Mammoth Cloning Reminds Us of "Jurassic Park" but Storm Replication Does Not: Naturalistic Settings Do Not Aid the Retrieval of Distant Analogs

After asking participants to propose analogies favoring a zerodeficit policy, Blanchette and Dunbar (2000) obtained a high proportion of base analogs lacking superficial similarities with the target, thus questioning the validity of a long experimental tradition demonstrating their centrality on retrieval. Through the use of culturally shared bases, we overcame two limitations in their study that preclude interpreting their results as evidence for superficially unconstrained retrieval: 1) a lack of discrimination between retrieved and invented bases, and 2) an assessment of the effect of superficial similarity based on counting superficially similar vs. superficially dissimilar bases, which disregards the number of available analogs of each kind. Our participants had to propose analogies that could be used to dissuade a person from pursuing certain objective. A movie seen in natural contexts could serve such a purpose. Whereas half of the participants had to retrieve it out of a superficially similar target, the other half had to retrieve it out of a superficially dissimilar target. In line with traditional findings, retrieval of superficially dissimilar sources was scarce and much lower than retrieval of sources maintaining such similarities. Results call into doubt the hypothesis that in natural settings analogical retrieval is less constrained by superficial similarity.