Experience and Age Guide Infants’ Attention to Meaning in Scenes

• In Event: Poster Session 07
  In Poster Session: PS 07 Topic - Attention, Learning, Memory

Abstract

Infants acquire a significant amount of information through looking. However, effective learning from the visual world requires attending to informationally rich regions while inhibiting distracting or irrelevant information — an ability that undergoes rapid development across the first year of life (Colombo, 2001). Younger infants’ gaze is drawn to highly salient regions of scenes, whereas older infants’ gaze is drawn to socially significant regions, such as the locations of human faces (Frank et al, 2014; Kwon et al., 2016). Much less is known about the factors that contribute to where infants look at complex real-world scenes. We know that when viewing such scenes, adults’ gaze is better predicted by the distribution of semantic content or meaning, than by physical salience (Henderson & Hayes, 2017, 2018), but this question has not been addressed in infants. This was the primary goal of the present work.

We used an SMI-Red eye tracker to record eye gaze (at 120 Hz) for 89 healthy, full-term infants between 4 and 12 months (39 boys) as they viewed an average of 30 5-s trials of real-world scenes (range 9 to 48 scenes per infant). The scenes were taken from Henderson and Hayes (2017), and adult ratings provided information about the meaning of scene regions, resulting in a meaning map for each scene. Specifically, adults rated the relative informativeness of small patches in the scenes. For example, a patch of blue sky would be judged less meaningful than a patch containing a recognizable part of a car tire. Infants were drawn from a region including large urban areas, suburban neighborhoods, and small rural towns; the infants, therefore, had a wide range of experience related to the scenes used in our set. We, therefore, asked parents to report their infant’s familiarity with each of the scenes. We compared infants’ patterns of fixation with the meaning maps. As can be seen in Figures 1 and 2, infants were more likely to fixate regions with high meaning ratings than to fixate regions with low meaning ratings. In Figure 1, the difference between the three lines shows that older infants (darker line) were more likely to fixate high meaning regions and less likely to fixate low meaning regions compared to younger infants (lighter line). In Figure 2, the three lines now correspond to differences in familiarity with the type of scene, rather than infant age. Once again, there was a larger difference in their likelihood of fixating more and less meaningful regions with increased experience with that type of scene (darkest line). Together, these results suggest that as infants develop and encounter more of the world, they refine their attention toward adult-defined meaningful regions.

Authors

• Shannon Marie Klotz, University of California - Davis
  Presenting Author

• Taylor R Hayes, University of California, Davis
  Non-Presenting Author

• Katherine Pomaranski, University of California, Davis
  Non-Presenting Author

• John M Henderson, University of California, Davis
  Non-Presenting Author

• Lisa M Oakes, University of California - Davis
  Non-Presenting Author