An economical method to study variability in the acquisition and extinction of operant behavior.

To examine operant response variability in detail, a lever was added to the Thumbstick of a video game controller that functions as a joystick. The control was mounted outside a standard rodent operant chamber with the lever extended inside the chamber. Lever movement was restricted to a downward vertical distance of 2 cm. Pellet delivery was used as a lever movement reinforcer. The criteria for reinforcement were that the lever had to be held between two criterion distances, one from the upper rest position of the lever and one from the final position of the complete movement, and the response had to have a certain duration. The construction of the device, the development of algorithms for response detection and the graphic analysis using actograms are described in detail. Use of the equipment is illustrated with two demonstration experiments examining response variability in extinction and during acquisition. The recording method allows considerable detail regarding variability to be displayed in real time with a resolution of 100 ms. For example, in Experiment 1, extinction of a relatively “simple” operant generated considerable new response forms with very little repetition of the previously reinforced response form.