Citekey: @chan1997

Chan, C. K. K., Burtis, J., & Bereiter, C. (1997). Knowledge Building as a Mediator of Conflict in Conceptual Change. Cognition and Instruction (Vol. 15, pp. 1–40). doi:10.1207/s1532690xci1501_1


This study examined how individuals and peers process scientific information that contradicts what they believe and assessed the contribution of this activity to conceptual change. Participants included 54 students in Grade 9 and 54 students in Grade 12, who were randomly assigned to four conditions: (a) individual conflict, (b) peer conflict, (c) individual assimilation, and (d) peer assimilation. Depending on the condition, students were asked to think aloud or discuss with their peers eight scientifically valid statements, which were presented in an order that either maximized or minimized the conflict between new information and existing beliefs. Pretest and posttest measures of prior knowledge and conceptual change were obtained, and student verbalizations were tape-recorded and coded for five levels of knowledge-processing activity. Two major approaches were identified from this analysis: direct assimilation, which involved fitting new information with what was already known, and knowledge building which involved treating new information as something problematic that needed to be explained. A path analysis indicated that the level of knowledge-processing activity exerted a direct effect on conceptual change and that this activity mediated the effect of conflict. Knowledge building as a mediator of conflict in conceptual change helps to explicate previous equivocal research findings and highlights the importance of students’ constructive activity in learning.


A major research theme in instructionalpsychology emphasizes students’self-regulatory strategyin their own learning (Bereiter & Scardamalia, 1992; Glaser& Bassok, 1989; Resnick, 1989). Theoretical notions such as surface-deep (Biggs,1984), mindfulness (Salomon & Globerson, 1987), explanation (Bielaczyc et al.,1995), and intentional learning (Bereiter & Scardamalia, 1989) have been used tocharacterize the distinction between a passive versus an active approach to learni (p. 3)

tion is particularly interestingin the context of conceptual change. Research inreading (Scardamalia & Bereiter, 1984), writing (Scardamalia & Bereiter, 1987),and intentional learning (Bereiter & Scardamalia, 1989; Chan, Burtis, Scardamalia,& Bereiter, 1992) has shown the importance of a problem-centered approach tolearning. Two contrasting approaches to understanding new conceptsin unfamiliardomains have been identified: Direct assimilation involves fitting new informationdirectly into existing knowledge, whereas knowledge building involves learnerstreating new conceptsas something problematic that they need to explain (Bereiter& Scardamalia, (p. 4)

knowledge-building approach would seem beneficial for the learning of difficultconcepts. Instead of directly assimilating new concepts, students employing knowl(p. 4)

The goal of this study was to examine how students construct scientific under-standingwhen confronted with information that contradicts what they believ (p. 5)

The second objective was to assess the efficacy of conflictual information andto examine its relations with a knowledge-building approach to learni (p. 5)

the first objective was to identify the kinds of knowledge-processingactivity students engagein when they are learning from scientific information andto assess the effects of such activity on conceptual change (p. 5)

edge-building activity are more likely to delay immediate interpretation, to recog-nize the difficulties of the new concepts, and to avoid equating ontologicallyincompatible new concepts with their prior conception (p. 5)

A third objectivewas to investigate the role of peer interaction inconceptualchange by examining how pairs of students learn from incompatible informati (p. 6)

Three research questions were addressed: (a) Do the hypothesized approaches ofdirect assimilation and knowledge building differentially affect conceptual change?(b) Does conflict lead to more conceptual change, and are its effects mediated byknowledge-processing activity? and (c) Does peer collaboration foster conceptualchange when students are confronted with contradictory informati (p. 6)


Surface Versus Deep Feature (p. 28)

l: A path analysis showed thatknowledge-processing activity, as defined by the scale, exerted a direct effect onconceptual change and mediated the effects of confli (p. 28)

These findings are consistent with the well-established idea that experts repre-sent problems at a deep level, whereas novices attend only to the surface features(Chi, Feltovich, & Glaser, 1981) (p. 29)

Piecemeal Editing Versus Delayed Interpretatio (p. 29)

Even though theyhad very inadequate knowledge of thedomain, they were trying to move in the direction of formulating problems at adeeper leve (p. 29)

Knowl-edge-building responses, however, suggested students’ concerns with a deeperproblem:the principle of evolution representedin the text and how it fits with theirprior belie (p. 29)

ly, knowledge-building responses reflected an active stance inidentifyingconflict and attempts to reconcile the discrepancies. Students employing a knowl-edge-building approach do not avoid conflict but use the contradictory statementsas opportunities for upgrading their domain understand (p. 30)

fs. The more expert learners, however, recognized that something was differ-ent; hence, they were able to subject their knowledge to inquiry. Just as theory-evi-dence coordination isimportantin scientificexperimentation (Klahr & Dunbar,1988; Kuhn, 1989), a problem-solving process is required for identifying andresolving the discrepancy between new information and domain understandi (p. 30)

The importance ofexplanationin science learning has been well documented (Bielaczyc et al., 1995;Chi, de Leeuw, et al., 1994 (p. 30)

In contrast to usinga problem-minimization approach, the more expert learners attend to deep princi-ples, construct coherence from discordant information, seek out knowledge con(p. 30)

s, students employinga knowledge-building approach may benefit more com-pared with those who provide a one-shot explanation to eliminate discrepant factsthat do not fit their beliefs. In viewing new information as problematic and asrequiring explanations, students are engagedin an ongoing process of problemrecognition and conflict resolution. Even ifthey have constructed inaccurateexplanations, they are more likely to detect anomalies in upcoming information andto revise their models continua (p. 30)

Patching Versus Explanation-Based Inquiry (p. 30)

Theory-Evidence Conflation Versus Problem Recognit (p. 30)

Indeed, focusing on the multiplicityof relations among pieces of information (Brown & Day, 1983) and emphasizingcoherence (Scardamalia & Bereiter, 1984) are crucial to meaning construct (p. 30)

flict, and engage inexplanation-based (p. 31)

A distinction needs to be made between external and internal conflict: Often,contradictory information is presented in the hope that it will produce cognitiveconflict. These results seem to show that, without knowledge-building activity,confrontation may not produce cognitive conflict but only assimilation or externalcompetiti (p. 33)

Collaborative Processing of Contradictory Informat (p. 33)

. The path analysis suggests that conflict may triggerknowledge-building activity, which then leads to conceptual change, but thatconflict in the absence of knowledge building will not produce conceptual change (p. 33)

Conflict in itself is not enough; it needs to be mediated by students’ knowledge-building activity. Viewing learning as problematic minimizes assimilation an (p. 36)


leads to more conceptual change (p. 37)

Blog Logo

Bodong Chen



Crisscross Landscapes

Bodong Chen, University of Minnesota

Back to Home