Components of Spatial Learning in the Rat

Abstract

A series of experiments were conducted to investigate the nature of how navigational systems interact in the rat (Rattus norvegicus) and the neural structures that support these interactions. The first set of experiments focused on geometry learning and how a reference frame based on the shape of the environment interacted with other non-geometric reference frames. The results revealed that rats were capable of rapidly integrating geometric cues with featural cues in only a single exposure to the cues in compound. This is a novel contribution to the current literature as it opposes the notion that featural information can only be ‘pasted on’ to a geometric reference frame over time. The effect of the rats’ sex on their propensity to use geometric and landmark cues was also investigated. The findings are the first to reveal no difference between male and female rats in the extent to which landmarks overshadow geometry learning when generalization decrement is controlled for. However, in a separate task, male rats were able to use both relevant geometric and landmark information better than female rats following changes to the relative reliability of environmental cues. In a separate series of experiments, the navigational strategies rats rely upon and the neural substrates underpinning these strategies was investigated. In a task requiring rats to use the colours of the enclosure walls to locate a hidden goal, it was found that the performance of rats with hippocampal damage and rats with dorsolateral striatum damage was identical to that of normal rats, i.e. they all solved the task using an allocentric strategy over an egocentric strategy. Importantly, the findings revealed that the hippocampus is not required to learn the spatial relationship between differently coloured features. A separate task revealed that hippocampal damage enhanced landmark learning (egocentric), and dorsolateral striatum damage enhanced room cue learning (allocentric) suggesting that these two systems compete for behavioural control in normal rats. Finally, the last experiment revealed that, under certain training conditions, the hippocampus is not critical for the acquisition of a place solution but is more likely involved in a path integration process. This result holds important implications for the role of the hippocampus in ‘knowing where’ versus ‘getting there’.