Poster #120 - What Can Eye Tracking Reveal About Children’s Event Processing as They Learn New Verbs?

• In Event: 2-163 - Poster Session 08
  In Poster Session: PS 08 Section - Language, Communication

Integrative Statement

To learn a verb, children must attend to objects and relations, often within a dynamic scene. Children can benefit from comparing events linked to a new verb (e.g., Childers, 2011; Scott & Fisher, 2012), but it is unclear how children perform these comparisons. This study asks whether children visually attend more to particular objects that vary across events as opposed to objects that stay the same. If they do, this suggests that they are comparing events to each other, noting which objects differ, and adjusting their visual attention appropriately, which could help them learn a verb.

Two ½-year-old (n=34) and 3½-year-old children (n=34) participated. Children sat in front of a monitor connected to a Tobii x3-120 eye tracker. An experimenter produced sentences containing a novel verb while children watched a video event (e.g., Look! She’s going to tam it.); this was repeated three times. During this learning phase, children saw 3 different tools, 3 different affected objects, or no variation while the eye tracker recorded fixations to specific objects (AOIs: tools, affected objects). At test, a split screen video was shown and children pointed to one of two new events in two test trials. This process was repeated until children learned four verbs (tam, jop, zim, and lorse).

If children can compare events to each other, they should notice objects that vary across trials and adjust their attention to them. To test this, we computed difference scores (Trial 3-Trial 1) to examine changes in visual fixations to a specific object during the learning phase. A repeated measures ANOVA with Age (2: 2½, 3½ years), Condition (3: Tool Change, AffectedObject Change, Control) as between subjects factors, and AOI type (tool, affected object) as within-subjects factor, dv=change in looking (Trial3-Trial1), revealed a main effect of AOItype, F(1,62)= 10.28,p=.002, and an AOItype x Condition interaction, F(2,62)=19.94,p<.001. Children who saw varied tools maintained their attention to tools and decreased their attention to the unchanging affected objects while children in other conditions showed a different pattern (see Fig.1). An ANCOVA analysis showed that, at test, adjusting visual attention to tools predicted successful verb extensions (Tool AOI, F(1,67)=6.31, p<.02) while looking to other AOIs did not. Additionally, there was a significant main effect of Age group favoring 3 ½-year-olds, F(1,67) = 18.10,p< .001, though children at both ages succeeded. In sum, children did attend to variation, particularly when the tool varied, and this attention was linked to success in verb learning.

Children can benefit from comparing events to each other during verb learning (e.g., Scott & Fisher, 2012). This study shows that children attend to objects differently depending on which object varies across examples, which suggests that they are actively adjusting their attention while learning verbs. This is consistent with structural alignment theory in which children may notice variation because they are aligning specific objects across events during verb learning (e.g., Gentner, 1989).

Authors

• Priscilla Tovar-Perez, Trinity University
  Presenting Author

• Blaire Porter, Trinity University
  Non-Presenting Author

• Sophia Arriazola, Trinity University
  Non-Presenting Author

• Jane Childers, Trinity University
  Non-Presenting Author