Poster #13 - Backward causal reasoning about successes and failures in 4- to 6-year-olds

• In Event: Poster Session 03

Abstract

Figuring out what went wrong when something does not work is the first step to fixing a problem. To do this, one has to reason backward from an observed outcome and diagnose the relevant causal factor(s) that may have resulted in a success or a failure. We asked whether young children can flexibly use learned causal knowledge to reason backward about the hidden cause of a success or failure in a “blicket-detector”-like task.
In Experiment 1 (Figure 1), 4- to 6-year-olds (N=44, M=5.86 years) were introduced to a ‘music machine’ and were asked to find ‘the right number of things’ that can make the machine play music. In the Exploration phase, children saw four sets containing one, two, three, or four objects, and were encouraged to try to find the set that, when placed on the machine, would activate it (only the set of three objects would activate the machine). Children completed four total Exploration trials, in which they could try as many sets as they wanted. The sets in each Exploration trial were made up of different objects on each trial to encourage children to attend to number. In each of the four Test trials, children were shown one visible object and a “treasure chest” and were told that they needed to figure out what was hidden inside the treasure chest. In Go trials, the visible object and treasure chest were placed on the machine and the machine activated. In No-Go trials, the visible object and treasure chest failed to activate the machine. Children then selected how many objects were inside the treasure chest from two alternative choices (one or two objects). We found that children succeeded in Go trials (65%, Wilcoxon p=.007) trials but failed in No-go (47%, p=.47) trials (Go vs. No-Go p=.004). Children’s performance was correlated with age in No-Go trials (p=.03), but not in Go trials (p=.28).
Experiment 2 (Figure 2) was a conceptual replication of Experiment 1. 4- to 6-year-olds (planned sample size N=44, current N=40, M=5.4 years) learned that ‘only two certain colors’ of things can make the machine play music (e.g. green and red together, but not green and yellow together). In Test trials, children were shown one object (e.g. a red object) and a treasure chest which together activated (Go trials) or did not activate (No Go trials) the machine, and were asked to select the color of the object in the chest from two options (e.g. green or yellow). Preliminary findings suggest children’s performance in Go (61%, p=.11) and No-go trials (63%, p=.05) were similar and were not related with age (ps>.25). Experiments 1 and 2 will be compared once the full sample is collected.
Together, our results suggest developing competence in backward causal reasoning about successes and failures in 4-6-year-olds, and may suggest that reasoning about a hidden cause that worked may be easier than reasoning about a hidden cause that did not work, at least in some contexts.

Authors

• Chen Cheng, Hong Kong University of Science and Technology
  Presenting Author

• Emilia Boak, University of California - Santa Cruz

• Melissa M. Kibbe, Boston University