The first phase of the experiment (expression or speech-reading) commenced with 15 practice trials in which feedback was given.
The data for the second task were entered into an ANOVA with the factors type of first task (speech-reading vs.
Experiment 1: Means and SDS of median reaction times (ms) on second (identification) task (N = 32) Identification Identification new old Mean SD Mean SD First task speech-reading Familiar 684 121 633 109 Unfamiliar 698 130 693 120 First task expression Familiar 687 101 639 78 Unfamiliar 747 159 741 128 There were no further significant effects or interactions, although the interaction between type of task and familiarity neared significance (F = 3.63, p = .067).
Experiment 1 replicated and extended Ellis et al.'s (1990) findings, using personally familiar faces and a speech-reading as well as an expression task in the first phase.
Natural, dynamic speech-reading can be affected by familiarity; in particular, it constrains susceptibility to audiovisual fusion illusions (Walker et al., 1995), so that familiar faces dubbed to unfamiliar voices show reduced susceptibility to such illusions.
The design of Expt 2 was similar to the speech-reading condition of Expt 1.
The tasks were identical to those described in Expt 1 for participants performing in the speech-reading condition of that study.
Experiment 2: Means of median reaction times i (and SDs) for correct decisions on second task (speech-reading) after an identification task (N = 16) New Old Familiar 854 (192) 859 (186) Unfamiliar 834 (236) 794 (185) Discussion
Experiments 1 and 2 explored, separately, effects of speech-reading as a first task on identity judgments and the opposite order of task performance.
This experiment used a between-subjects design, collapsing the speech-reading conditions of Expt 1 and Expt 2 and adding a further condition.
Identity decisions were speeded by prior exposure to a familiar face seen in an earlier speech-reading task.
In this experiment we also found priming for familiar faces when the first phase and the second phase of the study both were speech-reading tests.