Replication study and meta-analysis reveal that non-attended angry voices do not activate the amygdala, suggesting automatic threat detection models require revision.
Twenty years ago, an influential neuroimaging study showed that the auditory cortex and amygdala are active during the presentation of non-attended angry voices. These results had a large scientific impact and inspired subsequent clinical work but have not been replicated by an independent laboratory. We conducted a high powered pre-registered close replication study (n= 37) of unattended vocal anger processing and extended the study by examining anger specificity of activations. Additionally, we performed an effect size meta-analysis of neural processing of angry voices (total n= 140). Neither the close replication nor the meta-analysis supported activation of the amygdala or auditory cortex during the presentation of non-attended angry voices, while the meta-analysis supported a role of the auditory cortex in attention-dependent processing of anger. Together, the results do not support the existence of circuitry that is automatically activated to detect vocal threat, in contrast to evolutionary inspired models of automatic threat perception.
No takes yet. Share an insight, caveat, or question.
Witteman et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: