The Role of Situational Interest in Personalized Learning

• In Event: 41.023 - Exploring Links Between Interest and Learning

Abstract

Objectives and Framework
Consumer demand and national educational technology policies have sparked a proliferation of adaptive technologies that individualize learning based on cognitive factors like learners’ prior knowledge (US Department of Education, 2010, 2016). A second wave of technology now adapts learning based on non-cognitive factors – like students’ individual interests in out-of-school topics like sports or music – to personalize the learning experience. This presentation examines ways that personalized learning environments trigger students’ situational interest, and implications this interest has on learning outcomes.

Context personalization – the incorporation of students’ individual interests into learning tasks – has been shown to positively affect students’ learning and performance in mathematics (Walkington, 2013). Personalization may achieve its effects on learning via cognitive processes – such as grounding learning in familiar contexts (Goldstone and Son, 2005) – as well as motivational processes, like elicitation of situational interest (Hidi & Renninger, 2006). However, the mechanisms through which personalization impacts learning have yet to be identified. Incorporating students’ out-of-school interests is theorized to leverage enduring individual interests in order to trigger and maintain situational interest in school tasks. This study examines whether personalizing four units of algebra problems in an intelligent tutoring system (ITS) impacts in-tutor performance, performance on classroom exams, situational interest in the curriculum, and individual interest in mathematics. We also examine whether situational interest mediates the effect of personalization on outcomes.

Data Sources and Methods
Data from 150 (50% female) White, rural, high school Algebra I students who learned linear equations using an ITS are used to model the effects of personalization. All students completed surveys of their out-of-school interests and were randomly assigned to a standard curriculum or one that presented matched problems personalized to incorporate interests like games, sports, and food into problems. After each unit, students completed surveys of situational interest (Linnenbrink-Garcia, Durik, Conley, Tauer, Karabenick, & Harackiewicz, 2010), which were examined in conjunction with behavior logs recorded by the software. Students completed a final survey of individual interest in math (Linnenbrink-Garcia, et al., 2010) and a paper-based algebra test in the classroom six months after beginning ITS use.

Results and Significance
Path analysis indicates that students in the personalization condition reported greater triggered situational interest in the curriculum than the control condition, and that triggered interest predicted in-tutor outcomes (accuracy, learning efficiency; Figure 1), as well as maintained situational interest. A total effect of personalization was observed on classroom exam performance and individual interest in mathematics.

Interest triggered by personalization explained a significant proportion of variance in student performance, and deepened students’ individual interest in math. Interest is thus a key mechanism of context personalization (Walkington & Bernacki, 2014), and context personalization contributes to interest development (Hidi & Renninger, 2006). These findings provide support for the four-phase model of interest development, and recommend personalization as a practical intervention that “kicks off” the interest development cycle in a notoriously difficult subject area – high school mathematics. Because personalization can accommodate and activate interests of students from any social class, ethnicity, or culture, its potential is considerable.

Authors

• Matthew L. Bernacki, University of Nevada - Las Vegas

• Candace A. Walkington, Southern Methodist University