Taxonomy of Teams, Team Tasks, and Tutors

Dorneich, Michael
Bonner, Desmond
Gilbert, Stephen
Dorneich, Michael
Burke, Shawn
Walton, Jamiahus
Ray, Colin
Winer, Eliot
Gilbert, Stephen
Major Professor
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue
Aerospace EngineeringMechanical EngineeringVirtual Reality Applications CenterElectrical and Computer EngineeringIndustrial and Manufacturing Systems EngineeringVirtual Reality Applications CenterHuman Computer Interaction

While significant research has been done on teams and teaming (Salas, et al. 2004), less work has been done to characterize teams and team tasks in terms of the feasibility for them to benefit from intelligent tutoring. This theoretical paper begins to describe how the parameters of team structures addressed may affect the ways in which a team can accommodate external guidance. In addition, parameters of team tasks and resulting team tutors are also described. Examples of both team structures and team tasks are provided so that the resulting theoretical framework offers guidance for design decisions during the construction of intelligent tutoring systems (ITSs) for teams and the Generalized Intelligent Framework for Tutoring’s (GIFT) supporting team architecture. ITSs have been successful at improving performance in a wide variety of domains ranging from academic topics such as math (e.g., Koedinger, Anderson, Hadley & Mark, 1997) to work-based tasks such as management of power plants (Faria, Silva, Vale & Marques, 2009). However, there have been few ITSs designed for educating or training teams (Sottilare, Holden, Brawner & Goldberg, 2011). Despite much research on teaming since the 1970s, team performance is widely variable and difficult to predict (Sims & Salas, 2007), and there is a significant need for team-based ITSs. A taxonomy of team tutoring is present-ed (see Figure 29 for top level key elements). This paper describes three taxonomies: teams, team tasks, and relevant tutoring factors. The taxonomies are based on reviewing the teaming literature with a particular focus on the characteristics of each that would influence the design of a team-based intelligent tutoring system. This work leverages the extensive literature review of teaming by Burke et al. (in progress) as well as recent work that has sought to identify those major factors which impact team performance Salas, Shuffler, Thayer, Bedwell & Lazzara (in press).

The taxonomies provided below are designed to help guide the design of software architecture to support team ITSs within GIFT. GIFT is a powerful software architecture designed to support a wide spectrum of intelligent tutoring. It supports the traditional components of most ITSs: the learner model, the domain model, the pedagogical model, and the learner interface, but does so generically (Sottilare, Brawner, Goldberg & Holden, 2012; Sottilare, Graesser, Hu & Holden, 2013). Thus, a multitude of learners might manipulate a wide range of user interfaces as they engage with various domains while being taught using 190 a variety of pedagogies. However, the GIFT architecture does not naturally support teams. Team compo-nents are necessary if GIFT is to support team tutoring, but they are not present in the current release. In their 2011 paper, Sottilare et al., the creators of GIFT, describe the challenges of creating team tutors in detail.


This is a proceeding from the Second Annual Generalized Intelligent Framework for Tutoring (GIFT) Users Symposium (2014): 189. Posted with permission.