Native and Nonnative Number Agreement Processing Across Structural Distance: An ERP Study Gwisun Min Sungkyunkwan University
Native and Nonnative NumberAgreement Processing Across
Structural Distance: An ERP Study
Gwisun Min
Sungkyunkwan University
Introduction
Morphological
Processing
differences between
L1 and L2
Due to shallow processing and influence of age of acquisition (AoA) The Shallow Structure Hypothesis (SSH)
Due to limitations in general cognitive processing capacity The Declarative/ Procedural (DP) Model
Review of Literature
Can late Korean EFL learners with high proficiency process number feature checking as L1 speakers when there is structural separation between feature checking elements?
YES
The SSH is refuted.
NO
The DP model is refuted.
Review of Literature
Do different structural distances modulate L2 processing?
The SSH
L2 lacks syntactic hierarchies.
The DP Model
L2 processing of hierarchical structures can be as automatized
as L1 processing.
Research Questions
1. Do advanced late Korean EFL learners process ungrammatical number morpheme in simple NP in a nativelike way?
2. Do advanced late Korean EFL learners process ungrammatical number morpheme in partitive NP in a nativelike way?
Method
Participants
Materials
Procedures
Data acquisition and analysis
6
Method: Participants
• 9 native speakers
• 16 Korean EFL learners
Method
Participants
Materials
Procedures
Data acquisition and analysis
8
Method: Materials
(1a) Thomas added [QP many{+count} [NP exciting new items{+plural} ]] to the collection.
(1b) Thomas added [QP many{+count} [NP exciting new item{+plural} ]] to the collection.
(1c) Mary donated [QP many{+count} [NP (UNIT){+count} [PP of [DP her [NP
books{+plural} ]]]]] to the library.
(1d) Mary donated [QP many{+count} [NP (UNIT){+count} [PP of [DP her [NP
book{+plural} ]]]]] to the library.
(Adopted from Song, 2015)
Method
Participants
Materials
Procedures
Data acquisition and analysis
10
Method: Procedure
Method
Participants
Materials
Procedures
Data acquisition and analysis
12
Method: Data acquisition and analysis
• 28 electrodes, referenced online to FCz and offline to both mastoid electrodes
• The impedance threshold value was kept below 10kohm at all sites.
• 8.7% of trials were excluded across all conditions due to artefacts.
• Online filter and sampling: A 0.016 to 250Hz bandpass filter and 1000HZ sampling rate
• Offline filter and sampling: A 0.1 to 30 Hz bandpass filter and downsampled to 200Hz.
• Epochs: -200ms ~ 1000ms
• Time windows: 300-500ms, 500-700ms, and 700-1000ms.
• ANOVAs were performed within each time window separately for midline and lateral sites.
- Midline sites: Grammaticality x Electrodes (Fz, Cz, Pz)
- Lateral sites: Grammaticality x Anteriority (anterior, center, posterior) x Hemisphere
Results
Behavioral results
ERP results
- Native speakers
- Korean EFL learners
- Visual inspection of individual data
14
Results: Behavioral results
• Accuracy rates for comprehension questions
t (23) = 1.73, p > 0.05, d = 0.65
• Awareness of syntactic violations
t (23) = 1.51, p > 0.05, d = 0.66
Native speakers Korean EFL learners
94.79% 92.80%
Native speakers Korean EFL learners
46.66% 23.12%
15
Results
Behavioral results
ERP results
- Native speakers
- Korean EFL learners
- Visual inspection of individual data
16
Results: ERP Results
Native Speakers EFL Learners
Simple NP
Partitive NP
Simple NP
Partitive NP
17
Results
Behavioral results
ERP results
- Native speakers
- Korean EFL learners
- Visual inspection of individual data
18
ERP Results: Native speaker group
• Simple NPs • Partitive NPs
- No significant effect of grammaticality appeared.
19200 400 600 800 1000
-5μv
+5μv
_________ Grammatical--------- Ungrammatical
Fz
Cz
Pz
Results
Behavioral results
ERP results
- Native speakers
- Korean EFL learners
- Visual inspection of individual data
20
ERP Results: EFL learner group
• Simple NPs • Partitive NPs
21
Fz
Cz
Pz
200 400 600 800 1000
-5μv
+5μv
_________ Grammatical--------- Ungrammatical
Fz
Cz
Pz
Results
Behavioral results
ERP results
- Native speakers
- Korean EFL learners
- Visual inspection of individual data
22
Visual inspection of individual data
• Native speakers
• Korean EFL learners
Fz in 500-700ms in partitive NPs Pz in 500-700ms in partitive NPsCz in 500-700ms in partitive NPs
Fz in 300-500ms in simple NPs Pz in 300-500ms in partitive NPs
23
Topography of native speakers (n=6) excluding negative-response-dominant individuals in partitive NP condition
Topography of Korean EFL learners (n=16) in simple NP condition
24
Discussion
Absence of the P600
Functional nature of sustained anterior negativity (SAN) and early positivity
General discussion
Revisiting research questions
25
Discussion: Absence of the P600
Syntactic violation during comprehension task as opposed to GJT, may not elicit the P600.
1. The P600 is task-dependent (Brouwer & Crocker, 2017; Schacht et al., 2014).
2. Elicitation of the P600 depends on parsers’ subjective judgment of grammatical acceptability (Caffarra et al., 2015; Lemhöfer et al., 2014).
3. The P600 may not indicate detection of violations per se but sentence-level reanalysis after detection of linguistic violations (Tanner & van Hell, 2017).
26
Discussion
Absence of the P600
Functional nature of sustained anterior negativity (SAN) and early positivity
General discussion
Revisiting research questions
27
Functional nature of sustained anterior negativity (SAN)
• SAN was often associated with increased demand on working memory (e.g., Boudewyn, Long, Traxler, Lesh, Dave, Mangun, Carter & Swaab, 2015; Fiebach, Schlesewsky, and Friederici, 2002; Meltzer and Braun, 2013).
Not suitably applicable to the current experiment context since ungrammatical condition does not require additional WM in processing.
28
• SAN emerged in syntactic violation paradigm studies that involve comprehension question tasks (e.g., Hsu, Tsai, Yang, & Chan, 2014; Jiang, Tan, & Zhou, 2009; Jiang & Zhou, 2009; Ye & Zhou, 2008).
- A sign of automatic second pass semantic reinterpretation (Jiang et al., 2009)
Applicable to the current study. However, SAN in Jiang et al.’s study (2009) and that in the current study may manifest different stages of cognitive processes due to different time windows.
29
• SAN may signifies metacognitive, executive control processes that resolve conflicts, i.e., inhibitory processes (Hsu et al., 2014; Ye & Zhou, 2008).
- Inhibitory processes in the current study: comprehension vs. syntactic violation resolution
SAN
Ye & Zhou (2008) The current study
Task-takers with high cognitive control abilities
and low complexity sentence
Native speakers in simpler sentence condition (i.e.,
simple NP)30
Functional nature of early positivity
• Early positivity (300-500ms) in the current study
- The P300? However, it is usually elicited in oddball or memory recall tasks rather than to syntactic violations.
- Other possible interpretations:
1. The effort of integrating quantifiers with mismatching referents (Jiang et al., 2009)
2. A process of detecting or resolving conflict between an expected plurality and actual input of singularity (Ye & Zhou, 2008)
Effortful syntactic processing with attention involved
31
Discussion
Absence of the P600
Functional nature of sustained anterior negativity (SAN) and early positivity
General discussion
Revisiting research questions
32
General discussion
33
Native Speakers
Simple NPs Partitive NPs
Automatic syntactic violation detection and resolution
Controlled syntactic processing
Automatic syntactic violation detection and resolution
• Individual differences in language processing are reported in other studies (e.g., Tanner & van Hell, 2017; Ye & Zhou, 2008)
- Ye & Zhou (2008): Participants with high cognitive control abilities showed SAN in simple sentences while those with lower abilities showed sustained positivity to the same type of sentences, which is comparable to the current study findings.
Why different ERP patterns across structural
conditions?
Extra cognitive demand required for processing more complex sentences
Individual differences
34
35
Korean EFL Learners
Simple NPs Partitive NPs
Controlled detection/monitoring of syntactic violations
Discussion
Absence of the P600
Functional nature of sustained anterior negativity (SAN) and early positivity
General discussion
Revisiting research questions
36
Revisiting research questions
• The research questions were as follows:
1. Do advanced late Korean EFL learners process ungrammatical number morpheme in simple NPs in a nativelike way?
2. Do advanced late Korean EFL learners process ungrammatical number morpheme in partitive NPs in a nativelike way?
Late Korean EFL learners showed different ERP patterns from native speakers in both simple NP and partitive NP condition in spite of their high proficiency.
37
• This suggests that L2 processing is qualitatively different from L1 processing, in line with the SSH. However, the difference between L1 and L2 processing seems to lie in different degrees of automatization between the two rather than language deficits in L2. For both L1 and L2, the degree of automatization in processing was manipulated by the degree of sentence complexity, indicating that both L1 and L2 processing rely on general cognitive capacities, at least to some extent.
Implications and limitations
38
Implications and limitations
• Merit of the current study:
- An ERP study that investigated the effect of structural distance in L2 processing like the current study is scarce.
- With a comprehension task, this study revealed how syntactic violations were processed with parsers’ attention drawn to meaning.
• Limitations:
- Small sample size
- Bilinguals in the native speaker group
39
References
Boudewyn, M. A., Long, D. L., Traxler, M. J., Lesh, T. A., Dave, S., Mangun, G. R., Carter, C.S., & Swaab, T. Y. (2015). Sensitivity to referential ambiguity in discourse: the role of attention, working memory, and verbal ability. Journal of Cognitive Neuroscience, 27(12), 2309-2323.
Brouwer, H., & Crocker, M. W. (2017). On the proper treatment of the N400 and P600 in language comprehension. Frontiers in psychology, 8, 1327.
Caffarra, S., Molinaro, N., Davidson, D., & Carreiras, M. (2015). Second language syntactic processing revealed through event-related potentials: An empirical review. Neuroscience & Biobehavioral Reviews, 51, 31-47.
Clahsen, H., & Felser, C. (2006a). Grammatical processing in language learners. Applied psycholinguistics, 27(1), 3-42.
Clahsen, H., & Felser, C. (2006b). Continuity and shallow structures in language processing: A reply to our commentators. Applied Psycholinguistics, 27(1), 107-126.
Clahsen, H., & Felser, C. (2018). Some notes on the shallow structure hypothesis. Studies in Second Language Acquisition, 40(3), 693-706.
Fiebach, C. J., Schlesewsky, M., & Friederici, A. D. (2002). Separating syntactic memory costs and syntactic integration costs during parsing: The processing of German WH-questions. Journal of Memory and Language, 47(2), 250-272.
Jiang, X., Tan, Y., & Zhou, X. (2009). Processing the universal quantifier during sentence comprehension: ERP evidence. Neuropsychologia, 47(8-9), 1799-1815
Jiang, X., & Zhou, X. (2009). Processing different levels of syntactic hierarchy: An ERP study on Chinese. Neuropsychologia, 47(5), 1282-1293.
Lemhöfer, K., Schriefers, H., & Indefrey, P. (2014). Idiosyncratic grammars: Syntactic processing in second language comprehension uses subjective feature representations. Journal of Cognitive Neuroscience, 26(7), 1428-1444.
Meltzer, J. A., & Braun, A. R. (2013). P600-like positivity and left anterior negativity responses are elicited by semantic reversibility in nonanomalous sentences. Journal of neurolinguistics, 26(1), 129-148.
Schacht, A., Sommer, W., Shmuilovich, O., Martíenz, P. C., & Martín-Loeches, M. (2014). Differential task effects on N400 and P600 elicited by semantic and syntactic violations. PloS one, 9(3).
Song, Y. (2015). L2 processing of plural inflection in English. Language Learning, 65(2), 233-267.
Tanner, D., Grey, S., & van Hell, J. G. (2017). Dissociating retrieval interference and reanalysis in the P600 during sentence comprehension. Psychophysiology, 54(2), 248-259.
Ullman, M. T. (2001). A neurocognitive perspective on language: The declarative/procedural model. Nature reviews neuroscience, 2(10), 717-726.
Ullman, M. T. (2015). The declarative/procedural model. Theories in second language acquisition: An introduction, 135-158.
Ye, Z., & Zhou, X. (2009). Executive control in language processing. Neuroscience & Biobehavioral Reviews, 33(8), 1168-1177.
THANK YOU
40
Contact Information
• Name: Gwisun Min
• Position: Lecturer
• Affiliation: Sungkyunkwan University
• Email: [email protected]