Stereotype-based priming without stereotype activation: A tale of two priming tasks.

Dimitra Tsamadi,Marius Golubickis,Betül Sahin,C Neil Macrae,Linn M Persson,Siobhan Caughey,Johanna K Falbén

doi:10.1177/1747021820925396

Abstract

An extensive literature has demonstrated stereotype-based priming effects. What this work has only recently considered, however, is the extent to which priming is moderated by the adoption of different sequential-priming tasks and the attendant implications for theoretical treatments of person perception. In addition, the processes through which priming arises (i.e., stimulus and/or response biases) remain largely unspecified. Accordingly, here we explored the emergence and origin of stereotype-based priming using both semantic- and response-priming tasks. Corroborating previous research, a stereotype-based priming effect only emerged when a response-priming (vs. semantic-priming) task was used. A further hierarchical drift diffusion model analysis revealed that this effect was underpinned by differences in the evidential requirements of response generation (i.e., a response bias), such that less evidence was needed when generating stereotype-consistent compared with stereotype-inconsistent responses. Crucially, information uptake (i.e., stimulus bias, efficiency of target processing) was faster for stereotype-inconsistent than stereotype-consistent targets. This reveals that stereotype-based priming originated in a response bias rather than the automatic activation of stereotypes. The theoretical implications of these findings are considered.

Highlights

A fundamental supposition in the psychological literature is that stereotypes simplify thinking and doing
To replicate the established stereotype-based priming effect using a SCT, our sample size was based on the meta-analytic effect size reported by Kidder et al (2018), such that G*Power (d = .52, α = .05, power = 80%) indicated a requirement of 32 participants
To further explore the three-way interaction, separate 2 (Prime: female or male) × 2 (Target: feminine or masculine) multilevel analyses were conducted for each Task

Summary

Introduction

A fundamental supposition in the psychological literature is that stereotypes simplify thinking and doing. In a prominent article, Blair and Banaji (1996) demonstrated that forenames (e.g., Alice) were categorised faster according to sex (i.e., male or female) when they were preceded by gendercongruent (e.g., sensitive) compared with gender-incongruent (e.g., decisive) attributes Underlying effects of this kind is the assumption that, following presentation of a priming stimulus, activation spreads to semantically associated information in memory, thereby facilitating responses to related (vs unrelated) material (Collins & Loftus, 1975; Neely, 1991).. Prior to target presentation, exposure to the prime triggers the generation of a compatible or incompatible response, such that performance is enhanced when prime and target elicit congruent (vs incongruent) reactions In this way, priming can be driven by both spreading activation and/or response generation, a state of affairs that may explain why stereotype-based priming is more likely to emerge in response-priming than semantic-priming tasks. In the current investigation we sought to identify the pathway (or pathways) through which stereotype-based priming arises in response-priming tasks

Methods

Results

Conclusion