9514880 Mizumori To study the process by which sensory-based, organism- centered spatial information is transformed into an allocentric coordinate system, we will examine the relative sensory and mnemonic contributions within individual tectolimbic structures to accurate navigation. This first step is essential in order to define target structures within the neural system of interest. The original specific aims also proposed to test the functional connectivity between the target brain structures so that we could begin to evaluate the dynamical interactions between structures. However, the latter specific aim has been eliminated. To facilitate comparisons between structures, these experiments will focus on one well studied form of navigational learning, visual spatial learning by rats on a radial maze. Comparison of the response properties of information representation in a variety of brain structures during a single behavior will allow future studies to begin to postulate the sort of dynamical interactions that must exist between brain areas if adaptive consequences of spatial processing are to be realized. In order to understand how organisms utilize the tectolimbic circuit in a behaviorally meaningful way, it is essential that we determine how this system affects the ongoing behavior of the animal. We hypothesize that striatal structures such as the nucleus accumbens and caudate nucleus serve to integrate hippocampal spatial information with nonspatial reward-related information to direct the initiation of voluntary behavior via efferents to the ventral pallidum, a critical structure for the direct control of voluntary behaviors. Together, the results of this neural system analysis will not only provide new and important insight into how multiple brain areas might act in concert during navigation, but our findings should also afford new opportunities to learn about neuroplastic mechanisms that might be available during normal spatia l learning. Single unit responses of the same 12 brain structures listed in the original proposal will be recorded as rats perform a nonspatial and spatial learning task on the radial maze. The structures selected comprise the tectolimbic and limbic striatal systems. Preliminary data show that it is possible to classify these various cell types as spatial or nonspatial, and that we can determine the relative mnemonic and sensory contribution of these cells to navigation. A memory function will be suggested if 1) there is protracted shaping of the correlate during acquisition of a spatial task, 2) if there is evidence for contextual dependency of the representation, and/or 3) if there is short term persistence of the representation in the absence of environmental (visual) cues. Exps. 1-3 will test these possibilities. Exp. 4 and 5 will evaluate the sensory-dependence of the representation by observing cell responses immediately after the removal of visual cues, and after scrambling of the cues. Vestibular sensory contributions will be assessed by rotating the center portion of the radial maze, then observing unit and behavioral responses (Exp. 6). Intramaze olfactory sensory influences on the spatial representations will be tested by observing unit responses to rotation of the entire maze (Exp. 7). Exp. 8 will investigate the role of voluntary movement to spatial representations. Finally, since accumbens cells are sensitive to differences in reward magnitude, Exps. 1-8 will take place using rewards of different magnitude.
