Andrea Zukowski
1401 Marie Mount Hall
Linguistics Department
University of Maryland
College Park, MD 20782
(301) 405-5388
zukowski@glue.umd.edu

home home research publications personal useful links cnl lab

Current Projects

Language Competence and Performance in Williams Syndrome Williams Syndrome

These projects, funded by NICHD (2003–4), investigate knowledge and use of specific aspects of syntax in people with Williams syndrome. The broad question of interest is how mental retardation does and does not influence the ability to acquire different structures and the ability to use those structures in real time production and comprehension. The structures and constraints under investigation include:

1. Relative Clauses

2. Tag Questions

3. Affirmative and Negative Wh-Questions

4. Control in Temporal Adjuncts and the Locality Constraint on Reflexive Binding

5 . Constraints on Wanna Contraction

Related Projects with Typically Developing Populations

A repeated finding in this research is that people with Williams syndrome display exaggerated levels of difficulty with normally difficult syntactic structures. This includes exaggerated rates of production of errors that typically developing children make at younger ages. For some people it also includes failure to eventually recover from developmentally normal errors. In order to intepret these findings, it is critical to understand why certain structures cause difficulty for children even under the best of circumstances and how children normally manage to recover from errors.

1. The Status of Polarity in Children's Tag Question Rules

It is widely recognized that the most difficult aspect of tag questions for children to acquire is their polarity (knowing whether a particular tag question should be negative or affirmative). It is also known that the special difficulty of tag polarity lasts even longer in children with Specific Language Impairment or early focal brain lesions. Surprisingly, there has been little attempt to understand what the status of polarity is in children's tag question rules. For example, do children look like they are probablistically following an adult model (where tag polarity must be mis-matched to the polarity of the declarative)? Or do children have their own child-like polarity rules? (e.g. all tags are negative, etc.). This study looks at individual polarity patterns in both elicited production and judgment of tag questions in typically developing children in order to determine the current status of polarity in their tag question rules. This in turn suggests what children will have to do in order to eventually achieve adultlike polarity patterns in their tag questions. The findings indicate that most children genuinely overgenerate in the polarity patterns that they find acceptable, and this results in a learnability problem. We argue that the problem is resolvable if we assume, following recent suggestions in the literature, that the polarity mis-matching feature of tag questions is an entirely natural consequence of pragmatics.The key to recovering from matched polarity errors may lie in recognizing that the stressed auxiliary in the tag indicates focused polarity (a non-trivial task).

2. The Production of Sentences that We Fill Their Gaps

What happens when you start a sentence in such a way that there will be no way to end it grammatically without editing your original message? Our research suggests that in some cases, you go ahead and produce a sentence that you readily recognize as ungrammatical, rather than stopping mid-stream and restarting the whole sentence. This study investigates whether individual differences among normal adults in inhibition and working memory predict the liklihood of producing sentences that are known to be ungrammatical. This research has potential implications for explaining exaggerated levels of errors in the speech of people with Williams syndrome (or any other disorder where inhibition and/or working memory are compromised).