Representations, Interlocutors, and Their Influences on Apparent Knowledgeability

• In Event: Integrating Issues of Knowledge and Interaction in Analyses of Cognition and Learning

Abstract

Objective

One of the most challenging issues faced by education researchers and practitioners relates to variability in apparent student “knowledgeability.” That is, we may know that students possess particular bodies of instructed knowledge, but this does not allow us to predict when and where that knowledge will present itself. In many respects, this is related to the longstanding transfer problem of when and where existing knowledge will be used (Salomon & Perkins, 1989).

Our goal in this paper is to take advantage of a unique set of data – in which a student discussed the same chemical phenomenon in four different conversational contexts with different interlocutors (Brown, 2009) – to develop explanations of variations in apparent knowledgeability.
Perspectives and theoretical frameworks

Like the rest of this symposium we believe that analyses of knowledge (diSessa, 1993) and interaction (Jordan & Henderson, 1995) can be used together to provide traction on important issues in the learning sciences. In our paper, we show how interactional features including framing (Tannen, 2003), representational practices (Greeno & Hall, 1997), and interactional history can help explain apparent knowledgeability.

Data

An undergraduate chemistry student with some formal instruction in dynamic phase equilibrium was video recorded over four separate days while: (1) being clinically interviewed, (2) tutoring a friend, (3) tutoring a stranger, and (4) being tutored by a graduate student.

Analysis

To coordinate analyses of knowledge and interaction, we created transcripts (example in Figure 2) that juxtaposed notations capturing speech, eye gaze, gesture, and representation use (Erickson, 1992; Goodwin, 1981) with notations related to elements of knowledge (Sherin, Krakowski, & Lee, accepted). These transcripts allowed us to investigate the temporal coordination of interactional features with knowledge expression.

We then performed two types of analyses:
1) Comparative analyses comparing the four types of conversations, and
2) Micro-longitudinal analyses that considered the conversations as consecutive events.

Results

In all four conversations, the student explained state changes as involving molecules overcoming attractions and differentials of density. We observed variation in the background knowledge related to molecular structure and relationships between heat, energy, and movement. We also found new, non-normative ideas about molecules (e.g., they form discrete clumps in liquid state) jointly constructed in later conversations.

We then explained these behaviors by jointly considering:
a) how each conversation was framed by the researcher with respect to who had expertise about what (Engle, 1996, 2006; Jacoby & Gonzales, 1991; cf. Hammer, et al. 2005);
b) how different drawings served as “material anchors” (Hutchins, 2005) that played a pivotal role in initiating and mediating the articulation of knowledge related to dynamic equilibrium; and
c) the history of prior interactions that made particular knowledge more or less relevant for being reported and built upon over time.

Significance

This work both demonstrates and explains how apparent knowledgeability can exhibit stabilities and variations across settings. More generally it shows how combining analyses of interaction and knowledge leads to insights that neither analysis could provide alone.

Authors

• Nathaniel J.S. Brown, Boston College

• Joshua Adam Danish, Indiana University

• David DeLiema, University of California - Los Angeles

• Randi A. Engle, University of California - Berkeley

• Noel D. Enyedy, University of California - Los Angeles

• Victor R. Lee, Utah State University

• Orit Parnafes, Tel Aviv University