Abstract

Place memory is relevant for exploration and forage behaviour. When food supply is dispersed, a win-shift has advantage over a win-stay strategy. In the Olton Octagonal Maze, the rat follows a win-shift strategy using working memory. However, in the Olton 4x4 version, the rat follows a win-stay strategy, using both working and long-term memories. It has been suggested that the neocortex is required for the resolution of tasks demanding long-term, but not for that demanding working memory alone. The role of anteromedial/posterior parietal cortex (AM/PPC) was investigated here, using a reversible lesion induced by intracerebral lidocaine infusion. Long-Evans rats were implanted with guide cannulae into the AM/PPC and trained in an Olton 4x4 maze, counting working and long-term memory errors after a delay. Then, the animals were infused with lidocaine or saline during the delay phase and tested for three days. Another series of animals, treated as before, was tested in an Olton Octagonal Maze and subjected to the same injection schedule. In the Olton 4x4 Maze, lidocaine produced a significant increase in working and long-term memory errors, compared to saline and post-lidocaine conditions. In contrast, in the Olton Octagonal Maze, lidocaine did not induce any effect on working memory errors. Thus, AM/PPC is required when both working with previous information and long-term memories are needed, but not when only working memory is required, as it happens under ethological conditions. Whenever food supply is dispersed, a win-shift strategy is preferable