Title: In the Eyes of the Beholder: What Eye-Tracking Reveals About Analogy-Making Strategies in Children and Adults.
Abstract: In the Eyes of the Beholder: What Eye-Tracking Reveals About Analogy-Making Strategies in Children and Adults Jean-Pierre Thibaut, Robert French, Angelique Missault, Yannick Gerard, Yannick Glady {jean-pierre.thibaut, robert.french, yannick.gerard, yannick.glady}@u-bourgogne.fr; [email protected] LEAD-CNRS, UMR5022, University of Burgundy, Pole AAFE – Esplanade Erasme 21065 DIJON. FRANCE used geometrical shapes. In an A:B::C:D paradigm, children were influenced by the type and number of perceptual distractors. Thibaut, French, and Vezneva (2010b) studied the role of the semantic association strength between items making up the A-B and C-D pairs with 4- and 5-year-old children. They hypothesized that younger children, having more limited cognitive resources, would have more difficulty solving problems in which the A-B items and the C-Target pairs were weakly associated. In a classic A:B::C: ? paradigm with four possible responses, they compared weak and strong analogies (i.e., analogies in which the items of the A-B and C-D pairs were weakly, or strongly, associated) and manipulated the number of semantic distractors (1 or 3). Their results revealed a difference between weak and strong analogies, especially when the number of distractor items was high (i.e., three). This is compatible with the idea that a greater number of related distractors would be harder to inhibit (and thus, ignore) than a single semantic distractor. Interestingly, strong analogies were largely unaffected by the number of distractors, most likely because the relations between A- B and C-D item pairs were sufficiently strong that they were not interfered with by the semantic distractors. In contrast, when the problem involved weakly associated items, mapping the A-B pair onto the C-D pair requires more than simply accessing the obvious semantic dimensions of the items. For this reason, we consider analogy-making to be a search through a space of features and potential relations. The number of relations holding between any A-B pair is potentially large because, depending on the context, any number of different relations might be relevant (see Murphy and Medin, 1985; Chalmers, French & Hofstadter, 1992; Hofstadter et al., 1995; French, 1995; Mitchell, 1993; Thibaut, 1991; 1997). As mentioned above, the structure of the search space and the presence or absence of competing non-analogical solutions have an effect on the search, especially for young children, who have greater difficulty handling the cognitive load associated with a more elaborate search of the space of possible solutions. Abstract The present study uses eye-tracking technology to track differences in how children aged 5 and 8, and adults explore the space of possible answers to a semantic analogy problem. The main results were that adults looked more to A and B than to C and Target and that they start with A and B before looking at C and D. For children, the pattern was very different. They spent significantly more time than adults on C and the Target item (or distractors) and less on A and B. In addition, children start with an evenly distributed exploration of the stimuli before progressively converging on the C- Target relation. Keywords: Analogy-making, development, strategies, eye tracking executive functions. Introduction Extensive work suggests that analogy-making, in the sense of understanding and/or generating relations between objects or situations in the world, is a cognitive ability that develops only gradually (Gentner, 1988, Goswami, 1992). There are two main explanations of the development of the ability to make analogies. First, analogy-making can be explained in terms of the gradual increase of children’s structured knowledge of the world (Goswami & Brown, 1990; Vosniadou, 1995; see Thibaut, 1999, for a general overview of conceptual development). According to Goswami (1992, 2001), it is only the lack of conceptual knowledge in one of the domains involved in the analogy that prevents children from deriving the correct analogies. This view attaches little or no particular importance to processing constraints. An alternative explanation, however, is based on the development of children’s executive functions, and more particularly to their inhibition capacities and their cognitive flexibility. This explanation provides an explanation of observed analogy- making behavior for problems in which salient associations come immediately to mind, but are, in fact, irrelevant to the current analogy problem or when salient distractors are present in the solution set (Richland, Morrison, and Holyoak, 2006; Thibaut, French, & Vezneva, 2010a and b). In order to test the role of executive functions in analogy-making in children, Richland, Morrison, & Holyoak (2006) used scene analogy problems consisting of pairs of scenes illustrating relations between objects. When there were distractors perceptually similar to the focal item in the base scene, children made more errors than when the distractors were perceptually dissimilar from the focal item. Thibaut, French, Vezneva (2010a) Goals of the present paper The purpose of the present contribution is to study the development of analogy making with a combined set of measures – namely, the percentage of correct answers, the locus of the errors, and eye tracking measures – in a task in which we manipulate the number of distractors and the semantic strength in the A-B pairs and the C-D pairs (see Thibaut et al. 2010b). By means of an eye-tracker, we were able to record exactly where participants looked in their quest for a solution, which allowed us to develop a better idea of how a solution to a particular problem arose.
Publication Year: 2011
Publication Date: 2011-01-01
Language: en
Type: article
Access and Citation
Cited By Count: 20
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot