Layer-Specific Somatosensory Cortical Activation During Active Tactile Discrimination

See allHide authors and affiliations

Science  25 Jun 2004:
Vol. 304, Issue 5679, pp. 1989-1992
DOI: 10.1126/science.1093318


Ensemble neuronal activity was recorded in each layer of the whisker area of the primary somatosensory cortex (SI) while rats performed a whisker-dependent tactile discrimination task. Comparison of this activity with SI activity evoked by similar passive whisker stimulation revealed fundamental differences in tactile signal processing during active and passive stimulation. Moreover, significant layer-specific functional differences in SI activity were observed during active discrimination. These differences could not be explained solely by variations in ascending thalamocortical input to SI. Instead, these results suggest that top-down influences during active discrimination may alter the overall functional nature of SI as well as layer-specific mechanisms of tactile processing.

Anatomical and physiological evidence indicates that top-down inputs may have a significant effect on mechanisms of sensory processing (13). Yet the nature of top-down influences on cortical function in primary sensory areas remains poorly understood. As a rat actively samples a tactile stimulus, particularly one with salient behavioral relevance, several higher order processes that would not be activated by random passive stimuli will likely be engaged, such as attention, motivation, sensory-motor integration, stimulus-reward association/expectation, etc. However, what effect these or other higher order processes might have on mechanisms of tactile processing in the rat primary somatosensory cortex (SI) remains largely unknown.

We recorded the activity of ensembles of single neurons through different layers of the barrel region of SI in five rats trained to perform a whisker-dependent tactile discrimination task (4, 5). The task required rats to actively sample a variable-width aperture with their large facial whiskers, and then signal whether the aperture was “narrow” or “wide” (Fig. 1A). Video analysis showed that rats approached and sampled the aperture in a very repeatable stereotypical manner, using only their large facial whiskers to contact the aperture (Fig. 1B). Single-unit activity was recorded through all layers of SI with chronically implanted movable arrays of high-impedance (Z ≅ 2.5 to 3 megohms at 1 KHz) microwire electrodes. Electrodes were oriented perpendicular to the cortical surface, so that recordings along an individual electrode track were from the same cortical column (Fig. 1C). This allowed neural activity recorded at different depths through individual cortical columns to be compared (Fig. 1, D to F).

Fig. 1.

(A) Schematic of the behavioral apparatus. Trials begin when the sliding door opens. Rats enter the center discrimination chamber and sample the variable-width aperture with their facial whiskers. Rats then poke their nose into either the left or right reward nose poke to receive a water reward: the left nose poke if the aperture was narrow (60 mm), the right nose poke if it was wide (68 mm). Immediately after, the sliding door closes and the aperture is randomly reset to wide or narrow. The next trial begins 30 s later. (B) Video frame captures showing a rat approaching and sampling the narrow and wide apertures. The 0.0 s frames (topmost frames) show the rat breaking the infrared photo-beam; the middle frames (0.1 s later) show the whiskers initially contacting the aperture; the bottom frames (0.2 s) show the whiskers fully contacting the aperture. (C) Representative histological section showing an electrode track (dashed line) and marking lesion (solid arrow) at the bottom of the track used to reconstruct the depth of recorded units. (D to F) Representative examples of single-unit activity (displayed with two time scales) recorded at three different depths: (D), layer III; (E), layer IV; (F), layer V.

A total of 317 units were recorded bilaterally in the barrel region of SI while the rats performed the active tactile discrimination: 114 units in supragranular layers, 105 in layer IV, and 98 in infragranular layers. Sixty-seven percent (212) of these units displayed significant modulations in firing rate as rats performed the tactile discrimination: 37% (78) showed significant increases in firing (excitatory responses), 26% (56) showed decreased firing (inhibitory responses), and 37% (78) had multiphasic responses consisting of combinations of increases and decreases. The overall percentage of responsive units per layer did not differ significantly (64% in supragranular layers, 70% in layer IV, and 66% in infragranular layers).

These responses were compared with the activity of 244 units recorded in SI of 10 additional rats that received several different forms of passive whisker stimulation designed to simulate the spatiotemporal dynamics of whisker deflections that occurred during the active discrimination. Three of the rats received patterned ramp-and-hold stimulation of 16 individual whiskers with a multichannel whisker stimulator (5, 6) while lightly anesthetized (Fig. 2A). Three rats were habituated to calm, head-fixed restraint and received similar patterned ramp-and-hold stimuli (Fig. 2A) while fully awake (5, 7). Four rats received bilateral whisker stimulation with a movable aperture while lightly anesthetized or restrained and fully awake (Fig. 2B). This moving aperture (the same size as the aperture in the active discrimination task) was accelerated across the facial whiskers at velocities and trajectories that replicated the whisker deflection dynamics that occurred during active discrimination (5).

Fig. 2.

(A) The upper schematic shows the pattern of multiwhisker ramp-and-hold passive stimuli delivered to anesthetized rats (5). Large black dots represent stimulation of a particular whisker. Upward arrows show stimulation onsets. The lower schematic shows the stimulation pattern of the awake restrained rats. (B) (Left) Schematic of the moving-aperture stimulus. The aperture is accelerated across the facial whiskers (with variable onsets and velocities) by the pneumatic solenoid and also simultaneously deflected laterally in varying amounts by the dc servo in order to accurately replicate the range of whisker deflection dynamics that occurred during active discrimination (5). (Right) Video frame captures showing an example of the aperture moving caudally across the whiskers of an awake restrained rat while simultaneously deflecting laterally 5 mm (to the right) over a 200-ms interval. (C) Representative single-unit responses showing long-duration tonic activation during active discrimination. The upper portion of each panel is a raster plot where each line represents a consecutive trial in a recording session, and each dot is a unit spike; the lower portion of each panel shows summed activity for all trials in 5-ms bins. The 0 time point represents the moment when rats disrupted the aperture photobeam (Fig. 1). (D) Representative single-unit responses evoked by passive ramp-and-hold stimulation of 16 whiskers in lightly anesthetized rats (upper panel) and by passive stimulation of 8 whiskers in awake restrained rats (lower panel). The 0 time point represents stimulus onset. (E) Representative single-unit responses evoked by moving-aperture stimulation of awake restrained rats (the 0 time point represents the onset of aperture movement). (F) Mean (+SEM) excitatory response duration and magnitude evoked during active discrimination and by the different passive stimuli delivered to anesthetized or awake restrained rats.

Excitatory responses displayed a distinct shift from phasic activation during passive stimulation to tonic activation during active discrimination. This tonic activation (Fig. 2C) consisted of sustained increases in firing rate with a mean duration (5) that was significantly different across cortical layers as follows: supragranular layers, 282 ± 29 ms (mean ± SEM); layer IV, 207 ± 16 ms; and infragranular layers, 339 ± 19 ms [F (2,54) = 8.8, P < 0.001]. The magnitude of tonic responses (5) during active discrimination also varied significantly across layers as follows: supragranular layers, 7.0 ± 0.7 spikes per trial; layer IV, 4.3 ± 0.5 spikes per trial; and infragranular layers, 9.5 ± 1.1 spikes per trial [F (2,54) = 10.6, P < 0.0005].

In contrast, passive ramp-and-hold whisker stimulation (in either anesthetized or awake restrained rats) or moving-aperture stimulation (in anesthetized or awake rats) evoked excitatory responses that consisted of relatively brief, transient increases in activity (Fig. 2, D and E). The mean excitatory response durations and magnitudes evoked by these passive stimuli are summarized in Fig. 2F. Response duration during active discrimination was highly significantly different from that under each of the passive stimulation conditions [F (4,85) = 59.2, all P's < 0.0005, Tukey honestly significant difference (HSD)]. The same was true of the response magnitude measure [F (4,85) = 29.9, all P's < 0.00001, Tukey HSD]. In short, excitatory activation of SI by passive whisker stimulation was fundamentally different in nature than SI excitation evoked by the active discrimination. Several additional lines of evidence indicate that this shift from phasic to tonic activation during passive and active stimulation, as well as several other functional differences described below, could not have resulted solely from variations in whisker deflection dynamics during passive and active stimulation (see supporting online text).

Another functional difference between active and passive stimulation was a significant shift in the relative balance between excitatory and inhibitory responses. Only 7% of SI units responded to the ramp-and-hold or movingaperture passive stimulation with either purely inhibitory responses or inhibitory followed by excitatory activation (Fig. 3A); the remaining 93% responded with either purely excitatory or excitatory followed by inhibitory activity, results that are consistent with earlier studies (811). Moreover, there was no significant difference in the percentage of inhibitory responses evoked by the different passive stimuli delivered to either anesthetized or awake restrained rats. In contrast, nearly half (49%) of SI units responded during the active discrimination with either a purely inhibitory response (26%) or a response that was initially inhibitory followed by an excitatory phase (23%) (Fig. 3B). The mean duration of inhibitory responses during the active discrimination (246 ± 21 ms) was significantly longer than the duration of inhibitory responses evoked by the passive stimuli (12.4 ± 1.8 ms) [t (12) = 9.9, P < 0.0001]. During active discrimination, purely inhibitory responses were evenly distributed across layers: 33% in supragranular layers, 36% in layer IV, and 31% in infragranular layers; inhibitory-excitatory responses were asymmetrically concentrated in the granular layer: 23% in supragranular layers, 54% in layer IV, and 23% in infragranular layers. Collectively, these different inhibitory responses indicate that the activation dynamics of SI after passive whisker stimulation are fundamentally different than those during active discrimination. Passive multiwhisker stimulation causes a predominately excitatory deviation from prestimulus baseline activity, whereas similar stimulation during active discrimination evokes almost equally balanced excitatory and inhibitory shifts from baseline.

Fig. 3.

(A) Representative inhibitory responses evoked by passive ramp-and-hold stimulation. (B) Representative inhibitory responses evoked by active discrimination. See Fig. 2 legend for the description of plots.

The functional nature of these actively evoked inhibitory responses was examined using an artificial neural network, based on the learning vector quantization (LVQ) algorithm (fig. S1A and associated supporting online text). Results revealed that information about different aperture widths is encoded simultaneously by excitatory and inhibitory responses in SI (see LVQ-based analyses in supporting online text). Further, LVQ analysis of ensemble activity in different layers indicates that layer-specific tactile coding mechanisms may be engaged during active discrimination (fig. S1B). For instance, single-trial prediction of wide and narrow apertures by infragranular ensembles was significantly more accurate than prediction by supragranular ensembles [t (10) = –2.62, P < 0.05] (supporting online text).

Finally, cortical columns displayed significantly different functional properties during active discrimination and passive stimulation. SI recordings in monkeys, cats, and rats show that responses within a column evoked by passive stimulation share similar functional properties of place (all units have a common receptive field locus) and mode (all units respond to similar stimulus modalities) (8, 9, 1215). In contrast, during the active discrimination described here, 36% of infragranular units began responding significantly before the rats' whiskers contacted the aperture during active discrimination. More important, these units began responding significantly earlier than units recorded directly above them in supragranular or granular layers (Fig. 4A). Onset latencies between supragranular and granular layers did not differ significantly (Fig. 4B). In contrast, onset latencies of infragranular units were significantly earlier than units in the more superficial layers (Tukey HSD, P < 0.0005).

Fig. 4.

(A) Examples of single-unit responses recorded at different depths along three different electrode tracks during active discrimination (wide aperture). Units recorded in infragranular layers respond significantly earlier than units in more superficial layers and before whiskers contact the aperture. (B) Distribution of onset latencies for all responsive cells recorded in the different layers during active discrimination. (C) Summary diagram showing significant major effects across layers during active discrimination. The “Early-On” column shows the percentage of units showing early onsets per layer. LVQ: Mean (+SEM) performance of the LVQ for populations recorded at the different depths. Duration: Mean (+SEM) excitatory response duration. Magnitude: Mean (+SEM) magnitude of excitatory responses. MPI: Distribution of units with multiphasic responses that began with an inhibitory phase. Excitatory: Percentage of units with excitatory responses. Multi: Percentage of multiphasic units.

These early responsive units in infragranular layers appear to represent a functionally different class of neurons. First, early responses were not seen in layer IV, indicating that the afferent source(s) of these responses was not ascending thalamic input. Second, video analysis of rats performing the task shows that these responses occurred as the rats were moving toward the aperture, although no distinct tactile stimuli appeared to contact the whiskers. Third, the duration of these early responses was significantly longer than responses of other infragranular units that responded only when the whiskers contacted the aperture [389 ± 19 versus 293 ± 19 ms; t (28) = –3.33, P < 0.005]. Finally, the whiskers on one side of the face of one rat were cut before a behavioral recording session. Early-onset responses were still observed in the infragranular layers contralateral to the whisker cut. Together these results indicate that the early-onset units are not activated by whisker stimulation directly. As such, these early-onset responses do not appear to share the same functional properties of place and modality as units recorded more superficially in the same cortical column.

Numerous functionally significant differences in SI activity were observed in different cortical layers as rats performed an active tactile discrimination task (Fig. 4C). Moreover, fundamental differences in the functional nature of SI activity were observed during active discrimination and passive whisker stimulation. These results suggest that SI receives significantly different afferent input during actively acquired and passively delivered stimuli. However, these differences do not appear to result exclusively from changes in bottom-up ascending input to SI (see supporting online text). Instead, during active discrimination, top-down influences may also affect tactile processing in SI. For instance, the early-onset units were observed only in infragranular layers and not in layer IV, indicating that these responses did not arise from ascending thalamic input. Because motor cortex (MI) sends a significant projection to SI infragranular layers (16, 17), these early-onset responses might represent modulation from MI as rats initiate the discrimination. Also, excitatory response durations and magnitudes in supra- and infragranular layers were substantially greater than those in layer IV during active discrimination, indicating that the nongranular laminae received additional excitatory input from sources other than ascending thalamocortical input. Possible sources of this input include the secondary somatosensory cortex (18, 19) and the contralateral SI (1922), both of which innervate the nongranular laminae. Finally, the functional differences observed in SI during active and passive stimulation also indicate that passively evoked responses are a relatively poor predictor of layer-specific tactile processing mechanisms during active discrimination.

Supporting Online Material

Materials and Methods

SOM Text

Figs. S1 and S2


References and Notes

View Abstract

Navigate This Article