Disclaimer / Avis de non-responsabilité

Psychoshorts – 2014

Growth Curve Analysis to analyze visual preference data from both eye trackers and raw coding

Organized by Chris Fennell & Tania Zamuner

Registration by email to Chris Fennell. Free for students and postdoctoral fellows.

Invited speakers:Brock Ferguson (Northwestern University) & Helen Buckler (University of Toronto)

Date: Dec. 9th, 2014 from 9am to 5pm

Location: University of Ottawa, Room 2008 of the Vanier Building & Arts Building Rm 509

 

Schedule

9am: Welcome

9:15-12 noon: Introduction/refresher to R

12 Lunch

12:30-1:30 Talks in Arts Building Rm 509

  • Helen Buckle, Voicing alternations in the developing mental lexicon
  • Brock Ferguson, When veps cry: Can toddlers learn novel words from linguistic contexts alone?

1 :30- 5: Growth Curve Analyses

 

Abstracts

When veps cry: Can toddlers learn novel words from linguistic contexts alone?

Brock Ferguson, Northwestern University

As children acquire language, the input they receive includes not only child-directed speech but also overheard speech from nearby conversations. Here we ask whether 15-, 19-, and 24-month-olds can use these overheard conversations to learn the meanings of novel words by attending to the linguistic contexts in which these words are used. Using a novel eye-tracking paradigm, my colleagues and I document that, by 19 months, infants can use known verbs (e.g., The vep is crying) to infer the animacy status of novel nouns. Moreover, by 24 months, this mechanism undergoes a sharp increase in efficiency: Toddlers’ processing of novel words learned in this way becomes considerably faster, and they even begin to form hypotheses about the animacy of a novel noun from syntactic information alone (e.g., The vep is X-ing). These findings cast light on a powerful word learning mechanism that likely contributes to children’s well-documented vocabulary expansion during this time period.

 

Voicing alternations in the developing mental lexicon

Helen Buckler, University of Toronto

Both Dutch and German have a phonotactic constraint against voiced segments in final position (“final devoicing”), which gives rise to voicing alternations within morphological paradigms (e.g. Dutch ‘bed(s)’, be[t]-be[d]en). This alternation is notoriously difficult for children to acquire, however, previous studies have relied predominantly on the child’s (in)ability to produce this alternation correctly. For example, errors whereby the child says *betten instead of bedden persist long into childhood. In this talk I will discuss data from an experiment using the Looking While Listening Paradigm (aka Visual World Paradigm), a perceptual study designed to test how voicing alternations are represented in the developing mental lexicon. Dutch and German toddlers’ sensitivity to mispronunciations of voicing word-medially was measured (e.g. “Look at the bedden / *betten”). Results suggest that toddlers’ lexical representations of voicing alternations are more robust than previously thought. Cross-linguistic differences can be attributed to language-specific distributional patterns of the voicing contrast and voicing alternations.