Newly learned categories induce pre-attentive categorical perception of faces

Abstract

Face perception is modulated by categorical information in faces, which is known as categorical perception (CP) of faces. However, it remains unknown whether CP of faces is humans' inborn capability or the result of acquired categories. Here, we examined whether and when newly learned categories affect face perception. A short-term training method was employed in which participants learned new categories of face stimuli. Behaviorally, using an AB-X discrimination task, we found that the discrimination accuracy of face pairs from different learned categories was significantly higher than that of faces from the same category. Neurally, using a visual oddball task, we found that deviant stimuli whose category differed from standard stimuli evoked a larger N170. Importantly, the visual mismatch negativity (vMMN), starting from 140 ms after stimuli onset, was stronger with the between-category deviants than with the within-category deviants under the unattended condition. Altogether, our study provides empirical evidence indicating that CP of faces could be induced by newly learned categories, and this effect occurs automatically during an early stage of processing.One of the most remarkable abilities of the human visual system is recognizing individual faces quickly and efficiently. Many event-related potential (ERP) and magnetoencephalography (MEG) studies have investigated the time course of face processing and reported an early face-selective component peaking at approximately 170 ms after the stimulus onset (N170 or M170) 1,2 . Researchers proposed that the N170 is associated with the early structural encoding of faces [3][4][5][6] . Meanwhile, face perception is modulated by categorical information in faces (e.g., identity, race, and expression), i.e., discrimination of two faces is easier when the faces straddle a category boundary than when they belong to the same category 7 , which is a phenomenon known as categorical perception (CP). Here, we investigate how quickly the CP effect occurs during face perception.CP has been implicating in the processing of facial expressions [8][9][10][11] , race 12 , gender 13 and the identity of faces 7,14,15 . For example, Beale and Keil (1995) first reported that CP occurred in processing a morphed continuum between two familiar faces 7 . The authors used an AB-X discrimination task in which A and B were sequentially presented morphs followed by a target X that was identical to either A or B. The participants selected which morph (A or B) was identical to X. The discrimination accuracy was higher when A and B straddled the category boundary than when they were both from the same identity category. Moreover, face CP can occur early during the time window of the N170. For example, following the second face onset, the amplitude of N170 was larger when the two morphed faces stemmed from two different identities than that when the faces belonged to the same identity 16,17 . Studies exploring CP in facial expression have found similar results of N170 18 . However, it is unknown ...

Visual mismatch negativity and stimulus-specific adaptation: the role of stimulus complexity

“…Here, the shorter vMMN latency to the oblique bar pattern (a simple stimulus) indicates that deviancy is detected earlier in the context of complex stimuli than a complex deviant event within the sequence of more simple ones. When the standard and the deviant are similar [e.g., belonging to the same category (oblique lines: Kimura et al ( 2009 , 2010 ); facial categories and emotions: Yu et al ( 2017 ), Vogel et al ( 2015 ), Kreegipuu et al ( 2013 ); left vs. right hand: Stefanics and Czigler ( 2012 )], deviant-related negativity included longer latency ranges, or there were mismatch components in various ranges, whereas in the case of highly different standard and deviant (e.g., symmetric vs. asymmetric patterns), the difference potential was confined to an earlier and narrower latency range. Furthermore, using similar standards and deviants such as disappearing parts of an object (Sulykos et al 2017 ) and checkerboards with alternating locations of the dark and light squares (Sulykos et al 2018 ), an identical earlier phase of the vMMN appeared for both younger and older groups, whereas the later part was absent or diminished in the elderly.…”

No matter how: Top-down effects of verbal and semantic category knowledge on early visual perception

“…One way to control this is making the manipulated factors task-irrelevant (Firestone & Scholl, 2016). Additionally, for CP to count as an effect of language, we believe that effects should not be related to explicit perceptual training on a specific category boundary (Goldstone, Lippa, & Shiffrin, 2001;Notman, Sowden, & Özgen, 2005;Özgen & Davies, 2002;Yu et al, 2017). While perceptual training studies of CP are clearly interesting in a different context, they cannot provide evidence for a role of language in CP.…”

“…One recent study found color CP in the P1 component to predict access to visual consciousness in the attentional blink paradigm . Other studies found CP effects starting in the N1 time range (Boutonnet, Dering, Vinas-Guasch, & Thierry, 2013;Thierry et al, 2009;Yu et al, 2017), interpreted as a visual mismatch negativity, suggesting an effect of linguistic categories on preattentive perception (Thierry et al, 2009). In the context of visual object perception, the N1 is usually seen as an indicator of configural processing during early visual perception (Rossion & Jacques, 2011;Tanaka & Curran, 2001).…”

Category learning can alter perception and its neural correlates

“…Some learned CP effects are still open to the interpretation that they are not perceptual changes but a response bias from having learned to name the category (“naming bias” or “category label bias”): a tendency to judge things as less similar when their names are different and more similar when their names are the same [37–39]. One way to test whether CP effects are perceptual or verbal is to analyze brain activity during category learning.…”