Categorizing facial identities, emotions, and genders: Attention to high- and low-spatial frequencies by children and adults

Abstract

Three age groups of participants (5–6 years, 7–8 years, adults) matched faces on the basis of facial identity. The procedure involved either low- or high-pass filtered faces or hybrid faces composed from two faces associated with different spatial bandwidths. The comparison stimuli were unfiltered faces. In the three age groups, the data indicated a significant bias for processing of low-pass information in priority. In a second task, participants were asked to identify the emotion (smiling or grimacing) or gender (male or female) of hybrid high-pass/low-pass faces. Opposite results emerged in the two tasks irrespective of the age group; the gender discrimination task indicated a bias for low-pass information, and the emotion task indicated a bias for high-pass information. These differences suggest independent processing routes for functionally different types of information such as emotion, gender, and identity. These routes are already established by 5 years of age.

Introduction

Faces are polymorphous stimuli conveying various kinds of information such as gender, emotion, and identity. Most models of face processing posit that these different types of information are processed by functionally different neural/cognitive systems (e.g., Bruce & Young, 1986). The hypothesis that different systems process gender, emotion, and identity information (hereafter referred to as the hypothesis of functional separability) is supported in the literature by the following five lines of converging evidence. First, gender and identity are processed at different speeds than are emotions in judgmental tasks (e.g., Le Gal & Bruce, 2002). Second, variations of emotional expressions have little effect on facial identity judgments, but reaction times for emotion judgments are influenced by identity variations (e.g., Schweinberger & Soukup, 1998). Third, distinct patterns of cerebral activation emerge for the recognition of identity, expression, and gender (from event-related potentials [ERP] studies: e.g., Boles, Martin, Olivares, & Valdés-Sosa, 2000; from functional magnetic resonance imaging [fMRI] studies: e.g., McCarthy, Puce, Gore, & Allison, 1997; from positron emission tomography [PET] studies: e.g., Sergent, Otha, MacDonald, & Zuck, 1994). Fourth, brain lesions selectively affect adults’ recognition of facial identity and emotion (e.g., Humphreys, Donnelly, & Riddoch, 1993). Fifth, in fMRI studies, presentation of low- and high-pass filtered faces results in distinct patterns of processing of face and emotional expressions (e.g., Vuillemier, Armony, Driver, & Dolan, 2003). Interestingly, all of these findings supporting the hypothesis of functional separability are based on studies of normal adults or patients. There is surprisingly little developmental research on this topic.

In one of the few relevant studies, Bruce et al. (2000) studied the development of expression, lipreading, gaze, and identity processing in 4- to 10-year-olds. There were no significant correlations among matching-to-sample tasks in the different test conditions, suggesting that these various aspects of faces are already processed independently in children of that age. In another study (Bormann-Kischkel, 1986), 5-year-olds and adults were asked to sort cards that varied in terms of identity and emotion expressions. Adults and children showed a bias for sorting on an emotion basis rather than on an identity basis, suggesting that emotion and identity can be tapped independently. Finally, De Sonneville et al. (2002) tested children (7–10 years of age) and adults in face identity and face emotion discrimination tasks. Regardless of age, face recognition was faster than emotion recognition. These three studies tend to support the hypothesis of separability in children, and this would confirm as well as expand what is known about adults. Conclusions cannot be guaranteed, however, due to the reduced number of available developmental studies. These studies, moreover, convey no detailed information on the stimulus dimensions to which children paid attention when categorizing facial emotion and identity.

In that context, the current research was aimed at better documenting the relative development during childhood of perception of identity, gender, and emotion. To better characterize the processing of emotion, gender, and identity, participants at three ages (5–6 years, 7–8 years, and adults) were tested in the current research using spatial frequency filtered facial stimuli. Our reasoning was that controlling the frequency band available on faces should indicate whether different spatial frequency channels are used by children when processing identity, emotion, and gender facial information.

The two experiments reported here focus on the processing of identity (Experiment 1) and on the processing of both gender and emotion (Experiment 2). Although presented in succession, these two experiments were actually run in a balanced order, with half of the participants being tested at first in Experiment 1 and the other half being tested at first in Experiment 2.