Poster #164 - Face detection in 2- to 6-month-old infants is influenced by gaze direction and species

• In Event: 1-169 - Poster Session 04
  In Poster Session: PS 04 Section - Perceptual, Sensory, Motor

Integrative Statement

Humans and other primates efficiently detect faces, even when presented in complex visual displays (Simpson et al., 2017). Face detection appears especially efficient when the faces are making eye-contact. When presented with two faces, infants look longer at faces with open eyes and direct gaze, compared to faces with closed or averted eyes (Batki et al., 2000; Farroni et al., 2002; Lasky & Klein, 1979). Infants’ attention is also captured more by human than other animal faces. This privileged processing of human stimuli is known as an own-species bias (OSB; Scott & Fava, 2013).
Here, we longitudinally examined infants’ attention to human and animal faces varying in head and eye orientation. We hypothesized that OSB would increase with age, and that infants would be most attentive to gaze cues displayed by human faces. We tracked infants’ attention with a Tobii TX300 eye-tracker at 2, 4, and 6 months, as they viewed up to 16 arrays, containing chimpanzee and human faces (Figure 1). We used multilevel modeling to examine the effects of head/eye orientation and species on infants’ proportion of time looking to faces (out of the total time looking to all images).
As infants aged, there was a .02 decrease in looking to faces (γ10=-.02, p<.001), suggesting younger infants spent more time looking at faces. There was a .03 decrease in looking to chimpanzee compared to human faces (γ20=-.03, p<.001), reflecting infants’ greater interest in human compared to animal faces. Infants had a .03 decrease in looking to faces with averted compared to direct eye gaze (γ40=-.03, p=.01), indicating more interest in faces with eye-contact. There were no effects of head direction or any of the covariates (ps>.05). Infants looked more to faces with congruent head and eye orientations (e.g., direct head with direct eyes) relative to faces with incongruent head and eye orientations (e.g., direct head with averted eyes; γ50=-.03, p=.01). There was an interaction among species, head, and eyes (Figure 2; γ80=-.05, p=.05) Infants looked the most to the human faces with averted heads and eyes relative to other face types.
We did not find support for our hypothesis that OSB would grow stronger with age; rather, we found both species and gaze orientation influenced infants’ detection already by 2 months. Head and eye gaze were stronger social cues when displayed in unison (direct head and eyes or averted head and eyes), and were particularly salient in the context of human faces. While previous studies reported OSB for face detection by 6 months (Jakobsen et al., 2016), we suggest OSB in detection may occur even earlier, much like face preference (Heron-Delaney et al., 2011). Surprisingly, we found that human faces with both averted head and eyes received the most attention. This pattern may reflect the early emergence of gaze following—the ability to look where another individual is looking—which appears around this age (Morales et al., 2000). Infants may be especially interested in averted gaze faces, which may provide early scaffolding for joint attention.

Authors

• Sarah Emily Maylott, University of Miami
  Presenting Author

• Samantha Mitsven, University of Miami
  Non-Presenting Author

• Krisztina Jakobsen, James Madison University
  Non-Presenting Author

• Elizabeth Ann Simpson, University of Miami
  Non-Presenting Author