Victor J D, Ashurova A, Conte M M, 2006, "Classification image analysis of texture discrimination" Perception 35 ECVP Abstract Supplement
Classification image analysis of texture discrimination
J D Victor, A Ashurova, M M Conte
Classification images (CIs) are a psychophysical probe of the computations underlying visual perception. However, application of CIs to texture discrimination is not straightforward, since standard reverse-correlation will not capture the contribution of second-order or higher-order image statistics. Five subjects identified the location of a 16 × 64 pixel texture-defined target within a 64 × 64 pixel background array. Target and background were chosen from a two-dimensional space of binary Markov random field textures, parameterised by their mean luminance and a 2 × 2 fourth-order correlation. Stimuli spanned the range of performance from near threshold to near ceiling. 4320 trials per subject were collected. CIs were determined after preprocessing stimuli to create 'derived images' representing pixel-by-pixel estimates of luminance or higher-order statistics. Reverse correlation yielded CIs that identified the footprint of the target but did not reveal internal structure. However, CIs extracted by regression combined with regularisation identified features not seen in the reverse correlation CIs: an accentuation of the contribution of luminance statistics, but not fourth-order statistics, near the target edge. Thus, CIs reflecting nonlinear processes may be readily obtained via analysis of appropriate derived images, and regularisation techniques may provide insights beyond those apparent from standard reverse-correlation maps.
[Supported by NIH EY7977.]
These web-based abstracts are provided for ease of seaching and access, but certain aspects (such as as mathematics) may not appear in their optimum form. For the final published version of this abstract, please see
ECVP 2006 Abstract Supplement (complete) size: 2368 Kb