Transfer Across Disciplines

• In Event: 52.029 - Interdisciplinary and Integrated STEM Education: Research, Practices, and Perspectives

Abstract

Our theoretical framework (Rebello, Cui, Bennett, Zollman, & Ozimek, 2007) for transfer consolidates traditional and contemporary perspectives. We briefly discuss the framework, its relationship with other frameworks, and implications for transfer of learning in problem solving across disciplines. Finally, we demonstrate that instruction guided by our framework can support transfer of problem solving from mathematics to physics better than traditional methods.

Our framework purports two transfer mechanisms, both cognitively supported by the dynamic creation of associations between new information read-out by a learner and her internal knowledge (Rebello et al., 2005). In ‘horizontal’ transfer a learner’s read-out information activates a schema onto which new information is mapped. ‘Vertical’ transfer occurs where horizontal transfer fails. It involves modifying an existing schema or constructing a new schema to make sense of the situation. These two kinds of transfer align with others’ views (diSessa & Wagner, 2005; Salomon & Perkins, 1989; Schwartz, Varma, & Martin, 2008).

Schwartz et al. (Schwartz, Bransford, & Sears, 2005) have proposed that developing adaptive expertise requires guiding learners through an optimal adaptability corridor (OAC). Our transfer framework further suggests that navigating the OAC requires learners completing sequences of horizontal and vertical transfer tasks (V-H sequences) involving cognitive disequilibrium (Piaget, 1964) and productive failure (Kapur, 2008) to scaffold learning within a zone of proximal development (Vygotsky, 1978). Finally, the learner must reflect on their learning experience through V-H sequences. The goal is to enable learners to construct an adaptive schema to transfer their learning from the V-H sequences to other contexts.

Based on this framework, we designed an instruction to facilitate transfer of problem solving strategies in physics problems requiring integration (Nguyen & Rebello, 2011). Students enrolled in an undergraduate physics class completed a baseline task to ascertain their ability to solve integration problems.

The experimental group completed a V-H sequence designed as per the above framework. The goal of the V-H sequence was to construct a solution to a new problem and reflect on the process of solution construction. The control group attempted the same new problem by themselves, and then received a traditional written textbook solution to this problem. Both groups spent approximately 50 minute on training.

Each group completed the same transfer task – a problem in a different context. While the training problem involved finding the capacitance of a conical capacitor, the transfer problem involved finding the resistance of a truncated pyramidal resistor. Thus, the underlying physics was very different, but the problem solving strategies were similar. We designed the training to facilitate students to develop strategies transferrable across problems solving contexts.

We found no statistically significant difference between the two groups on the baseline task, indicating that they had comparable problem solving skills before beginning their training. However, the experimental group significantly outperformed the control group (p < 0.05) on the transfer task. These results seem to suggest that the V-H sequences designed based on our framework are more effective in promoting transfer between disciplines than traditional instruction.

Authors

• N. Sanjay Rebello, Purdue University

• Lili Cui, University of Maryland - Baltimore County

• Dong-Hai Nguyen, Kansas State University