The Functional Roles of Cognition and Emotion in Predicting Clinical Reasoning Performance

• In Event: 35.031 - Multimodal Data During Learning With Advanced Learning Technologies: What Does Evidence Reveal About Self-Regulated Learning?

Abstract

Self-regulated learning (SRL) refers to the thoughts, feelings and actions taken and adjusted to achieve one’s learning goals (Pintrich, 2000; Zimmerman, 2000). Since SRL is multi-componential, involving cognitive, affective, metacognitive, and motivational processes, these components must be considered jointly when examining learning and performance outcomes. Specifically cognition and emotions are attached to achievement activities or outcomes (Pekrun & Linnenbrink-Garcia, 2014) and have a critical role in determining medical students’ performances (Lajoie, Zheng & Li, 2018). This research explores the predictive value that cognitive and affective components play in medical students’ ability to make accurate and efficient clinical diagnoses in a computer based learning environment, BioWorld (2009). By using multi-channels of data (behavioral, verbal protocols, self-report), we examine how participants’ cognitive processes and achievement emotions jointly affect their diagnostic efficacy and confidence.

Secondary analysis of data collected from a prior study (Naismith & Lajoie, 2017) were used in this research. Participants (27 medical students from a North American university) volunteered to participate in this 2-hour study and solved three patient cases in BioWorld. Students were instructed to think aloud as they solved cases. An adapted Achievement Emotion Questionnaire (AEQ) for medical students was administered after each case and measured participants’ retrospective emotions (pride, relief, joy, shame, anger) (Naismith & Lajoie, 2017).

BioWorld log files of participants’ cognitive activities (e.g., order lab test, search library, add hypothesis) were analysed for each case. Participants’ cognitive tentativeness and cognitive certainty were inferred based on their think-aloud data using the Linguistic Inquiry and Word Count (LIWC) (Pennebaker, Boyd, Jordan, & Blackburn, 2015), a text mining program that provides an efficient method for studying participants’ cognitive components by comparing participants’ verbal samples with its psychologically meaningful categories (Tausczik & Pennebaker, 2010).

Considering the dependence between patient cases as each participant solved three cases, we performed linear mixed-effects models (LMMs) to test the influence of cognitive activity, cognitive-motivational processes (tentativeness, certainty), and achievement emotions (i.e. pride, anger, shame, relief and enjoyment) on participants’ diagnostic efficacy and confidence.

Both the number of cognitive activities participants engaged in and their levels of cognitive certainty positively predicted diagnostic efficacy. Cognitive tentativeness negatively predicted diagnostic efficacy. Furthermore, the achievement emotion of pride positively predicted efficacy, while relief negatively predicted efficacy (see Table 1 and Table 2).

In terms of confidence, the pride participants experienced positively predicted their confidence in diagnosing the case correctly, while the shame negatively predicted their confidence marginally. However, participants’ cognitive processes, including the number of cognitive activities, levels of cognitive tentativeness and certainty, did not predict confidence (see Table 3 and Table 4).

Cognitive activities directly predicted diagnostic efficacy but oddly enough did not predict confidence. Even when students perform well in the moment they are still not confident in their diagnoses. Achievement emotion of pride predicted both diagnostic efficacy and confidence. The relationship between cognition and affect needs further exploration to ensure that learners can accurately judge their performances in the self-reflection phase of SRL.

Authors

• Susanne P. Lajoie, McGill University

• Shan Li, McGill University

• Juan Zheng, McGill University