Improving students' diagnostic problem solving performance with expert support in a computer-based learning environment

Abstract

Problem solving has been pervasively used as an important method for learning and expertise development, particularly in such complex domains as medical education, where problem-based learning (PBL) has been widely adopted to help students to acquire knowledge through autonomous learning involving diagnostic problems. However, PBL is limited in its ability to help students to develop systemic knowledge structure and efficient reasoning processes, which may impede their diagnostic problem-solving performance and professional competency. The purpose of this study is to address the challenges of PBL by scaffolding students’ problem-solving and knowledge-construction processes, through the design of a computer-based learning environment supported by expert knowledge. Based on a design-based research approach, two rounds of design and evaluation were conducted to investigate how a computer-based, expert-supported learning environment could be designed to improve students’ diagnostic problem-solving, and to assess the effects of the designed learning environment on students’ performance. A computer-based, expert-supported learning environment was designed and developed that helped students to identify the gap between their performance and that of experts when they engaged in a number of problem-solving tasks in a simulated environment. To facilitate effective thinking and reflection during learning with problems, mental models were incorporated into the learning environment to help students to externalize the intellectual processes of problem-solving and knowledge-construction intertwined with complex learning activities. The effects of the proposed learning environment were examined in light of students’ perceptions of the designed learning environment, their performance on diagnostic problem-solving and subject-matter knowledge construction, and their professional diagnostic problem-solving capabilities. Medical students from a public medical college in southern mainland China participated in the study to learn to solve diagnostic problems using the designed learning environment. Data were collected from an experiment that compared the differences in learning outcomes between the experimental group and control group. The measures and instruments used include questionnaire surveys, pre- and post-tests, and learning products made by students during the learning process. Descriptive statistical analysis, correlation analysis, Mann-Whitney U tests, paired-sample t-tests, independent-sample t-tests, and content analysis were used to analyze the data. The evaluation results show that the students in the experimental group (i.e., those using the proposed learning environment with expert support) showed more positive perceptions toward the designed learning environment than those in the control group (i.e., those using another learning environment without expert support). In addition, students in the experimental group outperformed those in the control group in problem-solving and knowledge-construction performances, and in their professional capabilities for diagnostic problem-solving, though they showed no significant difference on traditional knowledge tests. The results of this study clearly showed that a computer-based expert-supported learning environment can be designed and developed to support students' complex learning in problem-based learning contexts.