Contradiction and information-seeking behaviour in ten-month-old infants

• In Event: Poster Session 06
  In Poster Session: PS 06 Topic - Cognitive Processes

Abstract

We know infants find impossible events more interesting (Baillargeon 2001, Stahl & Feigenson 2015). What happens after they encounter these? One possibility is that events that are contradictory based on their expectations lead to an inferential dead end. The other possibility is that infants interpret a contradiction as a cue that insufficient information is available. Only the latter case should lead to extensive search behaviour.

To test this question, we have shown 10-month-olds a series of contradictory events in an eye-tracking paradigm. In familiarisation an agent was introduced that visited suitable and unsuitable hiding places (the former, but not the latter, would cover the agent entirely). The test introduces a case of unexpected disappearance: the agent hid behind a central occluder, the occluder was then raised, revealing the absence of the agent.

Experiment 1 (n = 24) had two within-participant test conditions: in one, infants detected the disappearance of the agent in a visual scene with other occluders that were suitable hiding places for the agent. In the other, the remaining occluders were unsuitable hiding places. Infants’ scanning path revealed that they explored the remaining occluders longer when these were suitable hiding places, indicating that the contradictory event prompted information seeking behaviour. This was reflected in the ratio of infant fixations on the raised occluder versus the hiding places -- infants looked more at those locations that were suitable hiding places.

Experiment 2 (n = 24) provided additional evidence helping infants to explain away the disappearance of the agent: it showed that the central occluder functioned as a lift that could whisk away the agent when it was raised. In this case infants did not scan suitable hiding places longer, indicating that they did not require further information to solve the puzzle. Unlike in Experiment 1, infants did not look more at suitable versus unsuitable hiding places . These results suggest that 10-month-olds make a distinction between grounded and ungrounded disappearance and regard the latter as a cue that they don't know enough about the environment.

Authors

• Peter Racz, Central European University
  Presenting Author

• Erno Teglas, Central European University
  Non-Presenting Author