CogSci2014-kbs-2
- 1. The Role of Procedural Memory in Adult Second Language Acquisition CogSci 2014 Katherine A. Brill-Schuetz & Kara Morgan-Short July 25, 2014
- 2. Adult L2 Development • Adult second language (L2) learners have shown varying degrees of development and final attainment (Newport, 1990; DeKeyser, 2000; White & Genesee, 1996; Clashen & Felser, 2006) • This variation may be due to differences in cognitive abilities (Ehrman & Oxford, 1995) • The effect of these differences on L2 development may depend on learning context (Ellis, 1993; Robinson 1997, 2001)
- 3. Procedural Memory in L2 Development • Theory posits an important role for procedural memory in L2 development • Declarative/Procedural Model (Ullman, 2001, 2004, 2005) • L2: as proficiency increases, there is a potential shift of certain grammar processing to procedural memory • Skill Acquisition Theory of L2 (DeKeyser, 2007) • Procedural memory has significant role in final stages (proceduralization and automatization)
- 4. Empirical Support • Procedural memory has been shown to play a role in L2 development • High scores on measures of procedural memory correlated to success in learning an artificial morphophonological rule (Ettlinger, Bradlow & Wong, 2010) • Procedural memory correlated with successful learning of an artificial language in implicit training conditions (Carpenter, 2008; Morgan-Short, Faretta- Stutenberg, Brill-Schuetz, Carpenter, Wong, 2014)
- 5. Aptitude Treatment Interaction • Role of procedural memory may be affected by learning context • An aptitude-treatment interaction (ATI) suggests tasks will differentially recruit cognitive abilities (Cronbach, 1957, 1975) • Ability Differentiation Hypothesis - important to match L2 learners to conditions of practice that favor their cognitive strengths (Robinson, 2007)
- 6. Motivation and Goals • Investigate role of procedural memory in L2 development • Multiple measures of procedural memory • Implicit and explicit training conditions • Investigate ATI involving procedural memory
- 7. Predictions/Hypotheses • The role of procedural memory changes depending on learning context • Prediction: Participants with stronger procedural memory will show higher levels of proficiency at final assessment, particularly under implicit training conditions
- 8. Design • Mixed factorial design: 2 (High vs. Low Procedural Memory) x 2 (Implicit vs. Explicit training) x 2 (Time 1 vs. Time 2) Day 1: Training and practice session 1 Day 2: Grammar Judgment Task 1 Day 3: Training and practice session 2 Day 4: Grammar Judgment Task 2 C o g n i t i v e A s s e s s m e n t s
- 9. Experimental Design: Methods • Participants: • 26 (16 female) participants recruited from psychology subject pool and were paid participants • Age 18-30 years old • All native English speakers with no cognitive, learning or hearing impairments • Varied experience with L2
- 10. Methods: Artificial Language • Brocanto2 • Syntactically productive, based on universal requirements of natural languages • 4 nouns, 2 adjectives, 2 articles, 4 verbs, 2 adverbs • Follows a different word order than English • Example: • Blom neimo lu neep li praz noyka • Blom square the neep the switches vertically • “The square blom switches vertically with the neep.”
- 11. Methods: Artificial Language • Learned B2 in order to make game moves on a computer-based gameboard • Trained either in explicit or implicit conditions • Practice in meaningful context • Alternated between sets of 2 comprehension and 2 production modules (total 400 trials)
- 12. Grammar and Cognitive Assessments • Grammaticality Judgment Task: • 120 novel sentences, half correct and half violation • Administered at Time 1 and Time 2 • Procedural Memory: • Alternating Serial Response Task (Howard & Howard, 1997) • Weather Prediction Task (Knowlton, Squire, & Gluck, 1994)
- 13. Results • Participants performed significantly better at later stages of learning when compared to early stages
- 14. Results Explicit Training Implicit Training Implicit Explicit
- 15. Discussion • High procedural memory performed better overall (p = .06) • Greatest difference in performance was in implicit training condition at Time 2 • Consistent with predictions
- 16. Discussion • Consistent with previous studies • Supports theories in regards to procedural memory involvement (DeKeyser, 2007; Ullman, 2001, 2004, 2005) • Suggests there may be an ATI in L2 learning (Cronbach, 1957, 1975; Robinson, 2007)
- 17. Conclusions/Implications • Provides evidence for procedural memory as an individual difference in L2 learning • L2 learning may be enhanced by matching a training condition to specific cognitive strengths (Ability Differentiation Hypothesis) • L2 training conditions may differ in task demands such that cognitive abilities are recruited differentially across learning contexts, as well as over time
- 18. Thank you! • Special Thanks to: • CogSLA Lab members • Sam Cralli • Namrata Shah • Karla Cabrera-Perez • Fatima Saeed • Allison Lee
- 19. Questions?