On the distinctness of simple and complex cells in the visual cortex of the cat

The behaviour of neurones in cat striate cortex was examined in response to moving sinusoidal gratings and flashed bright and dark lines. The responses were summarized by three indices: discreteness was a measure of the degree of separation of inhibitory and excitatory regions in the receptive field; spatial summation ratio showed the degree of spatial summation within each region; relative modulation was a measure of the degree of modulation in the response to a moving grating. Some neurones had receptive fields with completely discrete excitatory and inhibitory regions; others responded equally to stimulus onset and offset throughout their receptive fields; however, some had overlapping excitatory and inhibitory regions. The degree of overlap varied continuously from complete separation to complete overlap. For neurones with discrete receptive fields, the widths of the regions were compared with the width of the bars in a grating of optimum spatial frequency to assess the degree of spatial summation within the regions. Most neurones with discrete receptive fields showed roughly predictable spatial summation, in that the two width measures agreed; but about 10% of them had receptive field regions that were too large by a factor of over two. The neurones which showed incomplete spatial summation also had considerable overlap of their excitatory and inhibitory regions. The waveforms of the responses to moving gratings of optimal spatial frequency were examined. The degree of modulation in the response was continuously distributed between low values typical of complex cells and high values typical of simple cells; the distribution was not bimodal. The degree of response modulation was closely correlated with the degree to which the excitatory and inhibitory regions in the receptive field were discrete. Both the degree of spatial summation and the degree of response modulation have been previously proposed as means for distinguishing simple and complex cells. In the present study, the continuity of the distributions of both indices ensured that neither index alone could be used to class all neurones unequivocally. However, a criterion based on two indices did allow classification. Simple and complex cells showed distinctive behaviour. However, complex cells with distinguishable excitatory and inhibitory regions in their receptive fields were not distinctly different from other complex cells.