Wheel-running reinforcement in free-feeding and food-deprived rats

Rats experiencing sessions of 30min free access to wheel running were assigned to ad-lib and food-deprived groups, and given additional sessions of free wheel activity. Subsequently, both ad-lib and deprived rats lever pressed for 60s of wheel running on fixed ratio (FR) 1, variable ratio (VR) 3, VR 5, and VR 10 schedules, and on a response-initiated variable interval (VI) 30s schedule. Finally, the ad-lib rats were switched to food deprivation and the food-deprived rats were switched to free food, as rats continued responding on the response-initiated VI 30-s schedule. Wheel running functioned as reinforcement for both ad-lib and food-deprived rats. Food-deprived rats, however, ran faster and had higher overall lever-pressing rates than free-feeding rats. On the VR schedules, wheel-running rates positively correlated with local and overall lever pressing rates for deprived, but not ad-lib rats. On the response-initiated VI 30s schedule, wheel-running rates and lever-pressing rates changed for ad-lib rats switched to food deprivation, but not for food-deprived rats switched to free-feeding. The overall pattern of results suggested different sources of control for wheel running: intrinsic motivation, contingencies of automatic reinforcement, and food-restricted wheel running. An implication is that generalizations about operant responding for wheel running in food-deprived rats may not extend to wheel running and operant responding of free-feeding animals.