Proceedings of the Annual Meeting of the Cognitive Science Society

Traditional laboratory studies have found that people are more likely to retrieve surface matches than distant analogs, suggesting that superficial similarities exert a stronger influence than structural similarities on retrieval. However, it has been contended that the observed supremacy of surface similarity may have originated in experimental conditions that are unfairly adverse for the retrieval of distant analogs, as well as in a faulty separation between surface and structural similarity during the construction of surface matches. In two experiments, we presented a target item that maintained only superficial similarities with one extra-experimental source and only structural similarities with another one. By using natural items, we were able to avoid the shallow processing often attributed to experimental analogs, while carefully controlling that surface matches did not maintain structural similarities. Converging with traditional results, our data showed a more frequent retrieval of surface matches than of distant analogs, indicating a supremacy of superficial similarities during retrieval.