We developed a novel behavioral task in which rats learn to recognize the configuration of objects in an animated scene displayed on a computer screen. The scene consisted of a moving bar and a stationary rectangle. Rats deprived of food were trained to press a lever for reward in a small chamber located in front of the screen. Lever presses were rewarded
only when the bar was at the rectangle. Rats anticipated the reward by gradually increasing frequency of lever pressing as the bar approached the rectangle. Control experiments showed that neither the timing nor the discrimination of rewarded and non-rewarded periods as two discrete conditions explain behavior of the rat. Because the changes in the
scene were generated by movement of the object, the presented task could be used for studying neural structures involved in spatial behavior of rats using virtual reality technology.