TR184: A Mediated Interaction Approach to Study the Role of Media Use in Team Interaction

Do differences in how teams use media relate to differences in how teams interact? Does media use play a role in meeting synergy or breakdowns? This dissertation explores these questions in the context of Architecture, Engineering, and Construction (AEC) project meetings, using an approach that I developed, called Mediated Interaction Approach (MIA). My observations of over 100 project meetings showed that good and bad patterns of mediated interaction recur in meeting practice. The observations also showed that each meeting is unique, but made up of hundreds of interactions and patterns of interaction that repeat themselves in a meeting and from meeting to meeting. However, practitioners, media designers, and researchers lack methods and metrics to discern, describe, assess, and compare different patterns of mediated interaction. Consequently, practitioners and media designers rely on intuition or anecdotal evidence to make changes to meeting practice or meeting media. Developing such methods requires analyzing the relationship between media use and team interaction at a micro-level to identify and abstract patterns of mediated interaction that practitioners and media designers can use as a resource to improve meeting practice and meeting media.

Prior approaches examining aspects of the relationship between team interaction and media use miss key aspects of this dynamic, fall short of operationalizing team interaction or media use concepts, or are ill-suited for the meeting context. Existing models of team interaction conceptualize interaction as multipurpose and analyzable with respect to three key processes - communication, reaction, and action - and these processes make contributions to project goals, to the meeting process (and its goals), and to interpersonal interactions. Existing studies operationalize at most two aspects of team interaction, e.g., communication and action, and address typically just one level of analysis, i.e., in relation to project goals, the meeting process, or interpersonal interactions. Thus, existing models and constructs are idealistic and unidimensional and do not capture the multi-purpose and multi-level nature of meeting interaction. Additionally, prior approaches operationalize team interaction and media use constructs that are task-, study-, or media- specific and are ill-suited for ad-hoc meetings that typically involve multiple tasks and multiple media. These studies limit their examination to feature-specific aspects of media use as opposed to general aspects of use, such as frequency and accessibility, level of interactivity, and instrumental purpose of media. This makes it difficult to compare patterns of media use involving multiple media. Finally, existing approaches miss the temporal aspect of meeting interaction and often rely on post-process data rather than observations. This makes it difficult to identify different patterns of mediated interaction that emerge and recur in meetings.

I developed MIA to address these shortcomings by observing over 100 AEC project meetings over a ten-year period and by analyzing 5,000 meeting interactions. MIA makes two key assumptions. First, the meeting process is analyzable as a set of discrete meeting interactions, each of which is analyzable from two distinct vantage points: how teams interact and how teams use media. Second, each meeting interaction v is analyzable relative to a standard of performance, regardless of task(s), that accounts for the multipurpose and multi-level nature of teams. MIA comprises the following four contributions:

(a) A model of the meeting interaction process, the Mediated Interaction Model (MIM), that integrates and builds on prior models of interaction and media use, applies to multi-task and multiple media contexts, and conceptualizes the meeting interaction as four interdependent processes: communication, reaction, action, and media use that make contributions to the project, meeting process, and interpersonal interactions.

(b) A Mediated Interaction Analytic (MIA) scheme to operationalize the MIM concepts by interpreting and coding video-recorded meeting interactions.

(c) An Interaction Spectra Method to operationalize and visualize the multi-categorical, temporal concepts of team interaction and media use as a spectrum: a) the Richness of Interaction spectrum, representing the range of interaction from breakdown to status quo to synergy and the extent to which teams achieve synergy and b) the Richness of Media Use spectrum (RMU), representing the range of media use from no use to rich use and the extent to which teams interact and engage with media use in the meeting interaction.

(d) The MIA Relational Spectra Method to describe patterns of mediated interaction and the process of how teams use media in relation to the process of how teams interact.

Findings from MIA show that it is not the task or media that matter, but the interaction and media use. Teams achieved synergy performing a range of tasks using a range of media. Teams that make media part of the team interaction, i.e., involve media in key aspects of team interaction, experience more synergy. Teams that enact a less rich role for media, i.e., infrequent use of media and minimal physical interaction with media, are more likely to maintain status quo and experience intermittent breakdowns. The findings suggest a mutually dependent symbiotic relationship between media use and team interaction in AEC meetings. MIA describes but does not yet explain the differences in meeting interaction and is a step towards developing normative models of media use and team interaction in natural contexts.