Poster #42 - Stable Individual Differences in Infants’ Detection of Violations and their Search for Explanations

• In Event: 2-055 - Poster Session 06
  In Poster Session: PS 06 Section - Cognitive Processes

Integrative Statement

Infants detect impossible events, looking longer at scenarios that violate adults’ expectations.  Furthermore, these surprising events motivate infants to preferentially explore objects that violated their expectations (Stahl & Feigenson, 2015).  Yet, no research has revealed why infants explore following surprising events, nor whether there are individual differences in their ability to detect surprising events or in their use of surprise to motivate exploration. Here, we first tested the possibility that infants’ exploration reflects a search for explanations of surprising events. We hypothesized that if infants are searching for explanations, then giving them explanations for surprising events should change their subsequent exploration. Next, we used a longitudinal approach to ask whether surprise systematically affects individual infants differently.

First, we showed 25 12-month-old infants either a surprising or expected event involving object solidity. Infants saw a ball pass through or become stopped by a solid wall, and their looking was measured. Next the experimenter turned the wall, revealing a large opening at its base—this offered infants who had seen the solidity violation a plausible explanation of the event. Infants’ looking at this “explanation” was measured. Finally, infants got the opportunity to explore the familiar ball versus a novel object. We found that infants looked longer at the surprising solidity violation than the expected non-violation, t (23) = 2.66, p = .007. We also found that the longer infants who had seen the surprising violation looked at the “explanation” (the wall revealed to have a hole in it), the more they preferred to explore the novel object, r (11) = .62, p = .02; infants who had been surprised but looked little at the “explanation” preferred to explore the ball that had behaved surprisingly, as in previous work. This suggests that infants’ surprise-driven exploration reflects a search for an explanation: when an explanation is presented, infants stop exploring.

Next, to measure individual differences in the ability to detect surprising events and to identify explanations for those events, we re-tested infants 6 months later in a different event involving a support violation (an object appeared to float). As before, infants were then provided an “explanation” for the surprising event (a hidden source of support was revealed). We found that infants who better detected the solidity violation at 12 months also better detected the support violation at 18 months, r = .59. Moreover, infants who showed more interest in the “explanation” at 12 months also showed more interest in the “explanation” at 18 months, r = .65 (Figure 1). Moreover, these two processing measures dissociated: violation detection was uncorrelated with explanation recognition at either age: r = .14, and, r = .13 (Figure 2).

These findings suggest that surprise motivates infants to seek explanations for anomalous events. Strikingly, stable individual differences in infants’ looking at violations versus explanations suggest that there are distinct processing stages involved in surprise-driven learning. Our ongoing work aims to better understand the longer-term developmental trajectory of these individual differences and the way their underlying component processes support learning.

Authors

• Jasmin Perez, Johns Hopkins University
  Presenting Author

• Lisa Feigenson, Johns Hopkins University
  Non-Presenting Author