Multimodal coordination of concurrent collaborative tasks
Responsable du projet Adrian Bangerter
Collaborateur Eric Mayor
Résumé Multitasking (i.e., the concurrent performance of two different tasks) is a frequent phenomenon in everyday life. Important practical issues involve for example the effect of multitasking on task performance. Research on multitasking has revealed much about the basic cognitive processes involved, showing that sharing processing resources (attention, working memory, executive control) between multiple tasks can impair task performance (increasing completion times or error rates). Much of this research, however, has focused on multitasking behavior of individuals engaged in solitary tasks. Some research focuses on situations where people coordinate collaborative tasks (either one individual task and one collaborative task or two collaborative tasks). Such research has shown that visual access to one’s partner or co-presence is an important determinant of multitasking: It is more difficult to multitask when one’s partner is face-to-face than when s/he is remotely located. Many studies of collaborative multitasking do not investigate the role of dialogue in task coordination. This is a significant oversight, because dialogue (which involves both verbal and nonverbal acts) is the commonest means of coordinating collaborative activities. Based on a theory of language use as a form of collaborative activity, the applicant and colleagues have recently developed a model of collaborative task suspensions and reinstatements. This model specifies the additional constraints that accrue in situations of collaborative multitasking, notably coordinating, on a moment-by-moment basis, the common ground between participants as well as managing social-affiliational aspects like the face of the participants. But more research is needed on face-to-face situations where people not only switch between tasks at a time scale of minutes, but also how they coordinate simultaneous performance of two tasks. Face-to-face situations afford visual information, and therefore multimodal coordination may play an important role in such situations. Starting from this model, we therefore seek to answer three research questions: • Research Question 1: How do partners engaged in collaborative multitasking reconcile task-related and social-affiliational demands? • Research Question 2: How is collaborative multitasking multimodally coordinated? • Research Question 3: How do participants minimize their collaborative effort in coordinating multitasking? We investigate these research questions in an experimental study where pairs of participants complete two collaborative concurrent tasks: Talking together while walking, or telling each other a story while following an itinerary on a map. This situation is compared to several baseline situations with less or no multitasking constraints (telling a story without moving, telling a story while walking in a straight line, and navigating an itinerary without telling each other a story). These conditions create a first, within-subjects independent variable. We also manipulate the division of labor within the pair by either giving the map to the narrator of the story or to the listener (between-subjects variable), thereby creating a 4 (degree of multitasking) X 2 (conversational role of navigator) experimental design. We will study the effects of these conditions on various dependent variables designed to measure conversational performance (e.g., disfluencies, back-channels, gestures, gaze) and locomotive performance (e.g., walking speed, navigation errors). Forty pairs of participants will be videotaped while performing these tasks. Recordings will be transcribed. Dependent variables will be coded and analyzed in detail using both quantitative procedures (analysis of variance, regression) and qualitative analyses (retranscription of selected cases, detailed multimodal analysis, creation of a taxonomy of coordination strategies). The planned research will contribute to filling important gaps in our understanding about how participants coordinate the partly conflicting task-related and interpersonal demands emerging during concurrent collaborative multitasking. The results obtained will yield findings for writing at least two manuscripts for peer-reviewed journals in the fields of cognitive science and/or psycholinguistics. The publications will also add value to the research credentials of the postdoctoral researcher.
Mots-clés multi-tasking, interruptions, conversation, communication
Type de projet Recherche fondamentale
Domaine de recherche Psychologie
Source de financement FNS - Encouragement de projets (Div. I-III)
Etat Terminé
Début de projet 1-1-2012
Fin du projet 28-2-2014
Budget alloué 122'076.00
Autre information http://p3.snf.ch/projects-138195#
Contact Adrian Bangerter