The (reverse) Colavita effect in emotion recognition: Children and adults come to different conclusions when categorizing incongruent emotional audio-visual stimuli

Previous work has shown that different sensory channels are prioritized across the life course, with children preferentially responding to auditory information and adults to visual information. The aim of the current study was to investigate whether this effect extends to more socially meaningful stimuli, namely emotional stimuli. 98 children and 66 adults viewed congruent and incongruent dynamic emotional body and emotional voice stimuli pairs, then were asked how the person was feeling. We saw no age-related difference in accuracy ratings for congruent audio/visual stimuli. When presented with incongruent stimuli however (e.g. Happy body movements and Fearful vocalisations), children responded that the person was feeling the emotion that they heard (auditory dominance) significantly more than chance across all emotional combinations used. Adults responded either at chance between the auditory and visual emotions or reported that the person was feeling the emotion that they saw (visual dominance). This provides the first evidence to our knowledge of a developmental difference in the emotional interpretation of incongruent bi-modal stimuli, moving from an auditory dominance in childhood to a visual dominance in adulthood. The potential underlying mechanisms of this phenomenon are discussed as well as potential practical implications in education and neurodiversity.Keywords: Colavita; emotion recognition; development; multisensory