Repetition priming for face speech images: Speech-reading primes face identification

Does speech-reading (classifying a facial image in terms of speech-sound) proceed independently of other face-reading skills? Repetition priming of face images was used to address this question. Earlier studies established that identity-based decisions are sensitive to earlier exposure of images of a well-known person, but that face-reading tasks such as expression or age classification may not be. We extended the repetition priming paradigm to speech-reading and in three experiments established that (a) identity decisions for personally familiar faces were sensitive to a previous image of that person's face making a speech-sound (identity decisions were primed by speech-reading), (b) speech-sound ma:ching vas not faster for known faces seen earlier in a different task (speech-reading was not primed by identification). We also found some evidence that knowledge of familiar faces could interfere with classification of face images for speech (unfamiliar faces were speech read faster) and that repeating a speech-classification task could give rise to form rather than face-based priming. These findings are discussed in relation to the separability of speech-reading from other face processing tasks.