Search for: All records

While existing student modeling methods focus on predicting students’ knowledge states, they often overlook the underlying cognitive processes contributing to learning. In this work, we integrate cognitive processes, specifically phases of rule learning, into student modeling, drawing inspiration from cognitive science. Rule learning involves rule search, discovery, and following, providing a systematic framework for understanding how individuals acquire and apply knowledge. We conduct two studies to explore rule learning phases in a real-world learning context. Moreover, we present a two-step approach to first predict the phases of rule learning students experience during problem solving with an intelligent tutoring system and then estimate the time spent on each predicted phase. Furthermore, we identify the relationships between the time spent on specific phases of rule learning and student performance. Our findings underscore the importance of integrating cognitive processes into student modeling for more targeted interventions and personalized support. 

The dual mechanisms of control framework describes two modes of goal-directed behavior: proactive control (goal maintenance) and reactive control (goal activation on task demands). Although these mechanisms are relevant to learner behaviors during interaction with intelligent tutoring systems (ITS), their relation to ITSs is under-researched. We propose a manipulation to induce proactive or reactive control during interaction with an online tutoring system. We present two experiments where students solved problems using either proactive or reactive control. Study 1 validates the manipulation by investigating behavioral measures that reflect usage of the intended strategy and assesses whether either mode impacted learning. Study 2 investigates if alternating between control modes during problem solving affects student performance. 

Response time has been used as an important predictor of student performance in various models. Much of this work is based on the hypothesis that if students respond to a problem step too quickly or too slowly, they are most likely to be unsuccessful in that step. However, something that is less explored is that students may cycle through different states within a single response time and the time spent in those states may have separate effects on students’ performance. The core hypothesis of this work is that identifying the different states and estimating how much time is devoted to them in a single response time period will help us predict student performance more accurately. In this work, we de-compose response time into meaningful subcategories that can be indicative of helpful or harmful cognitive states. We then show how a model that is using these subcategories as predictors instead of response time as a whole outperforms both a linear and a non-linear baseline model. 

Cognitive control and rule learning are two important mechanisms that explain how goals influence behavior and how knowledge is acquired. These mechanisms are studied heavily in cognitive science literature within highly controlled tasks to understand human cognition. Although they are closely linked to the student behaviors that are often studied within intelligent tutoring systems (ITS), their direct effects on learning have not been explored. Understanding these underlying cognitive mechanisms of beneficial and harmful student behaviors can provide deeper insight into detecting such behaviors and improve predictive models of student learning. In this paper, we present a thinkaloud study where we asked students to narrate their thought processes while solving probability problems in ASSISTments. Students are randomly assigned to one of two conditions that are designed to induce the two modes of cognitive control based on the Dual Mechanisms of Control framework. We also observe how the students go through the phases of rule learning as defined in a rule learning paradigm. We discuss the effects of these different mechanisms on learning, and how the information they provide can be used in student modeling.