Do Monkeys and Young Children Understand Exclusive “Or” Relations? A Commentary on Ferrigno et al. (2021)

Abstract

Ferrigno et al. (2021) claim to provide evidence that monkeys can reason through the disjunctive syllogism (given A or B, not A, therefore B) and conclude that monkeys therefore understand logical “or” relations. Yet their data fail to provide evidence that the baboons they tested understood the exclusive “or” relations in the experimental task. For two mutually exclusive possibilities—A or B—the monkeys appeared to infer that B was true when A was shown to be false, but they failed to infer that B was false when A was shown to be true. In our own research, we recently found an identical response pattern in 2.5- to 4-year-old children, whereas 5-year-olds demonstrated that they could make both inferences. The monkeys’ and younger children’s responses are instead consistent with an incorrect understanding of A and B as having an inclusive “or” relation. Only the older children provided compelling evidence of representing the exclusive “or” relation between A and B.

A dual-process account for diverse reasoning strategies underlying symmetries and asymmetries between inferences between disjunctions and conditionals

“…Finally, we compare human performances on disjunctive syllogisms and DCIs both of which are equivalent, and finds the difference that most participants accepted disjunctive syllogisms (Gautam, et., al, 2021;Manktelow, 1999;Mody & Carey, 2016;Newstead, et. al, 1984;Roberge, 1976a;1976b), but rejected valid inferences from disjunctions to equivalent conditionals (Espino & Byrne, 2013;Oberauer et al, 2011;Ormerod & Richardson, 2003;Richardson & Ormerod, 1997).…”

Thinking about possibilities: mechanisms, ontogeny, functions and phylogeny

“…While not directly addressed in this issue, studies have recently begun to examine whether non-human animals can similarly reason over mutually exclusive possibilities. Interpretations remain contentious [19,24,104,105], but the available evidence suggests that animals (including nonhuman primates) tend to perform better on 'epistemic uncertainty' tasks requiring them to consider two possible locations of a previously hidden reward [106][107][108][109] than on 'physical uncertainty' tasks requiring them to anticipate two possible future trajectories of a currently visible reward [17,105,110,111]. There are at least four plausible explanations for these conflicting findings.…”