1. Responsiveness within the hand region of the second somatosensory area of cortex (SII) was investigated in the marmoset monkey (Callithrix jacchus) in association with cooling-induced, reversible inactivation of the primary somatosensory area, SI. The aims were to determine whether thalamocortical systems in this primate species are organized according to a serial scheme in which tactile information is conveyed from the thalamus to SI and thence to SII as the next hierarchical level of processing and to establish whether primates are fundamentally different, in this respect, from mammals in which tactile information is conveyed in parallel from the thalamus to both SI and SII. 2. Inactivation of the SI had area was achieved when the temperature at the face of the silver cooling block over this SI region was lowered to < or = 13 degrees C. Inactivation was confirmed by abolition of the SI surface potential evoked by a brief tap stimulus to the hand and by the abolition of responsiveness in single SI neurons located beneath and around the edge of the block. 3. The effect of SI inactivation on SII-evoked potentials was investigated in 20 experiments by simultaneous recording of the SI- and SII-evoked potentials. The SII response was never abolished and was unchanged in the majority (12/20) of experiments. In the remainder, the SII-evoked potentials underwent a reduction in amplitude that was usually < 30% but never > 50%. 4. Tactile responsiveness was examined quantitatively in 47 individual SII neurons of different functional classes before, during, and after the inactivation of SI. Controlled tactile stimuli consisted of trains of sinusoidal vibration or rectangular pulses delivered to the glabrous or hairy skin of the hand. 5. Thirteen of the 47 SII neurons (28%) were unaffected in their response levels in association with SI inactivation. The remaining 34 SII neurons underwent some reduction in responsiveness, but in only 6% (3/47) was responsiveness abolished by SI inactivation. As the same range of functional classes of tactile neurons were represented among the affected and unaffected SII neurons, there was no evidence for a differential susceptibility among SII tactile neurons to the effect of SI inactivation. 6. Where reductions in amplitude of the SII-evoked potential or in response levels of SII neurons were observed, the effects were not attributable to direct spread of cooling from SI to the SII hand area as there was no cooling-induced prolongation of either the evoked potential or spike waveform in SII, an effect that is known to precede cooling-induced reductions in responsiveness. 7. These lines of evidence indicate that reductions in SII responsiveness in association with SI inactivation may be attributable to a loss of a background facilitatory influence rather than to a blockage of a component of peripheral input that comes over a putative serial path to SII via SI. First, as SI was cooled, there was a progressive increase in latency and time course of the SI responses before their disappearan...

Responsiveness of the first somatosensory area (SI) of the cerebral cortex was investigated in the marmoset monkey (Callithrix jacchus) in association with cooling-induced, reversible inactivation of the second somatosensory area, SII. The aim was to determine whether SI responsiveness to peripheral tactile stimulation depends on SII and therefore whether SI and SII in the marmoset occupy hierarchically equivalent positions in a parallel organizational scheme for thalamocortical tactile processing as appears to be the case in nonprimate mammals. Inactivation of SII was achieved when the temperature over SII was lowered to < or =12 degrees C, as indicated by abolition of the SII-evoked potentials generated by brief tap stimuli to the hand or foot, and by abolition of tactile responses in single SII neurons located at the margin beneath the block. The effect of SII inactivation on SI-evoked potentials was examined in 16 experiments by simultaneous recording of the SI- and SII-evoked potentials. SI-evoked potentials were never abolished and remained unaffected in 11 cases. In three experiments there was a small reduction in amplitude and inconsistent effects in the remaining two. Responsiveness to controlled tactile stimuli was examined quantitatively in 31 individual SI neurons of different functional classes before, during, and after the inactivation of SII. Tactile responsiveness in individual SI neurons was never abolished by SII inactivation, remaining unchanged in 20 neurons (65%) while undergoing some reduction in the remaining 11 SI neurons (35%). This reduction of tactile responsiveness in one-third of SI neurons is most likely attributable to a removal of a facilitatory influence emanating from SII, based on the observation that background activity of the affected neurons was also reduced. Furthermore, phase locking of SI responses to vibrotactile stimulation was unchanged when SII was inactivated. The retention of responsiveness in SI neurons when SII was inactivated by cooling in the marmoset demonstrates that tactile inputs can reach SI without traversing an indirect, serially organized path through SII. The present results, together with our previous observations that responsiveness in the majority of SII neurons survived SI inactivation, demonstrate that there is a parallel organization of the SI and SII areas for tactile processing in the marmoset monkey and that SI and SII occupy hierarchically equivalent positions in a parallel processing network. There is therefore no longer justification for the view that there are fundamental differences in the organization of thalamocortical tactile processing for SI and SII between simian primates, in general, and other mammals.

The functional organization of the primate somatosensory system at thalamocortical levels has been a matter of controversy, in particular, over the extent to which the primary and secondary somatosensory cortical areas, SI and SII, are organized in parallel or serial neural networks for the processing of tactile information. This issue was investigated for the marmoset monkey by recording from 55 single tactile-sensitive neurons in the lateral division of the ventral posterior nucleus of the thalamus (VPL) with a projection to either SI or SII, identified with the use of the antidromic collision technique. Neurons activated from the hand and distal forearm were classified according to their peripheral source of input and characterized in terms of their functional capacities to determine whether the direct thalamic input can account for tactile processing in both SI and SII. Both the SI- and SII-projecting samples contained a slowly adapting (SA) class of neurons, sensitive to static skin displacement, and purely dynamically sensitive tactile neurons that could be subdivided into two classes. One was most sensitive to high-frequency (> or =100 Hz) cutaneous vibration whose input appeared to be derived from Pacinian sources, while the other was sensitive to lower frequency vibration (< or =100 Hz) or trains of rectangular mechanical pulse stimuli, that appeared to receive its input from rapidly adapting (RA) afferent fibers presumed to be associated with intradermal tactile receptors. There appeared to be no systematic differences in functional capacities between SI- and SII-projecting neurons of each of these three classes, based on receptive field characteristics, on the form of stimulus-response relations, and on measures derived from these relations. These measures included threshold and responsiveness values, bandwidths of vibrational sensitivity, and the capacity for responding to cutaneous vibrotactile stimuli with phase-locked, temporally patterned impulse activity. The analysis indicates that low-threshold, high-acuity tactile information is conveyed directly to both SI and SII from overlapping regions within the thalamic VP nucleus. This direct confirmation of a parallel functional projection to both SI and SII in the marmoset is consistent with our separate studies at the cortical level that demonstrate first, that tactile responsiveness in SII largely survives the SI inactivation and second, that SI responsiveness is largely independent of SII. It therefore reinforces the evidence that SI and SII occupy a hierarchically equivalent network for tactile processing.

The marmoset monkey, Callithrix jacchus, has increasingly been the subject of experiments for the analysis of somatosensory system function in simian primates. However, as response properties of the mechanoreceptive afferent fibers supplying the skin have not been characterized for this primate, the present study was undertaken to classify fibers innervating the glabrous skin of the marmoset hand and determine whether they resembled those described for other mammalian species, including cat, macaque monkey, and human subjects. Forty-seven tactile afferent fibers with receptive fields (RFs) on the glabrous skin of the hand were isolated in fine median and ulnar nerve strands. Controlled tactile stimuli, including static indentation and skin vibration, were used to classify fibers. Twenty-six (55%) responded to static indentation in a sustained manner and were designated slowly adapting (SA) fibers, while 21 (45%) were selectively sensitive to the dynamic components of the stimulus. The SA fibers had well-defined boundaries to their RFs, lacked spontaneous activity in most cases (23/26 fibers), had an irregular pattern of discharge to static skin indentation, and displayed graded response levels as a function of indentation amplitude, attributes that were consistent with the properties of slowly adapting type I (SAI) fibers described in other species. The dynamically sensitive afferent fibers could be subdivided into two distinct functional classes, based on their responses to vibrotactile stimulation. The majority (15/21) responded best to lower frequency vibration (~10-50 Hz) and had small RFs, whereas the second class responded preferentially to higher frequency vibration (50-700 Hz) with maximal sensitivity at ~200-300 Hz. These two classes resembled, respectively, the rapidly adapting (RA) and Pacinian corpuscle-related (PC) fiber classes found in other species, and like them, responded to vibration with tightly phase-locked patterns of response over a wide range of frequencies. The results demonstrate that the functional classes of tactile afferent fibers that supply the glabrous skin in the marmoset monkey appear to correspond with those described previously for the cat and macaque monkey, and are similar to those supplying the human hand and fingers, although the SA fibers in the human hand appear to fall into two classes, the SAI and SAII fibers. With the increasing use of the marmoset monkey as a primate model for somatosensory system studies, these data now allow tactile neurons identified at central locations, such as the cerebral cortex and thalamus, to be classified in relation to inputs from the peripheral classes identified in the present study.

Transmission from single, identified hair follicle afferent (HFA) nerve fibers to their target neurons of the cuneate nucleus was examined in anesthetized cats by means of paired recording from individual cuneate neurons and from fine, intact fascicles of the lateral branch of the superficial radial nerve in which it is possible to identify and monitor the activity of each group II fiber. Selective activation of individual HFA fibers was achieved by means of focal vibrotactile skin stimulation. Forearm denervation precluded inputs from sources other than the monitored HFA sensory fiber. Transmission characteristics were analyzed for 21 HFA fiber-cuneate neuron pairs in which activity in the single HFA fiber of each pair reliably evoked spike output from the target neuron at a fixed latency. As the cuneate responses to each HFA impulse often consisted of 2 or 3 spikes, in particular at HFA input rates up to approximately 20 imp/s, the synaptic linkage displayed potent amplification and high-gain transmission, characteristics that were confirmed quantitatively in measures of transmission security and cuneate spike output measures. In response to vibrotactile stimuli, the tight phase locking in the responses of single HFA fibers was well retained in the cuneate responses for vibration frequencies up to approximately 200 Hz. On measures of vector strength, the phase locking declined across the synaptic linkage by no more than approximately 10% at frequencies up to 100 Hz. However, limitations on the impulse rates generated in both the HFA fibers their associated cuneate neurons meant that the impulse patterns could not directly signal information about the vibration frequency above 50-100 Hz. Although single HFA fibers are also known to have secure synaptic linkages with spinocervical tract neurons, it is probable that this linkage lacks the capacity of the HFA-cuneate synapse for conveying precise temporal information, in an impulse pattern code, about the frequency parameter of vibrotactile stimuli.