Working Memory and Relative Clause Attachment under

1
Working Memory and Relative Clause
Attachment under Increased Sentence
Complexity Akira Omaki Department of Second
Language Studies, University of Hawaii at Manoa
1. Introduction
3. Experiment 1

• (1) Someone shot
• the servant of the actress who
  was on the balcony.
• (N1) (N2)
  (RC)
• Cross-linguistic differences in RC attachment
  preferences
• Local Attachment (LA)
• English, Arabic, Norwegian, Romanian,
  Croatian, etc.
• Non-local Attachment (NLA)
• Spanish, Dutch, Afrikaans, French, Greek,
  Russian etc.
• Various Accounts for the cross-linguistic
  differences
• Implicit Prosody (Fodor, 2002)

Relative Clause Attachment
Method
Results

• Participants 36 English native speakers
• Materials 32 target items (4) 75 fillers
• (4a) Embedded Condition (EC)
• ? More complex (Storage cost at who 4)
• The babysitter that the sister of the
  schoolgirl who burned herself the other day
  adored was very nice.
• (4b) Sentence-complement Condition (SC)
• ? Less complex (Storage cost at who 2)
• The babysitter said that the sister of the
  schoolgirl who burned herself the other day was
  very nice.
• Example Question
• Who got burned the other day? 1. the sister 2.
  the schoolgirl
• Procedure
• A. Offline reading experiment on computer
• 1. Read the whole sentence on a computer
  screen
• 2. then press the space bar to show the
  question

Local Attachment
Non-local Attachment

• Significant negative correlation btw. RST scores
  and NLA responses
• 2-tailed correlation in
• EC condition r - .482 (pcondition r - .561 (p
• ? The greater RS leads to an LA preference.
  (Replicates previous findings)
• No significant difference between EC SC
  condition
• (Correlation between EC and SC condition
  was over .90)
• Weak trend for a local attachment preference
• Mean NLA responses in
• EC condition 42.19 (SD27.60) SC
  condition 46.35 (SD28.48)

Individual differences in RC attachment
preferences? Working Memory Capacity
Question Why the lack of sentence complexity
effect? A. Problems with the methodology? They
may not have read the whole sentence, since one
can answer the question by just looking at the
relevant RC attachment region ? Addressed in
Experiment 2, word-by-word self-paced reading task
The greater WM capacity leads to an LA
preference (i.e., Low-spans prefer NLA,
High-spans prefer LA)
4. Experiment 2 (in progress)
Preliminary RT Results
Method

• Main effect of complexity (F1F2) in of
  schoolgirl who burned
• ? Reflects the storage cost (Gibson, 2000)
• Main effect of complexity in was (F1F2)
• ? Reflects the integration cost (Gibson,
  2000)
• Main effect of attachment (though only F1) in
  himself/herself
• ? LA attachment preference, regardless of the
  sentence complexity
• No main effect of interaction in any of the
  regions (F1F2)

• Participants 32 English native speakers
• Materials 32 target items (5) 73 fillers
• Procedure Self-paced reading RST from Exp 1

2. Goal of the Present Study
(5a, c) Complex/Non-complex, forced LA
The babysitter (said) that the brother of the
schoolgirl who burned herself the
other day adored was very nice. (5b, d)
Complex/Non-complex, forced NLA The
babysitter (said) that the brother of the
schoolgirl who burned himself the
other day was very nice.
To further examine whether WM capacity
interacts with the RC attachment ambiguity in
off-line processing (Exp 1 ) and on-line
processing (Exp 2) by investigating (a)
whether high-spans and low-spans differ
in RC attachment preferences (as in
previous studies), and (b) whether the
sentence complexity (i.e., the
amount of memory cost) interacts
with RC attachment preferences
(cf. Eastwick Phillips, 1999)
Storage cost
LA preference
Integration cost
High vs. Low Between-subjects

• Mean RST score (n 32) 45.9, thus 46 as
  cut-off
• ? 16 low-spans, 16 high-spans
• ? Complexity (2) x attachment (2) x span size
  (2) design

• High-spans generally slower
• Integration, but NOT attachment, interacts with
  span size

5. Summary
Regions with main effect of span size
himself/herself and was

• The general LA preference was observed in
  offline and online processing
• RS significantly correlated with RC attachment
  ONLY in offline processing. In Exp 2, RS
  significantly interacted with integration cost,
  but not with attachment (cf. Caplan Waters,
  1999) this suggests that (at least) initial RC
  attachment preference is not influenced at all by
  memory capacity.

Main Effect RST pAttach p.053 No interaction
Main Effect RST pComplex pp
6. Acknowledgement

• I thank Matt Prior, Amy J. Schafer, Barbara
  Schulz, and Bonnie D. Schwartz for their help at
  various stages. I also thank the snack-eaters at
  the Psycholinguistics 3ji-no Oyatsu meetings.
• This research was partially supported by
  Elizabeth Holmes-Carr Scholarship, received with
  much appreciation.