Reduced sensitivity to contrast signals from the eye region in developmental prosopagnosia

Abstract

Contrast-related signals from the eye region are known to be important for the processing of facial identity. Individuals with developmental prosopagnosia (DP) have severe face recognition problems, which may be linked to deficits in the perceptual processing of identity-related information from the eyes. We tested this hypothesis by measuring N170 components in DP participants and age-matched controls in response to face images where the contrast polarity of the eyes and of other face parts was independently manipulated. In different trials, participants fixated either the eye region or the lower part of a face. In the Control group, contrast-reversal of the eyes resulted in enhanced and delayed N170 components, irrespective of the contrast of other face parts and of gaze location. In the DP group, these effects of eye contrast on N170 amplitudes were strongly and significantly reduced, demonstrating that perceptual face processing in DP is less well tuned to contrast information from the eye region. Inverting the contrast of other parts of the face affected N170 amplitudes only when fixation was outside the eye region. This effect did not differ between the two groups, indicating that DPs are not generally insensitive to the contrast polarity of face images. These results provide new evidence that a selective deficit in detecting and analysing identity-related information provided by contrast signals from the eye region may contribute to the face recognition impairment in DP.

Holistic face perception is impaired in developmental prosopagnosia

“…Based on this distribution, the reliability of an empirically observed ERP difference between conditions was determined for individual participants. If the probability of obtaining the observed difference by chance is below 5%, it is accepted as statistically significant (see Dalrymple et al, 2011;Oruc et al, 2011;Fisher et al, 2016Fisher et al, , 2017, for previous applications of this procedure in ERP studies of prosopagnosia). In the present experiment, this bootstrap procedure was based on EEG mean amplitudes obtained between 230 and 280 ms after S2 onset on full repetition and full change trials (collapsed the eight lateral posterior electrodes over the left and right hemisphere).…”

“…N250r components have been consistently observed over bilateral posterior occipito-temporal electrode locations, are accompanied by a fronto-central positivity, and are assumed to reflect a match between a face representation in working memory and an on-line perceptual representations of a particular face (e.g. Schweinberger, 2011;Schweinberger et al, 2004;Zimmermann & Eimer, 2013;Towler, Kelly, & Eimer, 2016;Fisher, Towler, & Eimer, 2016). Although the N250r is usually largest in response to repetitions of two physically identical face images (e.g., Schweinberger et al, 2002), this component is also reliably present for repetitions of different images of the same face (Bindemann et al, 2008;Zimmermann & Eimer, 2013, demonstrating that the underlying face identity matching processes are at least partially image-independent.…”

Face identity matching is selectively impaired in developmental prosopagnosia

“…Based on this distribution, the reliability of an empirically observed ERP difference between conditions was determined for individual participants. If the probability of obtaining the observed difference by chance is below 5%, it can be accepted as statistically significant (see Dalrymple et al, 2011;Eimer, et al 2012;Oruc et al, 2011;Fisher et al, 2016b, for previous applications of this procedure in ERP studies of prosopagnosia). In the present experiment, this bootstrap procedure was based on EEG mean amplitudes obtained between 220 and 320 ms after S2 onset on identity repetition and identity change trials where facial expression was repeated (collapsed the eight lateral posterior electrodes over the left and right hemisphere).…”

“…Most individuals with DP show normal N170 components to faces versus non-face objects (Towler, Gosling, Duchaine, & Eimer, 2012), suggesting that the ability to perceptually discriminate between faces and non-faces is largely intact. However, changes to the prototypical spatial configuration and contrast properties of face images (such as presenting face images upside-down, spatially scrambling internal facial features, or contrast-inverting the eye region) produce atypical N170 amplitude modulations in individuals with DP Towler, Parketny, & Eimer, 2016;Fisher, Towler, & Eimer, 2016b). This suggests that perceptual face processing mechanisms in DP may be less well tuned to the spatial configuration and contrast properties that are the defining features of a typical upright face, and are therefore less sensitive to deviations from a prototypical visual face template.…”

“…Importantly, instead of being generated at the stage where working memory representations are formed, identity-related deficits in DP may already emerge at earlier sensory-perceptual stages of face processing (see also Shah et al, 2016, for similar suggestions). Previous ERP studies of DP that focused on the N170 component have found evidence that DPs are less sensitive to the prototypical spatial configuration of upright faces and to contrast signals from the eye region (Fisher et al, 2016b). Such spatial-configural and contrast-related related signals, in particular from the eyes, provide important cues to identity (e.g., Gilad, Meng, & Sinha, 2009), because they remain invariant across changes in expression and other image changes (e.g., Burton, 2013).…”