| |
Abstract:
Functional magnetic resonance imaging was used to investigate
functional plasticity related to phonetic learning. Nine
monolingual English-speaking subjects were scanned while performing
an identification task both before and after a 2-week period of
phonetic training with a Hindi dental-retroflex non-native phonetic
contrast. During scaning, a native phonetic identification task was
used as a control. Behaviorally, subjects could not distinguish the
non-native contrast prior to training, but improved significantly
after training. Brain areas involved during theidentification of
the native contrast versus an acoustically matched noise baseline
included bilateral superior temporal gyrus and frontal opercular
regions. The pre-training non-native versus noise comparison
yielded only a left frontal opercular peak. After training, the
same comparison yielded results similar to those found for the
native versus noise comparison. Correlational analyses between
behavioral improvement and the BOLD signal obtained during the
post-training non-native scan revealed a positive relationship
between performance and signal in the left angular gyrus and
anterior left middle temporal gyrus, as well as a concurrent
negative relationship with activity in the posterior left middle
temporal gyrus and bilaterally in frontal opercular regions. We
suggest that neural substrates for processing newly learned
non-native phonetic contrasts are similar to those used for native
contrasts, but the degree of success in learning is accompanied by
more efficient neural processing in classical frontal speech
regions, while making greater processing demands in left
parieto-temporal regions.
|
