Purchase this article with an account.
Sean P. MacEvoy, Thomas R. Tucker, David Fitzpatrick; Characterizing V1 population responses to superimposed gratings. Journal of Vision 2005;5(8):429. doi: 10.1167/5.8.429.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Previous studies have shown that the responses of V1 neurons to an optimally oriented grating are usually suppressed by superimposition of a second grating at a non-preferred orientation. However, these single unit studies leave open the question of how two orientations are simultaneously represented in the distribution of population activity. To explore this question, we acquired intrinsic signal optical images of tree shrew V1 during presentation of a stationary full-field grating presented alone and in combination with a second grating at a range of orientations. Consistent with single-unit results, regions of cortex that are activated by each of the gratings presented singly are less activated by the combined stimulus. To quantify activity patterns we constructed a population response profile (PRP), a distribution which represents, for each orientation value, the summed activity of all pixels with that preferred orientation. We find that the height, width, and position of the PRP for the combined stimulus is well predicted by the mean of PRPs derived from the component gratings. This holds true for orientation differences between components ranging from 20 to 90° and over a range of contrasts. Thus the peak of activation for the combined stimulus is shifted away from the peaks for the components, and for gratings separated by 20° or less, the pattern of activity is often indistinguishable from that produced by a single lower-contrast grating at an intermediate orientation. As with gratings, responses to isolated line intersections are predicted by the mean of responses to the component lines, but only within 1mm of the cortical representation of their crossing point. Overall, we find that population responses to multiple orientations are governed by a simple scaling rule consistent with a local circuit-based mechanism of divisive cortical gain control.
This PDF is available to Subscribers Only