Feature-based inhibition underlies the affective consequences of attention

Abstract

Rapid selection of a target in the presence of similar distractors can cause subsequent affective evaluation of a distractor to be more negative than that for the selected object. This distractor devaluation effect has previously been attributed to an association of attentional inhibition with the distractor's representation. Here, we investigated whether the associated inhibition leading to distractor devaluation is object based or feature based. Using colour-tinted face and building stimuli in a two-item simple visual search, followed by evaluation of face stimuli on a trustworthiness scale, we report that emotional evaluation was modified by prior attention whenever the search stimuli and the to-be-evaluated face shared the distractor feature, regardless of whether face identity seen in the two successive tasks matched or not. These data support the notion that inhibition can be feature-based and show that such inhibition can have emotional consequences.

Acknowledgements

This research is supported by Integrative Analysis of Brain and Behaviour (IABB) initiative grant BBS/B/16178 from the Biotechnology and Biological Sciences Research Council (BBSRC), UK. A portion of the data in this paper was previously presented at the Object Perception and Memory (OPAM) Conference, 10 November 2005, in Toronto, ON, Canada, and a summary of this talk was published in an OPAM report in Visual Cognition: Goolsby, B. A., Raymond, J. E., & Shapiro, K. (2005). Affective consequences of attentional inhibition of faces depend on selection task. Visual Cognition, 14, 108–111. We would like to thank Alan Kingstone, Derrick Watson, and two anonymous reviewers for their very helpful comments on an earlier version of this paper.

Notes

¹We submitted the mean trustworthiness ratings from both Experiments 1A and 2 to a mixed-effects ANOVA response with response feature compatibility (match or mismatch) as a within-subjects factor and group (gender, large colour difference, and small colour difference) as a between-subjects factor. The main effect of novel trial type was not significant (match = 2.89, mismatch = 2.87; F<1), nor was its interaction with group (F<1). This validated our procedure of assessing ratings of novel faces on match trials, for which attention to the target item is irrelevant.

²In this and all subsequent experiments, there was no difference in participants’ ratings of faces on the trustworthiness scale and on the reverse-coded untrustworthiness scale, Exp. 1A: t(15) = 1.58, p>.10; Exp. 1B: t(18) = 0.79, p>.10; Exp. 2 (large colour difference group): t(17) = 1.56, p>.10, so we collapsed this factor.

³One participant made errors for all trials in one cell of the mixed-effects analysis for the attention task, and this empty cell resulted in a df _error of 17 instead of the expected 18.

⁴This condition was run chronologically after we had established that the response scale valence had no effect.