Abstract
A large body of literature reports a left visual field (LVF) bias in adult populations when viewing face stimuli. Adults tend to use the information from the right side of the owner's face (falling onto the LVF) to make decisions about identity (Gilbert & Bakan, 1973), gender, expression, attractiveness, age (Burt & Perrett, 1997) and lip-reading (Campbell, 1986). What remains unclear is how early this LVF bias emerges and whether it is exogenously driven by the stimulus or endogenously driven by the perceiver.
Experiment 1 examined potential visual field biases in infants between 6 and 9 months of age (M age = 239.60 days) with the use of an eye-tracking system. The faces in the 30 second video clips varied along two dimensions: familiarity (mother versus female stranger) and the presence or absence of mouth movement (talking versus not talking). Video presentation was counterbalanced for familiarity and talking. Strangers' faces were matched to the ethnic background of the mother.
Videos were coded off-line for fixations falling on the adult's face. Results revealed a visual preference for the right side of the owner's face corresponding to the LVF, F(1,15) = 8.38, p [[lt]]0.5. Additionally, infants looked more at the eyes during the no talking condition and at the mouth during the talking condition. There was no effect of familiarity.
Experiment 2 investigated the driving source of the LVF bias (stimulus versus perceiver). An additional group of 6- to 9-month-old infants were shown mirror reversals of the original video clips. Analyses revealed a significant decrease in the LVF bias. These findings suggest a stimulus driven component to the LVF bias in infants.
Overall, the LVF bias characteristic of adult face processing appears to emerge within the first year of life. Furthermore, these findings also lend support to an exogenous account of the LVF bias in infants.