Cross-modal associations of human body odour attractiveness with facial and vocal attractiveness provide little support for the backup signals hypothesis: A systematic review and meta-analysis

Potential mating partners are typically assessed for their attractiveness by integrating multiple sensory modalities, including olfactory, visual, and auditory cues. There have been diverging predictions on how the individual modalities should relate to each other. According to the backup signals hypothesis, the assessment of multimodal cues provides redundant information, whereas the multiple messages hypothesis suggests that different modalities provide independent and non-redundant information about an individual’s mating-related quality. The backup signals hypothesis predicts a positive association between assessments of different modalities, whereas no substantial correlation across modalities is expected under the ‘multiple messages‘ hypothesis. Previous studies testing the two hypotheses have provided mixed results, but a systematic evaluation is currently missing. We performed a systematic review and a meta-analysis of published and unpublished studies to examine the congruence in assessments between body odour and facial attractiveness, and body odour and vocal attractiveness, in humans. We found positive but weak associations between attractiveness ratings of body odours and faces ( r = 0.1, k = 25), and between body odours and voices ( r = 0.1, k = 9). No sex differences were observed in the magnitudes of effects. Our results suggest that assessments of human body odour and facial and vocal attractiveness may provide independent and non-redundant information about mating-related qualities. The findings of the resent study provide little support for the backup signals hypothesis and may be better explained by the multiple messages hypothesis