The McGurk effect
The experiment aimed to discover if real words actually could be used to cause a McGurk effect. Easton and Basala (1982) claimed the McGurk effect probably wouldn’t materialise using recognised words instead of motherese. Pisoni and Luce (1986) concluded that the McGurk effect was due to a simple bias effect. The experimental design was a repeated-measure and used a voluntary sample of 12 participants. Audio and audio-visual plus audio stimuli were both presented 66 times using a tape recording alone or partnered with a person miming a word for the two conditions. These conditions were counterbalanced. The phrase presented was either recognised as oh dear or oh beer. There were eleven variations of the phrase with each only being presented six times. These variations were produced through mixing beer and dear using a wavelength editor. Before the tape was made all the mixed and not mixed phrases were randomised to guard against order effects. It was believed that audio-visual stimulus had a strong influence on speech perception. This hypothesis was statistically supported by p. < 0.001. Both the theories mentioned above were supported through analysis of the data leading to the belief that the McGurk effe
For the audio-visual condition the same pre-recorded tape as described above was used but Dr. Watkins also synchronized the miming of gear over the dear:beer part of the phrase presented. Exact synchronising of gear over the audio stimulus was not required as the McGurk effect has been shown to not be affected by this time lag (Repp et al, 1983). A repeated-measures design was used for testing the hypothesis. Half of the complete sample was subjected to the audio condition first. These participants turned around in their chairs whilst the remaining participants were presented with the audio-visual stimuli. This was done as to balance order effects. Another benefit of this was that all participants could be tested at the same time. This process was then reversed so that all the participants were tested in both of the conditions. In each condition the stimuli was presented 66 times. The order of the 66 digital mixes of the combined words was randomized in attempt to eliminate order effects. A signal to noise ratio of 45dB was also employed so that background noise did not affect the speech perception. The group data was collected from a self-selected sample of 12 undergraduates from a much larger pool of participants. The independent variable is the nature of the stimulus presented whilst the dependent variable is the number of D responses recorded.
Some topics in this essay:
Grant Braida,
Method Participants,
Introduction McGurk,
Results Table,
Dr Watkins,
Easton Basala,
Pisoni Luce,
Results Wilcoxon,
Fowler Dekle,
mcgurk effect,
speech perception,
,
lip reading,
et al,
audio-visual condition,
audio condition,
dear beer,
dr watkins,
strong influence,
influence speech perception,
visual information,
raw data workings,
strong influence speech,
easton basala 1982,
words correctly identified,
Join now to see the rest of the essay!
Approximate Word count = 3109
Approximate Pages = 12 (250 words per page double spaced)
CUSTOMER SERVICES
| |
|
Example Essays Home
