What sorts of reasoning modules have been provided by evolution?: Some experiments conducted among Tukano speakers in Brazilian AmazoÃnia concerning reasoning about conditional propositions and about conditional probabilities

We investigate two claims made by Cosmides and her associates about content-speci®c reasoning processes (e.g., Barkow, Cosmides, & Tooby, 1992; Brase, Cosmides, & Tooby, 1998; Cosmides, 1989; Cosmides & Tooby, 1992, 1994; Fiddick, Cosmides, & Tooby, 2000; Tooby & Cosmides, 1992). The ®rst is a bioevolutionary argument that the environmental pressures on our Pleistocene ancestors resulted in a content-speci®c reasoning module for identifying violators of social contracts, but not in any content-general modules such as a mental logic for conditionals, of the sort we have proposed (e.g., O'Brien, 2004; O'Brien, Roazzi, Athias, Dias, BrandaÄo, & Brooks, 2003; see Braine & O'Brien (1998) for the most complete presentation of mental-logic theory and evidence in its support). The second claim is that the same human bioevolutionary history has provided a module for representing and reasoning about frequencies of events, but not for the probabilities of single events.