NEONATAL WHISKER REMOVAL REDUCES THE DISCRIMINATION OF TACTILE STIMULI BY THALAMIC ENSEMBLES IN ADULT RATS

Simultaneous recordings of up to 48 single neurons per animal were use d to characterize the long-term functional effects of sensory plastic modifications in the ventral posterior medial nucleus (VPM) of the tha lamus following unilateral removal of facial whiskers in newborn rats. One year after this neonatal whisker deprivation, neurons in the cont ralateral VPM responded to cutaneous stimulation of the face at much l onger minimal latencies (15.2 +/- 8.2 ms, mean +/- SD) than did normal cells (8.8 +/- 5.3 ms) in the same subregion of the VPM. In 69% of th ese neurons, the initial sensory responses to stimulus offset were fol lowed for up to 700 ms by reverberant trains of bursting discharge, al ternating in 100-ms cycles with inhibition. Receptive fields in the de afferented VPM were also atypical in that they extended over the entir e face, shoulder, forepaw, hindpaw, and even ipsilateral whiskers. Dis criminant analysis (DA) was then used to statistically evaluate how th is abnormal receptive field organization might affect the ability of t halamocortical neuronal populations to ''discriminate'' somatosensory stimulus location. To standardize this analysis, three stimulus target s (''groups'') were chosen in all animals such that they triangulated the central region of the ''receptive field'' of the recorded multineu ronal ensemble. In the normal animals these stimulus targets were whis kers or perioral hairs; in the deprived animals the targets typically included hairy skin of the body as well as face. The measured variable s consisted of each neuron's spiking response to each stimulus differe ntiated into three poststimulus response epochs (0-15, 15-30, and 30-4 5 ms). DA quantified the statistical contribution of each of these var iables to its overall discrimination between the three stimulus sites. In the normal animals, the stimulus locations were correctly classifi ed in 88.2 +/- 3.7% of trials on the basis of the spatiotemporal patte rns of ensemble activity derived from up to 18 single neurons. In the deprived animals, the stimulus locations were much less consistently d iscriminated (reduced to 73.5 +/- 12.6%; difference from controls sign ificant at P < 0.01) despite the fact that much more widely spaced sti mulus targets were used and even when up to 20 neurons were included i n the ensemble. Overall, these results suggest that neonatal damage to peripheral sense organs may produce marked changes in the physiology of individual neurons in the somatosensory thalamus. Moreover, the pre sent demonstration that these changes can profoundly alter sensory dis crimination at the level of neural populations in the thalamus provide s important evidence that the well-known perceptual effects of chronic peripheral deprivation may be partially attributable to plastic reorg anization at subcortical levels.