Start Date

10-5-2017 11:00 AM

End Date

10-5-2017 1:00 PM

Subjects

English language -- Phonetics, Categorization (Psychology), Cognitive learning -- Effect of working memory on

Description

Abstract: We examined the role of procedural-, declarative-, and working-memory systems in adults learning novel sound categories. Adults have fully developed declarative-memory skills that sometimes inhibit their ability to learn implicitly/procedurally (Filoteo, Lauritzen, & Maddox, 2010). Models of impaired language like the Procedural Deficit Hypothesis suggest that procedural-memory deficits are predictive of language-learning outcomes (Lum, Conti-Ramsden, Page, & Ullman, 2011). This study tested the hypothesis that language structure is best learned implicitly/procedurally, which has implications for L2 learning and language impairment. The novel sound categories presented to participants varied along a phonologically non-native dimension, pitch, and a native dimension, vowel quality (F2). Optimal learning required integrating information from both cues. We predicted that participants who demonstrated strong procedural-memory skills would be most successful in learning novel sound categories. In Experiment 1, participants completed procedural-, declarative-, and working-memory assessments to relate individual differences to speech-sound learning. Experiment 2 participants received twice as much speech-sound training and more training at the category boundary. Procedural-memory in Experiment 1 predicted participants’ ability to learn category boundaries, but Experiment 2 did not replicate this effect. It is possible that more explicit training on stimuli along the category boundary in Experiment 2 favored participants with strong working-memory skills. It is also possible that the second day of training reduced effects of individual differences by boosting the performance of individuals with weaker memory skills. Further analysis into the dynamics and time-course of these memory systems in language-learning may reveal more nuanced contributions of procedural-memory in Experiment 2.

Authors: Ben Carlstrom, Chelsea McGrath, Carolyn Quam, Alisa Wang, Andrew Lotto

Presenter: Ben Carlstrom

Persistent Identifier

http://archives.pdx.edu/ds/psu/20072

Share

COinS
 
May 10th, 11:00 AM May 10th, 1:00 PM

Individual and Group Differences in Sound Category Learning

Abstract: We examined the role of procedural-, declarative-, and working-memory systems in adults learning novel sound categories. Adults have fully developed declarative-memory skills that sometimes inhibit their ability to learn implicitly/procedurally (Filoteo, Lauritzen, & Maddox, 2010). Models of impaired language like the Procedural Deficit Hypothesis suggest that procedural-memory deficits are predictive of language-learning outcomes (Lum, Conti-Ramsden, Page, & Ullman, 2011). This study tested the hypothesis that language structure is best learned implicitly/procedurally, which has implications for L2 learning and language impairment. The novel sound categories presented to participants varied along a phonologically non-native dimension, pitch, and a native dimension, vowel quality (F2). Optimal learning required integrating information from both cues. We predicted that participants who demonstrated strong procedural-memory skills would be most successful in learning novel sound categories. In Experiment 1, participants completed procedural-, declarative-, and working-memory assessments to relate individual differences to speech-sound learning. Experiment 2 participants received twice as much speech-sound training and more training at the category boundary. Procedural-memory in Experiment 1 predicted participants’ ability to learn category boundaries, but Experiment 2 did not replicate this effect. It is possible that more explicit training on stimuli along the category boundary in Experiment 2 favored participants with strong working-memory skills. It is also possible that the second day of training reduced effects of individual differences by boosting the performance of individuals with weaker memory skills. Further analysis into the dynamics and time-course of these memory systems in language-learning may reveal more nuanced contributions of procedural-memory in Experiment 2.

Authors: Ben Carlstrom, Chelsea McGrath, Carolyn Quam, Alisa Wang, Andrew Lotto

Presenter: Ben Carlstrom