Presentation: Sally
Summary: Yu-chiao
Studies have shown that
both long-term native language experience and immediate linguistic expectation
may affect listeners’ use of acoustic information when making a phonetic decision.
In order to investigate how segments and tones are processed in different
ambient language conditions by listeners of different native languages, this
study utilized a Garner selective attention task to measure the response time of
discriminating different sounds. A total of 57 listeners aged form 18 to 30
years old were divided into three groups, including 20 Chinese listeners
completing the experiment in Mandarin Chinese (CC group), another 17 Chinese
listeners doing in English (CE group), and 20 American English listeners doing
in English (EE group). Identical stimuli recorded in Chinese were used for all
groups of participants. They were asked to distinguish groups of sounds differed
from one to three dimensions, which were tonal, vocal, and consonantal
dimensions. There were only two variants for each dimension, in which all
variants were linguistic meaningful for both English and Mandarin Chinese
(i.e., tone 2 vs. tone 4 or question vs. statement for tonal dimension, [i] vs.
[ai] for vocal dimension, and [ph] vs. [th] for
consonantal dimension). When a certain dimension was set as the target dimension,
the other two dimensions functioned as competitive interfering dimensions. Listeners
were asked to classify the difference in the target dimension regardless of the
presence or absence of the other two dimensions. If the stimuli only differed
in the target dimension, it was called the baseline condition; otherwise, it
was called the orthogonal conditions. RT results of the baseline condition showed
that vowels were significantly more discriminable than consonants or tones for the
CE and EE groups. Owing to such difference, vowels were excluded from the following
analyses. Further comparison between the baseline and orthogonal conditions on
consonantal and tonal dimensions indicated that both the CC and CE groups had
an interference effect, while such interference was not found for the EE group.
This suggested that English listeners were able to process the difference in
the two dimensions separately without being slowed down by the interference
from the other dimension. However, Chinese listeners seemed to perform symmetric
dimensional integrality between consonant and tone, and did not present
significant difference among different ambient languages.
