Kaoru Sekiyama and Yoh'ichi Tohkura
MCGURK EFFECT UNDER CONDITIONS
WITH OR WITHOUT NOISE:
No Visual Biasing to Hearing Japanese
Syllables of High Auditory Intelligibility
Abstract:The McGurk effect shows a visual influence on auditory perception under
audio-visual discrepancy conditions. Dubbed video tapes are used to
demonstrate the McGurk effect.
The main purpose of this study was thus to examine the relation between
auditory intelligibility and the McGurk effect. Our hypothesis was that the
McGurk effect occurs only when auditory stimuli do not have complete
intelligibility. In fact, the literature has suggested this point although it has not
been stated explicitly. This is perhaps due to lack of systematic experimentation:
in previous studies, the number of the stimuli was small or the auditory
intelligibility itself was not measured. To examine this relationship, auditory
intelligibility is defined for this paper as the percent of correctly heard responses
when auditory-alone stimuli are presented.
Our question arose: Are noisy or degraded auditory stimuli necessary to the
McGurk effect? More moderately, do degraded auditory stimuli promote the
McGurk effect? To examine this question, the present experiment compared the
McGurk effect in two experimental conditions: a noise-free condition and a
noise-additive condition. Auditory intelligibility was measured for both the
noise-free and the noise-additive conditions, presenting auditory-alone stimuli.
We were thus able to examine the following points: (1) How does the McGurk
effect depend on the auditory intelligibility? (2) What is the nature of the
McGurk effect for Japanese syllables?