannula-fed rats, but recovered more rapidly, confirming earlier findings (Salisbury & Wolgin, 1985). Such recovery was accompanied by a suppression of stereotyped head scanning movements during access to milk, but not before and after milk access. In contrast, cannula-fed rats given amphetamine showed stereotyped head scans throughout the session for the duration of the experiment. These results suggest that tolerance to the suppression of intake by amphetamine involves learning to suppress stereotyped head movements. The constraints on such learning are briefly discussed.