Flexible and inflexible task sets: Asymmetric interference when switching between emotional expression, sex, and age classification of perceived faces

Abstract

The present study investigated whether the processing characteristics of categorizing emotional facial expressions are different from those of categorizing facial age and sex information. Given that emotions change rapidly, it was hypothesized that processing facial expressions involves a more flexible task set that causes less between-task interference than the task sets involved in processing age or sex of a face. Participants switched between three tasks: categorizing a face as looking happy or angry (emotion task), young or old (age task), and male or female (sex task). Interference between tasks was measured by global interference and response interference. Both measures revealed patterns of asymmetric interference. Global between-task interference was reduced when a task was mixed with the emotion task. Response interference, as measured by congruency effects, was larger for the emotion task than for the nonemotional tasks. The results support the idea that processing emotional facial expression constitutes a more flexible task set that causes less interference (i.e., task-set “inertia”) than processing the age or sex of a face

Keywords:

• Asymmetric interference
• Task switching
• Task-set inertia
• Emotional expression

Acknowledgments

We thank Maria Gruno for preparing the stimulus material, Darja Rosocha for helping with data collection, and two anonymous reviewers for helpful comments on drafts of this article.

Notes

¹Note that the asymmetry between flexible and inflexible task sets results solely from a different amount of top-down control, with the flexible task being less strongly controlled in a top-down manner. The asymmetric interference does not result from different task strengths, as is the case when switching, for example, between colour naming and word reading in a Stroop paradigm. In Gilbert and Shallice's (2002) model, top-down control is represented by activation of the respective task-demand units, whereas task strength is represented by the strength of the stimulus–response associations for the different tasks.