SUMMARY

Recent data from research on spatial learning mechanisms with the water maze are discussed, with emphasis on the role of NMDA receptor-dependent long-term potentiation (LTP). Detailed analysis of behavior in the water maze has indicated a need to distinguish among different components of this seemingly simple task. Nonspatial pretraining allows for the separation of behavioral strategy learning and learning the spatial location of the hidden platform. Nonspatially pretrained rats can learn the location of a hidden platform as quickly as controls despite being treated with any of a variety of pharmacologic agents. When a water maze task of conventional difficulty is used, nonspatially pretrained rats given an NMDA receptor antagonist to block hippocampal LTP can learn the location of the hidden platform as quickly as controls. When an especially difficult water maze task is used, involving repeated one-trial learning with repeated reversal learning, blockade of hippocampal LTP produces a spatial memory impairment for long but not short retention intervals in nonspatially pretrained rats. These and other experiments point to an important issue of task difficulty in the research. It should prove useful to systematically vary experimental treatments and task difficulty as a means of identifying specific brain circuits that are important for specific components of the water maze or other tasks. This information can then be used to further evaluate the role of LTP in the learning.

Introduction

Long-term potentiation (LTP) has properties of central nervous system change that are inherently interesting and that could be relevant to a variety of behavioral phenomena. Chief among these is its relevance to systematic and enduring behavioral change, or learning and memory.