Activation is not always inference: Word-based priming in spontaneous trait inferences

Abstract

People infer, without any intention or awareness, personality traits about actors enacting diagnostic behaviors. This phenomenon is known as spontaneous trait inferences (STIs). The activation of a trait is considered to be a true inference when it results from processing the meaning of the whole behavioral description. However, a trait can also become activated due to intra-lexical associations with individual words in the description. Here, we suggest a method to distinguish the two sources of activation and explore the influence that word-based priming has on some of the most popular paradigms used to study STIs. Results show that in the probe recognition task, word-based priming plays a considerable role and can, in the absence of an appropriate control, mimic spontaneous trait inference occurrence. However, in the false recognition task and in the explicit trait judgment task, the role of this spurious activation is negligible and the real trait inference can be easily detected.