Features of Adaptive Assistance that Improve Peer Tutoring in Algebra (Walker, Rummel, Koedinger)
Collaborative Extensions to the Cognitive Tutor Algebra: Adaptive Assistance for Peer Tutoring
Erin Walker, Nikol Rummel, and Ken Koedinger
|Co-PIs||Nikol Rummel, Ken Koedinger|
Adaptive collaborative learning support, where an intelligent system assesses student collaboration as it occurs and provides assistance when necessary, is a promising area of research. While fixed forms of support such as scripting student interaction have had a positive effect on collaboration quality, they can overconstrain the interaction for some students and provide too little help for others. Using intelligent tutoring technology to support collaboration might be more effective, but little is known about how to build these adaptive systems for collaboration and what effects they might have. We explore this area of research by augmenting an existing intelligent tutoring system with a peer tutoring activity and providing automated adaptive support to the activity.
This project has focused on how to improve the construction of adaptive collaboration systems with respect to their suitability for classroom deployment and the breadth of the models they employ. Most currently implemented systems are prototypes which are limited both in the scope of interaction that they support and in their use by students. In our first PSLC project, “Collaborative Extensions to the Cognitive Tutor Algebra: A Peer Tutoring Addition, we explored the advantages of refactoring an existing intelligent tutoring system in order to transform it into a platform for collaborative research, such that interface and tutoring components can be added and removed in order to create different research conditions. We next demonstrated how individual intelligent tutoring models could be used as input to collaboration models in order to better assess peer tutoring behaviors, in the PSLC project [[Walker Adaptive Assistance for Peer Tutoring|Collaborative Extensions to the Cognitive Tutor Algebra: Adaptive Assistance for Peer Tutoring]. In this project, we extend this work by examining how individual models of student domain skills can be used as input to interaction models.
A related area of research is the potential of adaptive support for improving student interaction. The majority of the adaptive collaborative learning systems that have been developed have not been evaluated, and thus it is still unclear what influence adaptive support has compared to other forms of support. Our first step in this area was to develop adaptive domain support for the peer tutor, and compare it to a condition where the peer tutor is simply given problem solutions. While both types of support had advantages and disadvantages, it was clear peer tutors needed assistance that targeted collaboration skills in addition to domain knowledge. The next iteration of the system added adaptive interaction support to the adaptive domain support, and we compared the combined assistance to a fixed condition in a classroom study. As part of this project, we analyzed the study data in order to identify the broad impact both types of support have on the quality of student interaction, finding that adaptive support improves the quality of student help over fixed support. We now propose to investigate in more detail the potential role adaptive feedback could play in assisting student interaction by: 1) using HCI design methodologies to examine how students perceive and react to different features of support, and 2) empirically evaluating whether adaptive support has a cognitive or motivational influence on students. As an outcome of this research, we expect to add to understanding of the mechanisms by which adaptive support has an impact on student interaction, and how the support should be provided.
Background & Significance
Peer Tutoring: Learning by Teaching
Incorporating peer tutoring into the CTA might be a way to encourage deep learning. Roscoe and Chi conclude that peer tutors benefit due to knowledge-building, where they reflect on their current knowledge and use it as a basis for constructing new knowledge (Roscoe & Chi, 2007). Because these positive effects are independent of tutor domain ability, researchers implement reciprocal peer tutoring programs, where students of similar abilities take turns tutoring each other. This type of peer tutoring has been shown to increase academic achievement and positive attitudes in long-term classroom interventions (Fantuzzo, Riggio, Connely, & Dimeff, 1989). Biswas et al. (2005) described three properties of peer tutoring related to tutor learning: tutors are accountable for their tutee’s knowledge, they reflect on tutee actions, and they engage in asking questions and giving explanations. Tutee learning is maximized at times when the tutee reaches an impasse, is prompted to find and explain the correct step, and is given an explanation if they fail to do so (VanLehn et al., 2003).
Peer tutors rarely exhibit knowledge-building behaviors spontaneously (Roscoe & Chi, 2007), and thus successful interventions provide them with assistance in order to achieve better learning outcomes for them and their tutees. This assistance can target tutoring behaviors through training, providing positive examples, or structuring the tutoring activities. For example, training students to give conceptual explanations had a significantly positive effect on learning (Fuchs et al., 1997). It is just as critical for assistance to target domain expertise of the peer tutors, in order to ensure that they have sufficient knowledge about a problem to help their partner solve it. Otherwise, there may be cognitive consequences (tutees cannot correctly solve problems) and affective consequences (students feel that they are poor tutors and become discouraged; Medway & Baron, 1997). Domain assistance can take the form of preparation on the problems and scaffolding during tutoring (e.g., Fantuzzo, Riggio, Connely, & Dimeff, 1989). Although assistance for peer tutoring has generally been fixed, providing adaptive support may be a promising approach.
Adaptive Collaborative Learning Systems
In order to benefit from collaboration students must interact in productive ways, and collaborative activities can be structured (scripted) to encourage these behaviors (e.g., Rummel & Spada, 2007). However, fixed scripts implemented in a one-size-fits-all fashion may be too restrictive for some students and place a high cognitive demand on others (Rummel & Spada, 2007; Dillenbourg, 2002). An adaptive system would be able to monitor student behaviors and provide support only when needed. Preliminary results suggest that adaptive support is indeed beneficial: Adaptive prompting realized in a Wizard of Oz fashion has been shown to have a positive effect on interaction and learning compared to an unscripted condition (Gweon, Rose, Carey, & Zaiss, 2006). An effective way to deliver this support would be to use an adaptive collaborative learning system, where feedback on collaboration is delivered by an intelligent agent.
Work on adaptive collaborative learning systems is still at an early stage. One approach is to use machine learning to detect problematic elements of student interaction in real-time and trigger helpful prompts. Although implementations have lead to significant learning gains, the adaptive feedback appears to be disruptive to dyadic interaction (Kumar et al., 2007). Another promising approach has explored using an intelligent agent as one of the collaborators; students teach the agent about ecosystems with the help of a mentoring agent (Biswas et al. 2005). However, the agents do not interact with the students in natural language, one of the primary benefits of collaboration.
With respect to peer tutoring, intelligent tutoring technology could be applied either to supporting tutor behaviors or domain knowledge of peer tutors. As it is very difficult to build an intelligent tutor for collaborative processes, we decided to develop a general script for the peer tutoring interaction and then focus on providing adaptive domain assistance to peer tutors by leveraging the existing domain models of the CTA. A condition where students tutor each other with adaptive domain support provided to the peer tutor is likely to be better than a condition where the peer tutor merely has access to an answer key, because the support would be tailored to each individual tutor’s needs. It is also likely to be better than a condition where students use the CTA individually, because the students in the collaborative condition would be able to interact deeply about the domain material.