Infants in the First Year of Life Attribute Inference-Based Knowledge to Others

• In Event: Reasoning About Epistemic Processes: Children’s Understanding of how Knowledge is Acquired, Transmitted and Socially Constructed

Abstract

Adults interpret others’ behaviors with the assumption that people act consistently with the information available to them. This information may come from different sources, such as direct perception or inference. Thus, when a large toy is hidden in one of two boxes, one small and another large, we expect a person who saw the toy being hidden to go to the correct box when retrieving the toy (direct perception); we also expect a person who did not see the toy being hidden to go to the large box because the other box would be too small (inference). Research have shown that young infants track what information is available to others through direct perception. However, it was unclear whether infants track information that is available through inference. In two violation-of-expectation experiments, we examined whether infants would attribute inference-based knowledge to others by bringing to bear relevant general knowledge about the physical world. In Experiment1 (N=96), we built on the results suggesting that infants know that a wide object cannot be hidden inside a narrow container by 4-months, and showed 5-month-olds an experimenter who was searching for a wide toy that was hidden when she was absent in one of two locations: a narrow or a wide container. We found that infants expected the experimenter to search in the wide container. Three other conditions bolstered our findings. First, we showed that our results were not attributable to infants’ baseline expectations that the experimenter would always reach for the wide over the narrow container: Infants no longer expected the experimenter to search in the wide container when the toy was shown to be broken, and presumably no longer desirable. When both containers were equally wide, infants only expected the experimenter to search in the correct container when she saw the toy being hidden; and held no specific expectations when the experimenter did not see the toy being hidden.
In Experiment2 (N=60), we replicated and extended our results by examining the physical variable of height. By 7.5-months, infants understood that a tall object cannot be fully hidden in a short container. We showed infants an experimenter (E1) who wanted to find a tall toy, in either a tall box or a short box. When E1 was absent, a second experimenter compressed the toy to make it shorter and hid it in the short box. When E1 came back, infants expected her to have a false belief and search for the toy in the tall box; infants looked longer when she searched in the short box. However, if E1 saw the toy was being compressed, but not the hiding event, infants did not find it unexpected when E1 searched in either the tall or the short box. Together, these results suggest that even in the first year of life, infants could reason about inference as a source of knowledge for others. Our results also provide support for the notion that infants’ early psychological-reasoning system could integrate information from other cognitive processes, such as physical reasoning.

Authors

• Fransisca Ting, University of Toronto
  Presenting Author

• Lin Bian, University of Chicago

• Renee Baillargeon, University of Illinois at Urbana-Champaign