Purchase this article with an account.
Kirill N. Shokhirev, Tribhawan Kumar, Donald A. Glaser; Estimation of the parameters of a visual stimulus from the responses of a realistic population of model visual neurons. Journal of Vision 2003;3(12):81. doi: 10.1167/3.12.81.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Properties of the receptive fields of neurons in the primary visual cortex are mapped by superimposing known visuotopic and orientation preference maps. Individual neurons respond to a limited range of stimuli around their preferred stimulus. Therefore, localized visual stimuli excite only a small population of neurons. The distribution of preferred orientations in the excited population depends on the spatial location of the stimulus because of irregularities in the orientation preference map. Such a non-uniform coverage could lead to variability of the accuracy of estimating the stimulus orientation. However, no systematic variation in the psychophysical performance has been reported. In this work we attempt to quantitatively evaluate the information available about the stimulus, as well as test the performance of a simple model for estimating stimulus parameters with realistic distribution of simple neurons.
We compute the Fisher information matrix for orientation and location of short line stimuli from the responses of a population of model simple cells arranged according to experimentally measured feature maps, and calculate the Cramer-Rao bounds on the accuracy of estimating these parameters. We find that the main source of uncertainty is the uneven coverage of the parameter space resulting from the layout of the cortical feature maps. However, these theoretical bounds are much smaller than the psychophysically measured thresholds for the entire range of parameters studied. We also investigated the performance of the linear population vector model for evaluating the stimulus parameters. We found that the accuracy of the linear estimation is comparable to the psychophysical results. This suggests that the accuracy of parameter estimation is limited by the estimation mechanism.
WestheimerG.(2003). The distribution of preferred orientations in the peripheral visual field. Vision Research 43 (1): 53–57
BoskingW.H.CrowleyJ.C.FitzpatrickD.(2002). Spatial coding of position and orientation in primary visual cortex. Nature Neuroscience 5 (9): 874–882
DasA.GilbertC.D.(1997). Distortions of visuotopic map match orientation singularities in primary visual cortex. Nature 387 (6633): 594–598
KapadiaM.K.WestheimerG.GilbertC.D.(1999). Dynamics of spatial summation in primary visual cortex of alert monkeys. PNAS 96 (21): 12073–12078
DeAngelisG.CGhoseG.M.OhzawaI.FreemanR.D.1999. Functional micro-organization of primary visual cortex: Receptive field analysis of nearby neurons Journal of Neuroscience 19 (10): 4046–4064
IttiL.KochC.BraunJ.(2000). Revisiting spatial vision: toward a unifying model J Opt Soc Am A Opt Image Sci Vis. 17(11):1899–1917
This PDF is available to Subscribers Only