Spatial mapping takes time

Tests for learning and memory in rodent regulatory studies

For decades, regulatory guidelines for safety assessment in rodents for drugs, chemicals, pesticides, and food additives with developmental neurotoxic potential have recommended a single test of learning and memory (L&M). In recent years some agencies have requested two such tests. Given the importance of higher cognitive function to health, and the fact that different types of L&M are mediated by different brain regions assessing higher functions represents a step forward in providing better evidence-based protection against adverse brain effects. Given the myriad of tests available for assessing L&M in rodents this leads to the question of which tests best fit regulatory guidelines. To address this question, we begin by describing the central role of two types of L&M essential to all mammalian species and the regions/networks that mediate them. We suggest that the tests recommended possess characteristics that make them well suited to the needs in regulatory safety studies. By brain region, these are (1) the hippocampus and entorhinal cortex for spatial navigation, which assesses explicit L&M for reference and episodic memory and (2) the striatum and related structures for egocentric navigation, which assesses implicit or procedural memory and path integration. Of the tests available, we suggest that in this context, the evidence supports the use of water mazes, specifically, the Morris water maze (MWM) for spatial L&M and the Cincinnati water maze (CWM) for egocentric/procedural L&M. We review the evidentiary basis for these tests, describe their use, and explain procedures that optimize their sensitivity.

Infusion of GAT1-saporin into the medial septum/vertical limb of the diagonal band disrupts self-movement cue processing and spares mnemonic function

Degeneration of the septohippocampal system is associated with the progression of Dementia of the Alzheimer's type (DAT). Impairments in mnemonic function and spatial orientation become more severe as DAT progresses. Although evidence supports a role for cholinergic function in these impairments, relatively few studies have examined the contribution of the septohippocampal GABAergic component to mnemonic function or spatial orientation. The current study uses the rat food-hoarding paradigm and water maze tasks to characterize the mnemonic and spatial impairments associated with infusing GAT1-Saporin into the medial septum/vertical limb of the diagonal band (MS/VDB). Although infusion of GAT1-Saporin significantly reduced parvalbumin-positive cells in the MS/VDB, no reductions in markers of cholinergic function were observed in the hippocampus. In general, performance was spared during spatial tasks that provided access to environmental cues. In contrast, GAT1-Saporin rats did not accurately carry the food pellet to the refuge during the dark probe. These observations are consistent with infusion of GAT1-Saporin into the MS/VDB resulting in spared mnemonic function and use of environmental cues; however, self-movement cue processing was compromised. This interpretation is consistent with a growing literature demonstrating a role for the septohippocampal system in self-movement cue processing.

Multifactorial interactions involved in linear self-transport distance estimate: a place for time

As the vestibular system is the only sensory organ whose primary function is self-motion detection, we examined the conditions under which the otoliths, which detect the linear acceleration of the head, could be used to estimate traveled distance. In order to isolate the contribution of the otoliths (with the somatosensory system) from contributions of the visual and motor systems subjects were transported in darkness. We initially hypothesized that self-transport with continuously varying linear velocity should facilitate distance computation by continuously stimulating the otoliths, and that active control of self-motion should also help subjects estimate the distance traveled. However, it was found that the distance covered during self-motion is actually better estimated when transport velocity is quasi-constant. Nevertheless, such estimates strongly depend upon velocity magnitude; subjects show an idiosyncratic preferred self-motion velocity for which distance measurements are most accurate. Furthermore, the active control of self-transport improves estimates of self-motion mainly because the subjects can then adopt a constant velocity, and more precisely their preferred one. It was finally found that subjects mentally count in order to assess their displacement length, and that time perception is indeed disturbed by varying self-motion velocity.

Hippocampal lesioned rats are able to learn a spatial position using non-spatial strategies

In the last two decades, many experiments have demonstrated that the hippocampus plays a role in the learning and processing of spatial and contextual information. Despite these demonstrations, some recent publications have indicated that the hippocampus is not the only structure involved in spatial learning and that even after hippocampal lesions, rats can perform spatial tasks. However, it is not well established whether animals with hippocampal dysfunction still have some spatial learning capacities or develop non-spatial solutions; these may require lengthier acquisition training. We now report the effects of conventional, dorsal hippocampal ablation on rats' performance on the water maze. We tested rats using a short (4 days) versus a long (16 days) acquisition period. We demonstrated that animals with dorsal hippocampal lesions have some residual capacity for learning the localization of a hidden escape platform in a pool during both a reference memory task and a working memory task. The animals with dorsal hippocampal lesions learned to escape at a fixed location, but only with extended training. It is suggested that these animals used non-spatial strategies to compensate for a spatial memory impairment. The results are discussed with respect to the experimental procedure and the strategy applied by the lesioned rats.

Exploratory patterns of rats on a complex maze provide evidence for topological coding

Rats' exploratory patterns on a complex elevated maze were analyzed in both light and dark conditions. Rats were less active in the light than in the dark. In the light, they spent more time exploring the outer areas of the maze than the inner areas whereas exploration of both regions was similar in the dark. In both light and dark, rats spent more time investigating choice points (which provided multiple directions for movements) than runways that allowed only simple movements. In addition, choice points that provided more connections with other distant places were associated with more exploration. While such effects might be the result of stimulus-seeking of distant information in the light, increased exploration times in the dark presumably reflect the processing of local information associated with the maze connectivity. These results suggest that exploratory patterns in the dark reflect processing of the topological structure of the maze.

Rats alternate on a dry-land but not swimming-pool (Morris task) place task: implications for spatial processing

Groups of rats were rewarded with food for traveling from a start point to 2 different locations while their alternations in choice between those locations on 2 daily trials were recorded. In one experimental condition, the rats swam and received food once they climbed upon a platform that was hidden just below the surface of the water at the food location. In the other condition, the rats walked to reach the food. It was found that the rats did not alternate their choices between target locations when swimming but that they did alternate target choices when walking. Even experience in alternating when walking did not produce reliable alternation when swimming. It is proposed that rats treat escape (swimming) and search (walking) tasks in different ways, and this difference is discussed with respect to the possibility that different central processes may be used in the task solutions.