University Calendar

Wednesday, May 5, 2021
3:30-4:30PM on Zoom
Victor Kuperman, McMaster University
What makes us proficient readers in a second language: new eye-tracking corpora

Short bio: https://mira.mcmaster.ca/team/bio/victor-kuperman
Zoom link: https://montclair.zoom.us/j/83302979786?pwd=UTUzRGNaUC8wV3hscy83Z29yeU9tZz09

Research into second language (L2) reading has a short supply of comparable, ecologically valid data representing a variety of first languages (L1). This hinders the building of cross-linguistically validated theories. Our data resource called MECO (Multilingual Eye Movements Corpus) addresses this need: it reports an eye-movement record of text reading in L1 and in English as L2 from speakers of 11 different L1s. These data help answer a central question: what factors determine reading proficiency in L2?

Using eye-tracking data and a test battery of component skills of English reading from 485 university students in 11 countries, we conducted large-scale cross-sample analyses of the predictors of L2 reading fluency and comprehension. Using multiple regression models and classification trees, we quantified the contributions to L2 reading behavior stemming from four key proposed sources of variance: individual L1 reading behavior, proficiency in L2 component skills, extralinguistic factors, and the L1 of the readers.

Major findings included (i) a fundamental contrast between the determinants of L2 reading fluency (as reflected in eye-movement patterns) versus comprehension accuracy, (ii) high within-participant consistency in the real-time strategy of reading in L1 and L2, and (iii) an unexpectedly small role played by the L1 versus L2 distinction in explaining variance in L2 reading behavior.

These novel cross-linguistic data help adjudicate between competing proposals about sources of variability in L2 reading behavior. Demonstration of a stable within-participant strategy in L1 and L2 reading is a novel contribution to theories of eye-movement control in reading.