Spontaneous Inferences from Pictorially Presented Behaviors

Abstract

Previous research on spontaneous trait inferences (STI) was based on verbal stimuli. In this research, stimulus behaviors were presented pictorially and STIs were measured in terms of the time needed to identify a trait term that gradually appeared behind a mask. An attempt was made to demonstrate that the STI cannot be reduced to a side effect of language comprehension. A pilot study showed that the phenomenon extends to pictures and that a graphical encoding task leads to even stronger STIs than verbal recoding. Experiment 1 corroborated the basic finding using an improved methodology. In Experiment 2, specific encoding operations were manipulated in a verification task. STIs were strongest when the verification task referred to concrete stimulus aspects. The findings support neither an account in terms of mere language comprehension nor a verbal interference of inferential-distance account, but they are consistent with a concreteness advantage or picture-superiority effect.