[Corpora-List] *Corpus Data and Psycholinguistics*: 1-day seminar, 19/05/16, Lancaster University.

Hardie, Andrew a.hardie at lancaster.ac.uk
Mon Apr 25 13:39:16 CEST 2016

[Corpora list members may be interested in this forthcoming event hosted by CASS at Lancaster. Please circulate; my apologies for any multiple postings. – AH]

From: Hughes, Jennifer Sent: 12 April 2016 22:12 Subject: Upcoming CASS Psycholinguistics Seminar

Dear all,

CASS is excited to announce an upcoming half-day research seminar on the theme of “Corpus Data and Psycholinguistics” (http://cass.lancs.ac.uk/?p=1971). The event will take place on Thursday 19th May 2016 at 1-5pm in Furness Lecture Theatre 3.

We have an exciting set of talks lined up from both internal and external speakers, including Professor Padraic Monaghan (Lancaster University), Dr Phil Durrant (University of Exeter) and Professor Michaela Mahlberg (University of Birmingham). The talks cover a range of psycholinguistic methodologies including eye-tracking and EEG. Abstracts are given below.

All are welcome to attend!

Jen Hughes

________________________________ Computational modelling of corpus data in psycholinguistic studies Padraic Monaghan, Lancaster University

Computational models of language learning and processing enable us to determine the inherent structure present in language input, and also the cognitive mechanisms that react to this structure. I will give an introduction to computational models used in psycholinguistic studies, with a particular focus on connectionist models where the structure of processing is derived principally from the structure of the input to the model.

Revisiting collocational priming Phil Durrant, University of Exeter

Durrant & Doherty (2010) evaluated whether collocations at different levels of frequency exhibit psycholinguistic priming. It also attempted to untangle collocation from the related phenomenon of psychological association by comparing collocations which were and were not associates. Priming was found between high-frequency collocations but associated collocates appeared to exhibit more deep-rooted priming (as reflected in a task designed to reflect automatic, rather than strategic processes) than those which were not associated. This presentation will critically review the 2010 paper in light of more recent work. It will re-evaluate the study itself and suggest ways in which research could be taken forward.

Durrant, P., & Doherty, A. (2010). Are high-frequency collocations psychologically real? Investigating the thesis of collocational priming. Corpus linguistics and linguistic theory, 6(2), 125-155.

Investigating the processing of collocation using EEG: A pilot study Jennifer Hughes, Lancaster University

In this presentation, I discuss the results of an EEG experiment which pilots a procedure for determining whether or not there is a quantitively distinct brain response to the processing of collocational bigrams compared to non-collocational bigrams. Collocational bigrams are defined as adjacent word pairs which have a high forward transitional probability in the BNC (e.g. crucial point), while non-collocational bigrams are defined as adjacent word pairs which are semantically plausible but are absent from the BNC (e.g. crucial night). The results show that there is a neurophysiological difference in how collocational bigrams and non-collocations bigrams are processed.

Exploring corpus-attested patterns in Dickens’s fiction - methodological challenges of using eye-tracking techniques Michaela Mahlberg, Kathy Conklin, and Gareth Carrol, University of Birmingham

The study of the relationship between patterns and meanings is a key concern in corpus linguistics. The data that corpus linguists work with, however, only provides a partial picture. In this paper, we will look at how questions of frequencies in corpora can be related to questions raised by data from eye-tracking studies on reading times. We will also discuss challenges of designing experiments to address these questions. As a case study, we focus on examples of patterns identified in Dickens’s fiction, but the methodological issues we address have wider implications beyond the study of literary corpora. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 12612 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20160425/cb75d5c6/attachment.txt>

More information about the Corpora mailing list