CONICET | Buscador de Institutos y Recursos Humanos

Minervino, Oberholzer and Trench (2013) demonstrated that people tend to judge that two situations are analogous whenever they could be assigned to the same schema-governed category (SGC). For example, Peter dried the mugs could be considered analogous to Peter decorated his room by virtue of being two instances of housework, despite the fact that corresponding relations and objects are not similar when compared in isolation. In this type of analogies, the base-target differences that matter when assessing analogical relatedness are those related to the values of the compared facts along critical dimensions of the common SGC, and not differences between matched elements considered in an isolated fashion (Tavernini, Trench, Olguín & Minervino, in press). With regards to analogical inference, recent results in the same line (Minervino, Margni & Trench, in press) indicate that analogical inferences tend to be exemplars of the SGC that applies to the base information to be projected, and that these inferences rarely include relations similar to those that take part in the information to be transferred. For example, if Peter received a reward for his first housework, he would receive another reward for his second collaboration. But what criteria do people follow in selecting the appropriate exemplars? While several theorists have postulated that some sort of difference-detection during mapping is required to prevent the generation of senseless and irrelevant inferences (e.g., Holyoak & Thagard, 1995; Gentner & Smith, 2013), such phenomenon has not been empirically studied. The present experiment was aimed at determining whether the detection of base-target differences along dimensions of the SGC to which the analogous causes belong determine the values of the inferred effects in dimensions of the SGC of which these inferences are cases (henceforth, see Figure 1). Returning to the previous example, imagine that the base action of drying the mugs led Peter´s parents to buy him a laptop. If people perceive that decorating a room requires more effort than drying the mugs,they would probably anticipate that the second reward would also increase in importance (e.g., his parents will take him to Disneyworld). To assess this possibility, 40 participants were confronted with analogical comparisons whose base analogs comprised a cause that engenders an effect, and whose target analogs consisted of a target cause that belonged to the same SGC as the base cause. After reading the target of each set, participants were asked to postulate a consequence of the target cause that they deemed analogous to that of the base effect.Figure 1. Schema of the experiment, illustrated with a set of the employed materials. Note: SGC: schema-governed category.The target cause was the same in the two conditions of the experiment. The critical manipulation concerned the relative values of base and target causes along a critical dimension of the SGC to which they belonged: While in the increasing condition the base cause was chosen to exhibit lower values than the target cause (as in the presented example), in the decreasing condition it was chosen to display higher values along such dimension (in this set, the base cause was Peter built a barbecue area). Two judges were handed the target analogs together with the SGC that was applicable to the base effect (e.g., reward), and were asked to use a 5-point Likert scale to rate the target effects proposed by participants in terms of their value along a central dimension of the SGC (e.g., the importance of the reward). As predicted, target increments in values along dimensions of their corresponding SGC led to inferences displaying higher scores in dimensions of their SGC as compared to inferences from the decreasing condition. The implications of the results for theories of analogical mapping and inference generation are discussed.