BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Acquiring Verb Argument Structure from the Input Distribution: An 
 Artificial Language Study - Dr Elizabeth Wonnacott\,  University of Oxford
 \, Department of Experimental Psychology
DTSTART:20071016T150000Z
DTEND:20071016T163000Z
UID:TALK8358@talks.cam.ac.uk
CONTACT:Napoleon Katsos
DESCRIPTION:Adult language combines a complex mix of regular\, rule-like p
 rocesses and more conservative\, lexically based patterns. For example\, v
 erb argument structures may generalize to new verbs (Arthur flupped the ba
 ll to Ford->Arthur flupped Ford the ball) yet resist generalization with c
 ertain lexical items (Arthur carried the ball to Ford -> *Arthur carried F
 ord the ball). Pinker (1989) suggests that that learning whether a particu
 lar verb occurs with a particular argument structure involves learning a f
 ine-grained semantic representation. However\, frequency-based entrenchmen
 t effects in young children (Theakston 2004) and statistical effects in se
 ntence processing (Trueswell et al 1993)\, suggest that learners track ver
 b-structure co-occurrences. This concurs with recent approaches which emph
 asize the role of Statistical Learning processes in Language Acquisition (
 Saffran et al. 1996). Our work uses Artificial Language Learning to explor
 e whether the relationship between verbs and argument structures can be ac
 quired from input statistics. Adult participants were exposed to languages
  in which input distribution provided the only cue to verb subcategory (no
  semantic or phonological correlates). Each language had 12 verbs and two 
 synonymous argument structures\, VSO and VOS-Particle. Four languages were
  explored\, with differences among them in the degree to which their verbs
  exhibited lexically based versus language-wide patterns. After several da
 ys of exposure\, learners took Production\, Grammaticality Judgment and Ey
 e-tracked Comprehension tests. In this last\, participants heard sentences
  and viewed two scenes: correct and agent-patient reversed. If the verb is
  biased to occur with one structure\, eye-movement data should reveal a lo
 oking preference before the disambiguating particle.  \n \nOverall\, parti
 cipants proved able both to acquire verb subcategorization\, and to genera
 lize. Behavior in the different tests was very consistent\, and reflected 
 both the statistical preferences of particular verbs and across-verb stati
 stics (shown with 'new' verbs introduced during testing and not phonologic
 ally/semantically related to 'old' verbs). Eye-movement data revealed that
  these statistics influenced on-line processing. However\, the tendency to
  generalize was affected by a third source of information: the distributio
 n of verb types in the language. Learners exposed to languages in which th
 e majority of verbs occurred in multiple structures were more likely to ge
 neralize.  \n \nIn conclusion\, learning and processing of verb argument s
 tructure is strongly driven by various distributional properties of the in
 put\, and these purely formal phenomena may occur in the absence of semant
 ic cues. In addition\, Artificial Language learning is shown to provide a 
 fruitful methodology for exploring the relationship between Statistical Le
 arning and statistical effects in sentence processing. 
LOCATION:GR-06/07\, English Faculty Building
END:VEVENT
END:VCALENDAR