Navya Gullapuram
Pronouns: she/her
Research Mentor(s): Jonathan Brennan
Research Mentor School/College/Department: Linguistics / LSA
Program: UROPF
Session: Session 3 (11:00am – 11:50am)
Authors: Navya Gullapuram, David Abugaber
Abstract
Prior research suggests that conversational settings involve cognitive mechanisms that differ from individual cases of language use (e.g., reading, listening to audiobooks). For instance, in a conversation, speakers automatically predict the length of conversational turns in real time (Bogels & Torreira, 2015) and plan responses simultaneously to their interlocutors’ ongoing speech (Bogels et al., 2015). Our project extends previous studies on conversational alignment that have been limited to externally observable measures of performance (e.g., Boland, 2019) by using electroencephalography (EEG), which provides a covert and direct measure of neural activity during real-time language processing shown to detect effects not captured through behavioral methods alone (Tokowicz & MacWhinney, 2015). Specifically, we examine brain synchrony in oscillatory neural activity between individuals, which has been associated with better performance in interactive contexts such as word learning in children (Piazza et al., 2021) and collaborative puzzle solving (Reneiro et al., 2021). We ask whether brain synchrony is associated with two measures of conversational alignment used in previous studies: turn-transition times (the length of pauses between speakers’ utterances) and semantic similarity (the extent to which two speakers’ utterances involve the same meaning). In our study, EEG will be recorded from each of two participants using wireless headsets while they engage in an unscripted and naturalistic hour-long conversation. Turn transition times will be calculated as per Boland (2019); latent semantic similarity will be calculated as per Landauer and Dumais (1997). . In our analysis, we perform Pearson’s correlations to determine whether participants with higher brain synchrony show more aligned conversations as per these two measures. This project aims to extend prior research on conversational alignment via direct observation of neural activity while addressing a growing call for more ecologically valid methods in neurolinguistics (Willems, 2015) through the use of a naturalistic conversation paradigm.
