Amanda N. Nili

English, Psychology &
Social Behavior

Amanda Nili met her mentor at a Psi Chi induction ceremony, and Professor Pearl offered her the opportunity to participate in the Computation of Language Laboratory as a means of completing her Linguistics minor. Her project is unique in that it is an empirical analysis of the validity of a fairly ubiquitous assumption about syntax acquisition in the field of linguistics. The work is also characterized by a discussion of the implications of this assumption for the methodology of linguistics researchers. After graduation, Amanda plans to apply to graduate programs in Clinical Psychology and pursue a career in academia and clinical work.triangle.gif (504 bytes)




This study investigates an implicit assumption held by some language acquisition researchers that the frequency of linguistic structures correlates with how acceptable native speakers find these structures. The basic foundation of this assumption is that speakers have learned which structures are acceptable by how often they occur in the input. The authors compare Sprouse and Almeida’s acceptability judgment data of a wide range of linguistic structures with the frequencies of those same structures in child-directed speech. The utterances studied are from six American English corpora in the CHILDES database, which contains 170,263 child-directed utterances. Notably, the frequencies of very surface structural representations were not found to correlate with acceptability data. Future work can investigate whether acceptability judgments correlate with the frequencies of more abstract linguistic categories (e.g., using semantic abstraction or using additional syntactic abstraction), which would suggest that speakers use input frequency as a foundation for their linguistic knowledge, but do so in a more sophisticated way.triangle.gif (504 bytes)

back.gif (221 bytes) next


Faculty Mentor                                                                                                                

Lisa S. Pearl

School of Social Sciences

Amanda Nili’s research project investigates one of those assumptions that is often implicit in the field of language science, but rarely explicitly mentioned. As such, this assumption has subtly shaped the direction of research for theories of how we represent linguistic knowledge in our minds, how we deploy that linguistic knowledge in real time, and how we learn that linguistic knowledge in the first place. Amanda’s research not only highlights that this assumption exists, but pursues one of the testable predictions of it. Her findings are illuminating and suggest that this assumption is not obviously a good one when it comes to how we learn our linguistic knowledge.triangle.gif (504 bytes)

back.gif (221 bytes)

If you wish to view the paper in its entirety, please select the link given to the PDF file. pdf_logo.gif (126 bytes) [03_nili.pdf]

If you wish to download the Adobe Acrobat Reader,
please go to Adobe’s website (

Back to Journal 2014 Index

Copyright © 2014 by the Regents of the University of California.  All rights reserved.