The neural basis of intergroup threat effect on social attention

Abstract

Previous gaze-cuing studies found that intergroup threat is one of the modulators of gaze cuing. These findings indicate that intergroup threat would gate social attention by activating a network resembling that is thought to be involved in drawing or/and holding attention. The present study tested this hypothesis using a gaze-cuing task in which a particular in-group participants observed threatening out-group and nonthreatening out-group gazes, while undergoing functional magnetic resonance imaging. As expected, greater gaze cuing effect only emerged for threatening out-group when the in-group participants only felt inergroup threat from that out-group. Behaviorally, we found intergroup threatening out-group gazes did not draw attention faster than nonthreatening in-group gazes does. However, participants took more time to suppress the influence of the gaze direction of threatening out-group gazes, compared to nonthreatening in-group gazes, in the incongruent condition, which means intergroup threatening gaze holds attention longer than nonthreatening gaze does. Imaging results demonstrated that threatening cues recruited a fronto-parietal network, previously implicated in holding attention and execution functions. Our results, therefore, suggest that the mechanisms underpinning gaze cuing evolved to be sensitive to intergroup threatening stimuli, possibly because it is hard to disengage from such intergroup threatening cues once they are detected.

Figure 1. Mean reaction times (±SE) as a function of intergroup threat and cue congruency for 200-ms SOA in study a.

Figure 2. Mean reaction times (±SE) as a function of nonthreatening groups and cue congruency for 200-ms SOA in study b.

Figure 3. Mean reaction times (±SE) as a function of intergroup threat and cue congruency for 800-ms SOA in study a.

Figure 4. Mean reaction times (±SE) as a function of nonthreatening groups and cue congruency for 800-ms SOA in study b.

Figure 5. Maps showing brain areas in which activity was greater for nonthreatening in-group faces than for threatening out-group faces in study a (in the condition of congruent at 200 SOA) (FDR = 0.05, corrected).

Figure 6. Maps showing brain areas in which activity was greater for threatening out-group faces than for nonthreatening in-group faces in study a (in the condition of incongruent at 200 SOA) (FDR = 0.05, corrected).

Figure 7. Schematic illustration of gaze-cuing task.

The figure shows an example of a single trial. The direction of the male gaze was informative as to the location of the letter “L” target.