The Department of Linguistics is pleased to announce
Daniel Michel’s dissertation defense.
Monday, June 23, 2014, 2:00 pm- 4:00 pm
Applied Physics and Mathematics Building, Room 4301.

Individual cognitive measures and working memory accounts of syntactic island phenomena
Daniel Michel

This dissertation examines the on-line processing and off-line acceptability judgments of whether-islands using an individual differences approach in order to test processing accounts of island phenomena. Processing accounts of islands propose that the unacceptability of an island violation (1) can be attributed to difficulties in on-line processing compared to control sentences (2), but accounts differ in how this difficulty is characterized.

(1) * Who had the sailor inquired whether the captain befriended openly?
(2) Who had the sailor assumed that the captain befriended openly?

The capacity-constrained account (e.g. Kluender 1991), based on the capacity-constrained view of working memory (Just & Carpenter 1992), predicts that the greatest difficulty in the island violation will be at the clause boundary (whether) where a confluence of storage and processing costs overwhelm the capacity of the parser. Sprouse, Wagers and Phillips (2012) recently tested this account by looking for co-variation of acceptability scores and working memory scores. The three experiments reported here (acceptability judgments, Chapter 4; self-paced reading, Chapter 5; event-related potentials- ERPs, Chapter 6) follow a similar approach, but additionally test a novel ‘similarity-interference’ account of islands based on the similarity-based interference view of working memory (e.g. Gordon, Hendrik and Johnson 2001; Lewis, Vasishth & Van Dyke 2006).

In order to clarify the relationship between off-line acceptability and on-line processing, two frameworks are introduced (Chapter 4): the Cognitive Co-variation Intuition (CCI) and the Processing Benefits Schedule (PBS). Together with an examination of individual differences (i.e. reading span, memory interference), these frameworks allow for a clearer comparison of on-line and off-line data. Ultimately, the data reported here do not support a view where processing factors directly and transparently lead to the unacceptability of an island violation (neither do they directly support a grammatical account of islands). However, the ERP data indicate the importance of real time prediction for the on-line processing of islands. This is formalized as the gap predictability account of processing islands.

Specifically, high span readers are better able to adjust their predictions for a gap online (evidenced by an N400 response at the embedded gap, suggesting lowered expectation for a gap in an island context), but both high and low span readers show evidence of filler-gap association (evidenced by post-gap LANs). There was no evidence of a failed parse or reanalysis in any condition or in any group of participants, as predicted by both processing and grammatical accounts, yet these same participants still rated island violations as the least acceptable sentences. Even though the brain responses to the island and non-island embedded gaps were basically the same (modulated only for predictability of the gap), the acceptability responses still differed. There is no apparent evidence of a large on-line processing cost that would account for this difference in acceptability. These island violations appear to be unacceptable, but not unparseable.

The Department of Linguistics is pleased to announce
Lara Hochstein’s dissertation defense
Thursday, June 26, 2014, 10:00 am – 12:00 pm
Applied Physics and Mathematics Building, Room 4301.

Epistemic reasoning and implicature computation in typically-developing children and individuals with an Autism Spectrum Disorder
Lara Hochstein

This thesis explores the role of epistemic reasoning (i.e., reasoning about other people’s knowledge, beliefs, and intentions) in implicature computation by addressing two seeming paradoxes: first, the fact that typically-developing children fail at specific inferences known as scalar implicature until relatively late in development despite exhibiting basic epistemic reasoning abilities from an early age, and, second, the fact that individuals with Autism Spectrum Disorders (ASD) have been claimed to succeed at scalar implicature despite exhibiting deficits with epistemic reasoning in other domains.

Chapter 2 provides evidence that 5-year-olds successfully compute ignorance implicatures – inferences that involve significant epistemic reasoning about speaker knowledge and utterance informative – despite failing to compute scalar implicatures. On the basis of this finding, we argue that children’s failures with scalar implicature do not stem from any difficulty with the epistemic reasoning involved and, instead, are most likely due to an inability to access the specific lexical alternatives involved in scalar implicature.

Results from Chapter 2 also show that 4-year-olds fail to compute both ignorance and scalar implicature, suggesting that the ability to compute basic Gricean inferences in language emerges around 5 years of age. This finding is somewhat at odds with claims in the literature that children can compute other forms of pragmatic inference such as ad hoc implicatures at an earlier age (e.g., Stiller et al, 2011). Chapter 3 therefore explores the role of epistemic reasoning in children’s ability to compute ad hoc implicatures. We show that 4-year-olds successfully compute ad hoc implicatures despite failing to compute ignorance implicatures, which raises the possibility that children compute ad hoc implicatures before they are able to engage in epistemic reasoning about speaker knowledge and utterance informativeness.

Finally, chapter 4 tests epistemic reasoning abilities in high-functioning children and adolescents with ASD. Results show that high-functioning adolescents with ASD successfully compute both scalar and ignorance implicature, while younger children with ASD fail at both inferences. These results indicate that high-functioning individuals with ASD are capable of the kind of epistemic reasoning required to make basic inferences in language despite their other pragmatic deficits, although this ability nevertheless appears to be somewhat delayed in ASD.

Congratulations to Hope Morgan, who is one of three winners of the 2014 Linguistic Society of America’s Student Abstract Awards this year, for her paper “The Emergence of Syntax in Kenyan Sign Language: Constituent Order and Space”. See the LSA website for details: http://www.linguisticsociety.org/news/2013/09/13/2014-lsa-awards-announced. Congratulations, Hope!