Putting the pieces together: Revealing face–voice integration through the facial overshadowing effect

ABSTRACT

Two experiments are reported which explore the limits of the facial overshadowing effect on voice recognition. In Experiment 1, the facial overshadowing effect was replicated. Voice recognition was significantly impaired when the voice had been studied in the presence of the face rather than when studied alone. However, this effect was removed when the accompanying face was inverted. In Experiment 2, the facial overshadowing effect was replicated using more realistic moving faces rather than static faces through presentation of a video sequence at study. Surprisingly, the magnitude of the facial overshadowing effect was not influenced by whether faces were static or dynamic. Moreover, the effect remained even under temporal asynchrony. Together, these results show the facial overshadowing effect to be a robust phenomenon. However, it appears to depend critically on one factor: the facial nature of the distraction. The results are discussed in terms of the integration of faces and voices in a recognition context, and in terms of a holistic view of face–voice processing in a more general person perception framework. This offers an explanation of current findings, and serves to guide predictions for future work.

KEYWORDS:

• Facial overshadowing effect
• face–voice integration
• holistic person perception

Acknowledgements

The authors would like to thank Professor Andy Young and Dr Greg Neil for valuable conversations on early versions of the ideas contained here.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Sarah V. Stevenage http://orcid.org/0000-0003-4155-2939

Notes

1 An audio lead was preferred to an audio lag so that any decline in the facial overshadowing effect could not be attributable to mere habituation to a pre-exposed face, as shown by Cook and Wilding (2001).

2 One may consider that ceiling effects in accuracy and d’ may have compromised the ability to reveal greater overshadowing under asynchronous than synchronous conditions. However, when reaction time data were considered, there was no pattern of facial overshadowing at all. These data are not reported further given variance in the accuracy and d’ data.

Additional information

Funding

This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) [grant number EP/J004995/1] awarded to the second author.