Some quantitative aspects of vibrissa-driven neuronal responses in rat neocortex.

1. Single-unit responses to whisker deflections were studied in the rat somatosensory cortex (SI) under urethan anesthesia. 2. Experiments were concerned primarily with cortical layer V neurons, by selecting those units that 1) displayed initially positive biphasic spikes with an amplitude of at least 2 mV, 2) were located between 1,000 and 1,500 pm below the cortical surface of whisker C3 representation, and 3) responded optimally to either dorsal or ventral deflection of a “center” whisker. 3. Manual stimulation tests showed that 3 1% of the total sample responded to deflections of one whisker only, while the remainder responded to two or more whiskers (up to 14 whiskers). The average of the receptive-field (RF) size thus determined was 3.9 in terms of the number of effective whiskers. 4. Manually assessed directional preference classes, “strong,” “weak,” and “nondirectional,” were corroborated by threshold measurements using a calibrated whisker stimulator. 5. The latency of unit responses to supramaximal deflections of the center whisker in the most preferred direction averaged 10.6 ms, When tested with deflections of 5 ms duration, the velocity threshold (measured at 10 mm from the whisker base) was 23.5 mm/s (or 134.4”/s) on the average, the amplitude threshold was 117 pm (0.68”). 6. Analysis of correlations among the four measures, the latency, the threshold, the RF size, and the degree of directional selectivity, revealed that 1) the longer the latency of a neuron was, the higher was its threshold; and 2) neurons with larger RFs tended to show lower thresholds and lesser degree of directional selectivity. 7. Threshold tuning curves were constructed by measuring the amplitude thresholds at different time constants of exponential deflections of a whisker and by plotting them on a log-log graph as a function of the equivalent sine-wave frequency. All tuning curves were negatively sloped, and most of the slopes were between 0 (ideal amplitude detector) and 1 (ideal velocity detector). The highly significant linearity of individual slopes suggested a constant contribution of the amplitude and velocity components of whisker deflection within the range of freqencies tested (0.31-17.8 Hz) in a given neuron. Although the ratio of contribution of the two components was different for different neurons, these neurons, as a population, were more sensitive to the velocity than to the amplitude, as indicated by a mean slope value of -0.78.