Assessing Visual Speech Communication Abilities: Cross-Situational Consistency in Lipreadability

Abstract Speech perception during face-to-face conversations is perhaps the most functionally significant aspect of multisensory processing in people’s everyday lives. Although they may not be aware of it, people use lipreading to both supplement auditory processing under poor listening conditions and compensate for age-related declines. Thus, understanding the sender’s role in communication is as important as understanding the receiver’s role, and in both cases, a fundamental question concerns whether individual differences reflect abilities and not just skills. From a psychometric perspective, an ability is distinguished by a high degree of cross-situational consistency in performance, which in turn has implications for how one remediates weaknesses clinically. Methodologically, round-robin designs, in which each participant is paired with everyone else in their group as both a sender and a receiver, represent an efficient way to measure cross-situational consistency, but although such designs have been used to study lipreading, the data have never before been analyzed in ways that address the ability/skill issue. Accordingly, the present investigation reanalyzed data from a previous round-robin study to assess consistency in both the accuracy with which different participants lipread others in their group and the accuracy with which individual talkers were lipread by the others in their group. The results provide evidence of two distinct abilities in visual speech communication: Receivers were consistently either good or poor at lipreading regardless of who the talker was, whereas talkers were consistently lipread either accurately or inaccurately regardless of who the lipreader was. Thus, just as communication under good listening conditions depends on a talker’s intelligibility and the listener’s hearing ability, under poor conditions, it also depends on two other abilities — both the receiver’s lipreading ability, and the talker’s ability to be lipread, which we have termed their ‘lipreadability’. Importantly, these two visual speech abilities appear to be uncorrelated.