Understanding students' emotions and their interplay with motivation, behavior, and learning in game-based learning environments Andres Felipe Zambrano Jacobo

Format:

Book

Thesis/Dissertation

Author/Creator:

Zambrano Jacobo, Andres Felipe, author.

Contributor:

University of Pennsylvania. Education., degree granting institution.

Language:

English

Subjects (All):

Educational technology.

Educational psychology.

Education.

0710.

0525.

0515.

Local Subjects:

Educational technology.

Educational psychology.

Education.

0710.

0525.

0515.

Genre:

Academic theses

Physical Description:

1 online resource (211 pages)

Contained In:

Dissertations Abstracts International 87-12A

Place of Publication:

Ann Arbor : ProQuest Dissertations and Theses, 2026

Language Note:

English

Summary:

Educational games are increasingly recognized as powerful environments for fostering meaningful learning. However, one dimension that remains relatively underexplored in game-based learning research is the emotional one. While prior studies have shown positive associations between certain emotions (e.g., joy, happiness, and engaged concentration) and learning during gameplay, less is known about the mechanisms through which these emotions emerge and influence learning. Most existing research relies on sparse self-reports or classroom observations that lack the temporal granularity needed to align emotions with specific in-game behaviors or to analyze the persistence and transitions among affective states. Other studies have employed commercial facial recognition software, which, although capable of capturing fine-grained affective data, presents ethical, logistical, and validity concerns due to biases in training data, high cost, and the potential to alter students' emotional experiences. To address these gaps, this dissertation leverages machine learning-based detectors trained on the target student population to investigate how emotions interact with motivation, behavior, and learning in Crystal Island, a narrative-driven microbiology game. Two suites of sensor-free affect detectors were developed based on self-reports and classroom observations. The same interaction logs were then analyzed using Ordered Network Analysis and clustering techniques to identify four behavioral archetypes-Scanners, Worksheet Users, Conversers, and Roamers-representing distinct engagement patterns. Across five studies, the dissertation examines how affective and behavioral patterns interact with motivational factors and influence learning outcomes. Findings reveal that students with higher situational interest and self-efficacy tended to sustain positive emotions for longer periods. In addition, students who engaged in self-regulatory behaviors also experienced more episodes of confusion and frustration but transitioned more frequently from these emotional states to positive ones, patterns that are linked to productive struggle, effective emotional regulation, and greater learning gains. In contrast, students lacking these characteristics tended to remain in boredom or other negative states and failed to achieve the intended learning goals of the experience. Overall, this dissertation connects affective, motivational, behavioral, and cognitive processes in game-based learning environments, offering both methodological innovations and theoretical insights into how emotions shape learning during gameplay

Notes:

Source: Dissertations Abstracts International, Volume: 87-12, Section: A.

Advisors: Isotani, Seiji Committee members: Baker, Ryan S.; Jiang, Shiyan

Ph.D. University of Pennsylvania 2026

Vendor supplied data

Local Notes:

School code: 0175

ISBN:

9798247973287

Access Restriction:

Restricted for use by site license