Top-down guidance in visual search for facial expressions

Hahn, Sowon; Gronlund, Scott D. · 2007 · OpenAlex-citations

DOI: 10.3758/bf03194044

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates how top-down search goals influence the well-documented attentional bias toward threatening facial expressions, specifically angry faces. While previous research established that angry faces capture attention more efficiently than happy or neutral faces—often attributed to automatic, subcortical processing pathways involving the amygdala—it remained unclear whether voluntary, goal-directed attention could modify this bias. The authors hypothesized that search efficiency depends on the combined influence of stimulus characteristics and top-down goals, rather than being solely driven by the automatic salience of threat. The researchers conducted two experiments using a visual search paradigm with schematic faces. In Experiment 1, 16 participants searched for a discrepant facial expression (either angry or happy) within a crowd of neutral faces without a specific target instruction. Results replicated the "angry face superiority effect," showing a significantly shallower response time (RT) slope for angry targets (37 msec) compared to happy targets (52 msec), indicating faster detection of angry faces when no specific goal was present. Experiment 2 introduced a top-down goal by instructing 11 participants to search specifically for either a happy or an angry face. The design included target-present trials, nontarget singleton-present trials (where the opposite emotion was present but not the target), and singleton-absent trials. When searching for a target, the angry face search remained more efficient than the happy face search (slopes of 28 msec vs. 44 msec). However, when an angry or happy face appeared as a nontarget singleton (opposing the goal), the search slopes were equivalent (59 msec vs. 57 msec). RT distribution analyses using the ex-Gaussian model further revealed that the special status of angry faces—characterized by constant leading-edge RTs indicative of parallel processing—only occurred when the angry face was the target. When the angry face was a nontarget distractor, its processing resembled that of happy faces. The findings suggest that the attentional advantage of angry faces is not purely automatic but is modulated by top-down goals. While angry faces guide attention effectively in the absence of a specific goal, their efficiency in guiding search is dependent on the alignment between the stimulus and the task goal. The mere presence of an angry face does not confer a search advantage if it opposes the current goal. This implies that visual search for facial expressions involves a joint influence of bottom-up stimulus priority and top-down goal-directed attention, challenging the view that threat detection is entirely automatic and independent of cognitive context.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-25; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• visual search