The autoshaping deficit is a robust and replicable phenomenon, yet the underlying cause of this deficit remains unexamined. Experiment 1A and 1B investigated whether a failure in processing reward information underlies the autoshaping deficit. In Experiment 1A we used a delayed differential and nondifferential outcomes task. Hippocampal and control animals performed similarly. In Experiment 1B we used the suboptimal procedure as this task may represent a finer measure of reward processing. Consistent with previous research, our animals chose suboptimally, and there was again no difference in performance between hippocampal and control animals. The results of Experiments 1A and 1B suggested that a failure in processing reward information does not appear to underlie the autoshaping deficit.
In Experiment 2, we examined whether a failure in contingency detection is the cause of the autoshaping deficit by using differential and nondifferential autoshaping conditions to manipulate the predictive relationship between the conditioned stimulus (light) and the unconditioned stimulus (food). Hippocampal animals were not impaired at detecting a change in contingency. Indeed, if anything, hippocampal animals seemed more sensitive to the change in contingency than control animals. Hence, we concluded that a failure in detecting contingencies also does not seem to underlie the autoshaping deficit.
In Experiment 3 we used a spatial task known to be sensitive to hippocampal lesions. Unlike control animals that were not impaired, hippocampal animals were severely impaired on an open field analogue of the radial arm maze task.
Our finding that hippocampal animals are not impaired in processing reward or contingency, leads us to conclude that the underlying cause of the autoshaping deficit may be due to a failure in directing movements through space.
