The neural basis of trajectory computations in rodent posterior parietal cortex and hippocampus

Abstract

Space is a fundamental property of nature, so it is not surprising that the processing of spatial information involves many brain structures, including the posterior parietal cortex (PPC) and hippocampus. The PPC represents body position in multiple frames of reference, aligning the reference frames of different parts of the body to cooperate in the performance of a task. Hippocampal "place cells" increase their firing rates in specific locations, and place cell responses can be based on information encoded in multiple reference frames, including external, sensory-based reference frames or internal, memory-based frames. The computation of whole-body trajectories is common to all navigation tasks, but little is known about how this computation is performed. Rats were trained to run to distant targets that were presented either randomly or as segments of sequences. When rats were trained to make direct trajectories to distant targets that were presented in random order, no evidence was found that hippocampal place cells encoded distant goals. When rats learned a sequence of goals that contained a repeated segment, hippocampal place cell activity along the repeated segment remained the same. This showed that differential hippocampal codes were not required for rats to differentiate overlapping sequential contexts. Differential hippocampal codes could be formed, however, if rats learned sequences of goals under specific conditions, reflecting differential activity from neural structures outside the hippocampus. At the initiation of trajectories, the phase of hippocampal theta was reset at the peak acceleration of the rat. The activity of some parietal neurons was modulated by hippocampal theta, though the magnitude of modulation was not as great as that for hippocampal units, and the preferred firing phase of parietal units differed from that of theta-modulated place cells. Two classes of units in PPC responded differentially during the early and late stages of trajectories: one class with an increased firing rate during the early stages of trajectories, and the other with an increased firing rate during the late stages. These results provide a foundation for future studies of parieto-hippocampal interactions during trajectory planning.