The rodent barrel field cortex integrates somatosensory information from two separate thalamic nuclei, the ventral posterior medial nucleus (VPM) and the rostral sector of the posterior complex (POm). This paper compares the sensory responses of POm and VPM cells in urethane-anesthetized rats as a first step in determining how cortex integrates multiple sensory pathways. A complete representation of the contralateral body surface was identified in POm. Trigeminal receptive fields (RFs) of POm and VPM cells were mapped by computer-controlled displacement of individual whiskers; responses were quantified by using peristimulus time histograms. Average RF size was similar in POm (5.1 whiskers) and VPM (4.4 whiskers), but evoked responses in the two nuclei differed significantly according to all other measures. VPM cells were maximally responsive to one single whisker--the "center RF." Stimulating this whisker evoked, on average, a response of 1.4 spikes/stimulus at a latency of 7 ms; surrounding whiskers evoked responses of less than 1 spike/stimulus at latencies of greater than 8 ms. In contrast, POm cells were nearly equally responsive to several whiskers. Quantitative criteria allowed us to designate a single whisker as the "center RF" and stimulating this whisker evoked, on average, a response of 0.5 spikes/stimulus at a latency of 19 ms. VPM cells, but not POm cells, were able to "follow" repeated whisker deflection at greater than 5 Hz. We conclude that, when a single whisker is deflected, VPM activates the related cortical barrel-column at short latency--before the onset of activity in POm. The timing of activation could allow POm cells to modulate the spread of activity between cortical columns.

This study tested the hypothesis that the receptive fields (RFs) of neurons in the adult sensory cortex are shaped by the recent history of sensory experience. Sensory experience was altered by a brief period of "whisker pairing": whiskers D2 and either Dl or D3 were left intact, while all other whiskers on the right side of the face were trimmed close to the fur. The animals were anesthetized 64-66 h later and the responses of single neurons in contralateral cortical barrel D2 to stimulation of whisker D2 (the center RF) and the four neighboring whiskers (Dl, D3, C2, and E2; the excitatory surround RF) were measured. Data from 79 cells in four rats with whiskers paired were compared to data from 52 cells in four rats with untrimmed whiskers (control cases). During the period of whisker pairing, the RFs of cells in barrel D2 changed in three ways: (i) the response to the center RF, whisker D2, increased by 39%, (u) the response to the paired surround RF whisker increased by 85-100%, and (ui) the response to all clipped (unpaired) surround RF whiskers decreased by 9-42%. In the control condition, the response ofbarrel D2 cells to the two neighboring whiskers, Dl and D3, was equal. After whisker pairing, the response to the paired neighbor of D2 was more than twice as large as the response to the cut neighbor of D2. These findings indicate that a brief change in the pattern of sensory activity can alter the configuration of cortical RFs, even in adult animals.The purpose of this study is to examine the influence of sensory experience on the functional properties of neurons in the somatosensory cortex (SI). The rodent whisker sensory system offers several advantages for studying the mechanisms of experience-dependent cortical plasticity. The projection from the whisker follicles to the contralateral SI preserves the spatial organization of the sensory receptors, resulting in a somatotopic map of cortical columns (1, 2). Layer IV of each column contains a "barrel," a discrete cluster of closely packed cells that is readily identified by several histological markers (3, 4). The one-to-one correspondence between each whisker and its cortical barrel makes it possible to relate whisker-evoked single-unit cortical activity to distinct thalamocortical and corticocortical pathways. For example, cells in barrel D2 are excited quickly (6-to 10-ms latency) and powerfully (1-2 spikes per stimulus) by deflection of whisker D2 (5). Because this response depends upon direct inputs from the thalamic ventral posterior medial nucleus (6), it is convenient to call whisker D2 the "center receptive field" (CRF) of barrel D2. Deflection of neighboring whiskers (e.g., Dl, D3, C2, or E2) excites cells in barrel D2 less strongly (approximately one spike every second or third stimulus) at a longer latency (20 ms on average). Since the response to these whiskers is generated in large part by a separate pathway-intracortical inputs from surrounding barrels (7)-the neighboring whiskers are referred to as the "excitatory surround recept...

The projection from the whiskers of the rat to the S-I (barrel) cortex is segregated into two separate pathways--a lemniscal pathway relayed by the ventral posterior medial nucleus (VPM) to cortical barrels, and a paralemniscal pathway relayed by the rostral sector of the posterior complex (POm) to the matrix between, above, and below barrels. Before investigating how the barrel cortex integrates these sensory pathways, it is important to learn more about the influence of the various inputs to the two thalamic nuclei. Based on the greater density of descending versus ascending projections to POm, it seemed likely that corticofugal inputs play an important role in the sensory activity of POm. To test this, the responses of POm and VPM cells to sensory stimuli were measured before, during, and after suppression of the S-I cortex. S-I was suppressed by application of magnesium or by cooling; the status of the barrel cortex was assessed continuously by an electrocorticogram. All VPM cells (n = 8) responded vigorously to whisker movement even when the barrel cortex was profoundly depressed. In contrast, all POm cells (n = 9) failed to respond to whisker movement once the barrel cortex became depressed, typically about 25 minutes after the start of cortical cooling or magnesium application. POm cells regained responsiveness about 30 minutes after the cessation of cortical cooling or the washoff of magnesium. These findings indicate that the transmission of sensory information through the lemniscal pathway occurs independently of the state of cortex, whereas transmission through the paralemniscal pathway depends upon the state of the cortex itself.

1. Extracellular spike recordings were made from single cells in various layers of barrel cortex in adult rats anesthetized with urethan. Response magnitude and latency differences to brief 1.14 degrees deflections of mystacial vibrissae of center (principal) and surround receptive-field vibrissae were measured. Latency differences for pairs of cells in the same penetration to stimulation of the principal vibrissa were also collected. In separate experiments the domains of layer IV cells were mapped for their influence by a single vibrissa and their latencies to this vibrissa were recorded. In all experiments precise locations of layer IV cells in each penetration were identified using dye-lesioning and cytochrome oxidase staining of tangential sections. 2. The results suggest that principal vibrissa data are relayed radially in a column of neurons before parallel relay to adjacent columns. To the principal vibrissa, layers IV and Vb neurons discharged earliest, with layers II and III on average 2 and 3 ms later, respectively. Serial relay from layers IV to III to II was suggested to be the most common event. Although layer Va cells fired next, a single-column organization is not suggested for them because differences in latency or response magnitude to their principal and immediate surround vibrissae were not significant. Layer II, III and IV cells showed no statistical difference in latency to the nearest surround vibrissa but fired significantly later than to their principal input. 3. Because, from our previous studies, surround receptive fields of barrel cells in rat S1 cortex appear to be constructed intracortically, these data suggest a parallel column-column relay for their construction. Horizontal relay between barrels occurred first within the septae between barrels. Mean intracortical transmission velocities were calculated at approximately 0.05 m/s for column-column information transfer.