Academic journal article Cognitive, Affective and Behavioral Neuroscience

Eye-Movement Assessment of the Time Course in Facial Expression Recognition: Neurophysiological Implications

Academic journal article Cognitive, Affective and Behavioral Neuroscience

Eye-Movement Assessment of the Time Course in Facial Expression Recognition: Neurophysiological Implications

Article excerpt

Happy, surprised, disgusted, angry, sad, fearful, and neutral faces were presented extrafoveally, with fixations on faces allowed or not. The faces were preceded by a cue word that designated the face to be saccaded in a two-alternative forced-choice discrimination task (2AFC; Experiments 1 and 2), or were followed by a probe word for recognition (Experiment 3). Eye tracking was used to decompose the recognition process into stages. Relative to the other expressions, happy faces (1) were identified faster (as early as 160 msec from stimulus onset) in extrafoveal vision, as revealed by shorter saccade latencies in the 2AFC task; (2) required less encoding effort, as indexed by shorter first fixations and dwell times; and (3) required less decision-making effort, as indicated by fewer refixations on the face after the recognition probe was presented. This reveals a happy-face identification advantage both prior to and during overt attentional processing. The results are discussed in relation to prior neurophysiological findings on latencies in facial expression recognition.

The processing of emotional facial expressions has attracted considerable behavioral and neurophysiological research (see reviews in Calder & Young, 2005; Eimer & Holmes, 2007; Frischen, Eastwood, & Smilek, 2008; Palermo & Rhodes, 2007; and Vuilleumier & Pourtois, 2007). Recognition times vary for the six basic emotional expressions (fear, anger, disgust, sadness, surprise, and happiness; see Calvo & Lundqvist, 2008; Palermo & Coltheart, 2004). The present study investigates the time course of such processing differences, with special interest in the earliest and the typical latency of conscious recognition. By using an eye-movement methodology that provides temporally precise measures at consecutive perceptual and cognitive stages, we relate behavioral and neurophysiological research on facial emotion recognition and show that happy faces are processed more efficiently than are other expressions across various stages.

An Advantage in the Recognition of Happy Faces

In behavioral studies using recognition and categorization tasks, happy facial expressions have been found to be identified faster and more accurately than other expressions. This "happy-face advantage" has been observed for separate comparisons of happiness and sadness (Kirita & Endo, 1995), happiness and disgust (Leppänen & Hietanen, 2004), happiness and anger (Juth, Lundqvist, Karlsson, & Öhman, 2005, Experiment 4; Leppänen, Tenhunen, & Hietanen, 2003), and happiness and fear (Juth et al., 2005). Furthermore, in two studies (Calvo & Lundqvist, 2008; Palermo & Coltheart, 2004), the recognition of all six basic emotional facial expressions was compared. An equivalent pattern of findings appeared in both studies, with recognition performance being fastest and most accurate for happy faces. The fact that different facial stimulus sets (Karolinska Directed Emotional Faces [KDEF; Lundqvist, Flykt, & Öhman, 1998]-Calvo & Lundqvist, 2008; vs. Pictures of Facial Affect [PFA; Ekman & Friesen, 1976] and other stimulus databases-Palermo & Coltheart, 2004) and different response systems (manual: Calvo & Lundqvist, 2008, vs. verbal: Palermo & Coltheart, 2004) were used in each study shows that the happy-face advantage is a robust and generalizable finding.1

Furthermore, Calvo and Lundqvist (2008) found that less visual information is required for recognition of happy expressions than of others, since the happy-face advantage became even greater when stimulus display duration was reduced: Only minimal impairment in the recognition of happy expressions was observed as display duration decreased from unlimited time to 500, 250, 100, 50, and 25 msec, whereas recognition of all the other emotional expressions decreased almost linearly as a function of display duration. Convergent evidence shows lower identification thresholds for happy expressions than for angry ones (Esteves & Öhman, 1993), and indicates that happy faces are less effectively masked than are angry faces (Maxwell & Davidson, 2004). …

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.