An experiment was conducted to examine whether cannula-fed rats could learn to suppress amphetamine-induced head movements when milk infusion was contingent upon suppression of these stereotyped movements. To test this hypothesis, rats in two cannula-fed conditions, Cannula/No Lick and Cannula/Lick, received injections of amphetamine (2.0 mg/kg) for 42 consecutive days. Pilot data had suggested that suppression of these movements was unlikely to occur unless some other stereotyped behavior (e.g., licking) could be made, hence the availability of a drinking tube in one of the conditions. Although neither group recovered to baseline intake levels, milk intakes measured over 6 consecutive weeks revealed that both groups had recovered from the initial hypophagic effect of amphetamine and had learned to suppress stereotyped head movements in order to receive milk infusion. There was no significant difference between the intakes of the two groups. These findings suggest that instrumental learning may be an appropriate model to describe the development of tolerance to amphetamine-induced stereotyped movements. They also imply that the channeling of one form of stereotyped head movement (e.g., head scanning) into another form (e.g., licking) is not necessary for tolerance to occur.