Entorhinal inputs to dentate gyrus are activated mainly by conditioned events with long time intervals.

Abstract

Brain mechanisms of mammalian learning and memory have long been associated with the hippocampus. Although the role of the hippocampus in spatial behavior is well established, there is no general consensus on the function of the hippocampus in nonspatial tasks. Task-related changes in evoked potentials were first reported by Deadwyler and colleagues in the dentate gyrus, with a peak at 100 ms in rats trained in an auditory discrimination task (Deadwyler et al., Brain Res 161:211-225, 1979). Surprisingly, these results have not been confirmed by other investigators. In the present report, we set out to assess the task parameters that induce, modulate, and suppress this potential. Using multielectrode probes and current source density analysis, we monitored the entorhinal input to the dentate gyrus in behaving rats. Both differential and simple auditory conditioning led to the appearance of a large negative potential at 100 ms in the perforant path zone of the dentate gyrus. This negativity was found in averaged and in single-trial, evoked potentials. Current source density analysis revealed sinks in the perforant path zone of the molecular layer of the dentate gyrus with corresponding sources in the hilus. Once trained for differential conditioning, decrements of target probability consistently increased the amplitude of this negativity. When using a single-tone, stimulus-response task, the negative potential occurred with long (70-s), but not with short (10-s) intertrial intervals. Appearance of the potential coincided with a switch in response strategy and disappeared under stereotypical behavior. These data are best explained under the assumption that long intertrial intervals lead to transitions in task-related reference frames enabling a switch to more appropriate response strategies. Such transitions are not required during short interval stereotyped performance.