Group formation influences on over-imitation
With this data, we investigated influences of group membership on (i) eliciting over-imitation after an inefficient demonstration (demonstration of 4 irrelevant actions and 1 relevant action) and on (ii) maintaining over-imitation after an efficient demonstration (demonstration of only the relevant action) were studied. Two of the irrelevant actions involved physical contact with the puzzle box and two did not. In a no-group condition, 5-to-6-year-old preschoolers (N = 28 per condition) first observed an experimenter performing an inefficient strategy to extract a token from a puzzle-box, then it was the child's turn. Second, they observed another experimenter extracting the token efficiently (condition: 1=no group). In three further conditions, we formed groups before the actual over-imitation task. These three conditions varied in the emphasis of artificial group membership (conditions: 2=minimum group manipulation, 3=medium group manipulation, 4=maximum group manipulation). In these group conditions, the preschoolers observed a model belonging to an artificial in-group using an inefficient strategy including irrelevant actions to retrieve a reward (Phase 1) and subsequently an artificial out-group model using only the relevant action. In a baseline condition (condition=0) children did not get any demonstration at all. This condition was conducted to control for performance of irrelevant actions due to curiosity. Sex of the models was counterbalanced (female=0, male=1), and sex-group influences were exploratorily investigated. Our dependent variable was the oi-score, which was a binary coding for each demonstrated irrelevant action (0=not performed, 1 = performed). We predicted, that children would over-imitate after the inefficient demonstration and that their over-imitation after the efficient demonstration of the out-group model would be dependent on the emphasis of the group membership: If the group feeling is strong children would perform irrelevant action even if they know that they are not necessary. In a further group condition, the inefficient strategy was presented by an artificial out-group model (Phase 1) and the efficient strategy by an artificial in-group model (condition: 5=maximum group - inefficient/out-group). Here we predicted, that children would not over-imitate in Phase 1 if the inefficient strategy was presented by an out-group model. Results revealed that even though children’s over-imitation was influenced by general aspects of the group formation process, group membership itself did not affect their tendency to over-imitate. Whether the children observed either strategy by an in-group or out-group model did not make a difference. The pairing of children’s and experimenter’s sex seems to influence children’s over-imitation. Further research is needed to investigate these effects.