Effects of Shared Gaze on Audio- Versus Text-Based Remote Collaborations

Abstract

Remote collaborations are becoming ubiquitous, but, despite their many advantages, face unique challenges compared to collocated collaborations. Visualizing the collaborator's point of gaze on a shared screen has been explored as a promising way to alleviate some of these limitations by increasing shared awareness. However, prior studies on shared gaze have not considered the medium of communication and have only studied its effect on audio. This paper presents a study that compares the effects of shared gaze on collaboration performance during audio- and text-based communication. We find that for text, shared gaze improved task correctness and led collaborators to look at and talk more about shared content. Similar trends are found for gaze-augmented voice communication, but contrary to the slower performance in text, it also saw improvements in completion time as well as in cognitive workload. Our findings demonstrate the differences in how shared gaze impacts audio- versus text-based communication and highlight the need to further understand the nuances of the medium of communication when designing novel tools to support remote collaborators.

I Cannot See Students Focusing on My Presentation; Are They Following Me? Continuous Monitoring of Student Engagement through "Stungage"

“…Afterwards, a few works [14,15,27,47] captured the eye gaze signal through the eye tracker for collaborative reading, writing, problem-solving, and learning. Later on, the authors in [28,49] explored acoustic signal along with the eye gaze for improving the remote collaborative performance. While eye gaze monitoring by eye tracker is a promising technique, the cost and availability of the tracker to all the students is a major obstacle.…”

How Can High-Frequency Sensors Capture Collaboration? A Review of the Empirical Links between Multimodal Metrics and Collaborative Constructs

“…Body metrics (e.g., amount of movement [27,52,91,92], distance between members [58,59]) achieved a high rate of success for hands-on tasks while being generalizable across different contexts. Combinations of verbal (e.g., speech participation [25,27,29,30]) and gaze metrics (e.g., area of interest [78], joint visual attention [26,70,77,83,86]) were usually successfully linked to increased performance, although one paper found low-level gaze features to be unsuccessful predictors [72]. Physiological metrics, on the other hand, were largely unsuccessful and there was notable divergence in the success levels of physiological synchrony [62,63,65,66].…”

“…Other notable findings revealed that body distance was significantly related to physical engagement [79]. Additionally, during voice-based communication between collaborators, the presence of shared gaze reduced cognitive workload, potentially because it provided a shared referent [86]. However, the authors also suggest that, in text-based communication, shared gaze may increase cognitive load since participants must divide their attention.…”

“…In the second (and largest) category, a theory was mentioned in the introduction or related work section of the paper to justify the choice of a particular collaborative outcome or data-derived feature (e.g., [48,79,86,91]). Studies often drew upon theory to argue for the significance of a particular sub-dimension of collaboration.…”

“…One example is the study of joint attention [124], the mechanism by which a shared reference helps collaborators build common ground (i.e., grounding theory [125]). Joint visual attention (JVA) was extensively studied using eyetrackers in MMCA studies (e.g., [24,70,71,73,77,82,84,86]). In this case, theory did more than inform the choice of collaborative features; it created a space where researchers could go back and forth between theoretical constructs and data-generated metrics to obtain a clearer understanding of the relationship.…”