Persistence of spatial memory in the Morris water tank task.

Persistence of the spatial memory record was examined in 30 rats overtrained in the working memory version of the Morris water tank task. In Experiment 1, the animal had to find during the acquisition trial an invisible underwater platform randomly located at one of 4 possible sites in the pool. Retention test was performed 1 min, 60 min, 4 h or 24 h later with the same position of the platform and same or changed position of the start. Whereas in the acquisition trials the rats reached the goal after 12 s on average, the latency in the retrieval trials increased with the acquisition-retrieval interval. It was 5 s with the 1-min delay and exponentially approached, but did not quite reach the acquisition trial latency with the 24-h delay. In Experiment 2 the rats were started from the same position of the tank to goal positions changing from trial to trial in a prearranged sequence. There were either 1-h or alternating 1-min and 2-h intervals between the 6 daily trials. The latencies were long during the first trial and whenever the goal changed and short when the goal remained the same as on the previous trial. The latencies were not significantly influenced by intertrial interval in the 1 min to 2 h range. It is concluded that the decay of spatial memory in the water tank task is slow and is little affected by proactive interference.