Academic journal article Cognitive, Affective and Behavioral Neuroscience

Depth of Facial Expression Processing Depends on Stimulus Visibility: Behavioral and Electrophysiological Evidence of Priming Effects

Academic journal article Cognitive, Affective and Behavioral Neuroscience

Depth of Facial Expression Processing Depends on Stimulus Visibility: Behavioral and Electrophysiological Evidence of Priming Effects

Article excerpt

Participants performed a priming task during which emotional faces served as prime stimuli and emotional words served as targets. Prime-target pairs were congruent or incongruent, and two levels of prime visibility were obtained by varying the duration of the masked primes. To probe a neural signature of the impact of the masked primes, lateralized readiness potentials (LRPs) were recorded over motor cortex. In the high-visibility condition, responses to word targets were faster when the prime-target pairs were congruent than when they were incongruent, providing evidence of priming effects. In line with the behavioral results, the electrophysiological data showed that high-visibility face primes resulted in LRP differences between congruent and incongruent trials, suggesting that prime stimuli initiated motor preparation. Contrary to the above pattern, no evidence for reaction time or LRP differences was observed in the low-visibility condition, revealing that the depth of facial expression processing is dependent on stimulus visibility.

Facial expression serves as a major source of nonverbal information in social interactions (Adolphs, 2002). Consistent with such social significance, previous research has suggested that facial expression processing is privileged and may take place independent of awareness. For example, it has been reported that reaction to a target is influenced by a facial expression prime stimulus preceding it, even when the prime was briefly presented and masked so that participants were unaware of its occurrence (Murphy & Zajonc, 1993; Neidenthal, 1990). In line with these behavioral results, neuroimaging studies have reported that the amygdala, a subcortical brain region believed to be important for emotional processing, evokes responses to emotion-laden faces when they are presented in backward masking conditions (Morris, Öhman, & Dolan, 1998; Whalen et al., 2004; Whalen et al., 1998). In addition, event-related potential (ERP) studies have reported that unconsciously processed facial expressions may exhibit differential Pl (90-140 msec) and N2 (150-200 msec) component responses, suggesting that unaware facial information can be registered at early visual processing stages (Kiss & Eimer, 2008; Li, Zinbarg, Boehm, & Paller, 2008; Liddell, Williams, Rathjen, Shevrin, & Gordon, 2004; Williams et al., 2004). Despite evidence supporting unaware perception of facial expressions, contrasting results have been presented recently. These studies have reported differential amygdala responses to fearful faces in aware conditions, but not in unaware conditions (Pessoa, Japee, Sturman, & Ungerleider, 2006; Phillips et al., 2004), and have suggested that many participants are able to detect briefly presented and masked face stimuli (Pessoa, 2005; Szczepanowski & Pessoa, 2007).

Although a vast literature has addressed the challenging question of visual awareness, relatively little is known about how different levels of stimulus visibility affect behavior and the brain, especially in the domain of facial expression processing (Wiens & Ohman, 2007). Indeed, much of the work on visual awareness has focused on establishing the types of processing that do or do not occur in unaware conditions. This emphasis is present, we believe, given the overall tendency of researchers to view awareness as a binary phenomenon-for instance, a participant is either aware or unaware of a stimulus. Here, we reasoned that if perception is not all or none, and instead is more gradual (Cleeremans & Dienes, 2008), facial expression processing should vary as a function of stimulus visibility.

Studies supporting the notion that unaware processes may involve higher levels (e.g., semantic) of processing have capitalized on the congruency priming paradigm, in which priming effects are obtained when prime and target stimuli are congruent (Dehaene et al., 1998; Greenwald, Draine, & Abrams, 1996; Greenwald, Klinger, & Liu, 1989). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.