The dynamic interplay between goal setting, performance, and emotions in self-regulated learning: A computational modeling approach.

Citation

Lin WM, FitzGibbon L, Theobald M, Breitwieser J, Brod G, Murayama K & Sakaki M (2026) The dynamic interplay between goal setting, performance, and emotions in self-regulated learning: A computational modeling approach.. Journal of Educational Psychology, 118 (5). https://doi.org/10.1037/edu0001022

Abstract
In self-regulated learning (SRL), students often set goals that influence how they subsequently perform on tasks. Furthermore, emotions have been considered to play a key role in this process; yet, the exact dynamic relationship between goals, performance, and emotions has not been specified. In the present study, we employed computational modeling to delineate the specific dynamic interplay of goals, performance, and emotions in multiple goal-striving episodes. We developed and applied our computational model of SRL to data collected from an online math task (Study 1) and from an online learning app used to study for a high-stakes state exam in Germany (Study 2). Across both studies, we found that students who set higher goals (compared with their previous performance) had higher subsequent performance, highlighting the importance of setting high goals for learning. Furthermore, emotions are not only influenced by previous goals and performance, but they also influence subsequent goal setting and performance. We found that stronger positive emotions (particularly enjoyment) predicted higher levels of goals, whereas predicting lower performance in the online math task and higher performance when studying for the high-stakes exam. Our work highlights computational modeling as a valuable tool to theorize and empirically analyze the processes of SRL in education research.

Keywords
self-regulated learning; goals; emotions; computational modeling; bayesian inference

Journal
Journal of Educational Psychology: Volume 118, Issue 5

• The dynamic interplay between goal setting, performance, and emotions in self-regulated learning: A computational modeling approach