Abstract

What can differences tell us about the process of comparison? The structural alignment model posits that psychologically salient differences arise out of commonalities (Markman & Gentner, 1993) . This leads to the counterintuitive prediction that high similarity pairs (e .g ., bicycle/tricycle) should elicit more differences than low similarity pairs (e .g ., broom/ambulance), since high similarity pairs have more commonalities from which to derive differences, Gentner and Markman (1994) tested this prediction using a speeded difference task . Participants were asked to list one difference for as many word pairs (of high and low similarity) as possible in 5 minutes . As predicted, participants listed a difference for more high similarity pairs than low similarity pairs . The authors concluded that the high similarity pairs have a difference advantage because they have a larger common system from which to derive differences. However, another possible explanation is that many high similarity pairs (e .g ., hotel/motel) have been compared in the past, resulting in a stock of pre-stored differences . Experiment I addresses this alternative by directly testing whether producing commonalities facilitates listing differences . Forty-eight participants first listed a commonality for high and low similarity pairs . They were then given a speeded difference task, as described above . Half of the pairs were "old" (i .e ., used in the commonality task), and half were "new ." Differences were produced for more old pairs (M=5 .9) than new pairs (M=5 .4, p< .04), as predicted . This effect was obtained despite the presence of word pairs that were unlikely to have pre-stored differences (low similarity) . However, we still needed to rule out other explanationse.g .,, that the difference facilitation was a general result of recent co-activation of the two terms . In Experiment 2, half of the 48 participants performed a commonality task and the other half performed a thematic connection task on the same word pairs (which where chosen to facilitate either task) . For example, the pair "tree/child" could elicit the commonality "both grow" or the thematic connection "a child climbs a tree ." After this setting task, all participants were given the speeded difference task . Sample responses are presented in Table 1 . The results showed the predicted interaction (p= .004): for the Commonality group, differences were easier to list