Contrast versus identity encoding in the face image follow distinct orientation selectivity profiles

Abstract

Orientation selectivity is a fundamental property of primary visual encoding. High-level processing stages also show some form of orientation dependence, with face identification preferentially relying on horizontally-oriented information. How high-level orientation tuning emerges from primary orientation biases is unclear. In the same group of participants, we derived the orientation selectivity profile at primary and high-level visual processing stages using a contrast detection and an identity matching task. To capture the orientation selectivity profile, we calculated the difference in performance between all tested orientations (0, 45, 90, and 135˚) for each task and for upright and inverted faces, separately. Primary orientation selectivity was characterized by higher sensitivity to oblique as compared to cardinal orientations. The orientation profile of face identification showed superior horizontal sensitivity to face identity. In each task, performance with upright and inverted faces projected onto qualitatively similar a priori models of orientation selectivity. Yet the fact that the orientation selectivity profiles of contrast detection in upright and inverted faces correlated significantly while such correlation was absent for identification indicates a progressive dissociation of orientation selectivity profiles from primary to high-level stages of orientation encoding. Bayesian analyses further indicate a lack of correlation between the orientation selectivity profiles in the contrast detection and face identification tasks, for upright and inverted faces. From these findings, we conclude that orientation selectivity shows distinct profiles at primary and high-level stages of face processing and that a transformation must occur from general cardinal attenuation when processing basic properties of the face image to horizontal tuning when encoding more complex properties such as identity.

Psychophysical profiles in super-recognizers

“…Meanwhile, in our second experiment, we find horizontal selectivity commensurate with previous studies 31 , 32 , 34 , 35 , as suggested by improved model fits when adding orientation as a factor (Figs. 2 , 3 ).…”

“…In the first, we test whether consistent exploitation of retinally available spatial frequency information (across simulated viewing distances) 27 during unfamiliar face matching differentiates SRs from a sample of control observers. In the second, testing the same two groups of observers, we assess whether such (potential) differences extend to increased consistency in horizontal versus vertical spatial frequency structure as shown in previous research 31 – 35 .…”

Super-Recognizers: More Consistent or Qualitatively Different Psychophysical Profiles?

“…In our second experiment, we find horizontal selectivity commensurate with previous studies (Goffaux, Poncin & Schiltz, 2015;Goffaux & Greenwood, 2016;Pachai et al, 2017;Jacobs et al, 2020). Specifically, horizontal spatial frequency information is more diagnostic for facial identity matching than vertical, among both groups of observers.…”

“…In the first, we test whether consistent exploitation of retinally available spatial frequency information (as in Figure 1a) during unfamiliar face matching differentiates SRs from a sample of normal observers as controls (See Method section for details). In the second, testing the same two groups of observers, we assess whether such (potential) differences extend to greater selectivity for horizontal spatial frequency structure as shown in previous research (Goffaux, Poncin & Schiltz, 2015;Taubert et al, 2016;Goffaux & Greenwood, 2016;Jacobs et al, 2020) (See Method section for details).…”

Backward masking reveals coarse-to-fine dynamics in human V1

“…Since manipulations in the Fourier domain operate on the whole image (face and background), the presence of a uniform background artificially increases the LSF energy of scrambled images. In order to minimise contamination of the phase-scrambled image by its originally uniform background, we employed an iterative phase scrambling procedure (as in Jacobs et al, 2020; Petras et al, 2021; Petras et al, 2019; see Figure 1a). Iterative scrambling involves first generating a phase scrambled version of the face image, then superimposing the original face back onto the scrambled image, scrambling this new image again, and repeating the procedure several times (for all our scrambled stimuli, 500 iterations were used).…”