Evitts P, Gallop R. Objective eye-gaze behaviour during face-to-face communication with proficient alaryngeal speakers: a preliminary study.
INTERNATIONAL JOURNAL OF LANGUAGE & COMMUNICATION DISORDERS 2011;
46:535-549. [PMID:
21899671 DOI:
10.1111/j.1460-6984.2011.00005.x]
[Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/31/2023]
Abstract
BACKGROUND
There is a large body of research demonstrating the impact of visual information on speaker intelligibility in both normal and disordered speaker populations. However, there is minimal information on which specific visual features listeners find salient during conversational discourse.
AIMS
To investigate listeners' eye-gaze behaviour during face-to-face conversation with normal, laryngeal and proficient alaryngeal speakers.
METHODS & PROCEDURES
Sixty participants individually participated in a 10-min conversation with one of four speakers (typical laryngeal, tracheoesophageal, oesophageal, electrolaryngeal; 15 participants randomly assigned to one mode of speech). All speakers were > 85% intelligible and were judged to be 'proficient' by two certified speech-language pathologists. Participants were fitted with a head-mounted eye-gaze tracking device (Mobile Eye, ASL) that calculated the region of interest and mean duration of eye-gaze. Self-reported gaze behaviour was also obtained following the conversation using a 10 cm visual analogue scale.
OUTCOMES & RESULTS
While listening, participants viewed the lower facial region of the oesophageal speaker more than the normal or tracheoesophageal speaker. Results of non-hierarchical cluster analyses showed that while listening, the pattern of eye-gaze was predominantly directed at the lower face of the oesophageal and electrolaryngeal speaker and more evenly dispersed among the background, lower face, and eyes of the normal and tracheoesophageal speakers. Finally, results show a low correlation between self-reported eye-gaze behaviour and objective regions of interest data.
CONCLUSIONS & IMPLICATIONS
Overall, results suggest similar eye-gaze behaviour when healthy controls converse with normal and tracheoesophageal speakers and that participants had significantly different eye-gaze patterns when conversing with an oesophageal speaker. Results are discussed in terms of existing eye-gaze data and its potential implications on auditory-visual speech perception.
Collapse