[CaCL] CaCL: Winter 2011 update

[Fred Mailhot]

Greetings to all,

CaCL (the discussion group on Computation and Cognition in 
Language/Linguistics) will be resuming its weekly meetings for the 
Winter 2011 quarter. Those of you who were not able to attend last 
quarter are encouraged to join us.

TIME: Mondays 10:00-11:30
PLACE: Arps Hall Rm. 03

Our first meeting will be held on January 10th, during which one of 
Florian Jaeger's students will present some of her work (title & 
abstract below). N.B. this talk is in conjunction with two other talks 
on Jan 10(previously announced on the [Scholars] and [Lingosu] lists): 
Judith Degen at 1:30 in University Hall 147, and Florian Jaeger at 3:30 
in Oxley 122.


Looking forward to a great quarter,
Fred.



Maryia Fedzechkina
"Are there functional biases in language acquisition?"

Why do languages share structural similarities? The functionalist 
tradition has argued that languages have evolved to suit the needs of 
their users (e.g. Bates & MacWhinney 1982; Hawkins 2004). /How/ or /by 
what means/ functional pressures may come to shape grammar over time 
has, however, remained unclear. We investigate whether functional 
pressures operate during language acquisition biasing learners to 
(subtly) deviate from the input they receive.

We address this question in a series of artificial language learning 
experiments with adult native speakers of English. In these experiments 
we expose participants to miniature languages devised in the laboratory 
that are similar to natural languages in many ways but are small enough 
to be learned during the experiment.

Artificial language learning experiments can be used to investigate 
biases in acquisition (Newport & Aslin 2005; Culbertson & Smolensky 
2009). Experiment 1 investigates the trade-off between word order and 
case. Experiment 2 investigates optional case-marking (as in e.g. Korean 
or Japanese). In all experiments, we find a tendency to reduce the 
uncertainty (or ambiguity) in the input language. Experiment 1 provides 
evidence that language learners are more likely to fix the word order if 
a language does not have case. Experiment 2 provides evidence that 
language learners are more likely to mark untypical objects with case 
and that they prefer to apply case-marking in the less frequent 
(‘marked’) word order (cf. word order freezing).