Abstract
Attention is a cornerstone of effective learning, yet students engage in mind wandering—task-unrelated thoughts—approximately 30% of the time. While mind wandering often degrades learning outcomes, its relationship with Self-Regulated Learning (SRL) remains under-explored. This systematic review analyzes 134 studies (2010–2024) to bridge this gap, situating research within the Self-regulated learning process, Multimodal data, and Analysis (SMA) grid. The findings reveal a heavy reliance on self-reports and a significant research "blind spot" regarding the affective and motivational dimensions of mind wandering.
Introduction: The Challenge of Sustained Attention
In educational settings, maintaining focus is a self-regulatory challenge. Mind wandering (MW), or the shift of attention from a primary task to internal thoughts, is a pervasive phenomenon. Despite its impact, MW research has traditionally been siloed from SRL frameworks, which categorize learning through Cognitive, Affective, Metacognitive, and Motivational (CAMM) processes. This review seeks to integrate these fields, exploring how MW can be understood not just as an attentional failure, but as a potential regulatory mechanism.
Methodology
The authors conducted a preregistered systematic review following PRISMA guidelines. To manage an initial pool of 4,009 reports, they employed an AI-assisted "Researcher-in-the-Loop" process using the ASReview tool, utilizing models like TF-IDF/Naive Bayes and SBERT/Random Forest. This narrowed the selection to 134 empirical studies. These studies were then mapped onto the SMA grid to evaluate their analytical depth:
- Unimodal: Single data stream, no SRL context.
- Horizontal: Single data stream integrated with SRL processes.
- Vertical: Multiple data streams (multimodal) without SRL.
- Integrated: Multimodal data combined with SRL processes.
Key Insights: Definitions and Meta-Awareness
A critical finding is the lack of nuance in how MW is defined. While 84% of studies use the broad "Task-Unrelated Thought" (TUT) definition, very few distinguish between intentional and unintentional MW (10%) or account for meta-awareness (2%). From an SRL perspective, this distinction is vital: unaware MW represents a failure in metacognitive monitoring, whereas intentional MW might be a strategy to cope with boredom or high cognitive load.
The SMA Grid: Mapping the Research Landscape
The review highlights a significant imbalance in how MW is studied in relation to SRL:
- Cognitive and Metacognitive Focus: 16% and 11% of studies respectively linked MW to these processes.
- The Affective/Motivational Blind Spot: Only 2% of studies addressed motivation, and 0% addressed affective processes.
- Data Streams: Self-reports remain dominant (81%), though eye-tracking (26%) and other physiological sensors are increasing.
- Machine Learning (ML): 27% of studies used ML for automated detection. Multimodal models (e.g., combining eye-tracking and physiology) showed up to an 11% performance boost over unimodal approaches.
Analytical Approaches
The mapping revealed that the majority of research (59%) is unimodal. Only 5% of the reviewed literature achieved an "integrated" approach—combining multimodal data with SRL frameworks. This suggests that while we are getting better at detecting mind wandering through sensors, we are not yet effectively connecting those detections to the underlying psychological processes of the learner.
Conclusion and Future Directions
To advance the field, future research must move beyond simple detection. Key recommendations include:
- Refining Definitions: Incorporating meta-awareness and intentionality into MW models.
- Addressing the CAMM Gap: Investigating how MW interacts with emotions (affect) and goals (motivation).
- Developing Attention-Aware Systems: Leveraging multimodal ML to create adaptive educational technologies that support real-time self-regulation without disrupting the learner's agency.
By integrating MW into SRL frameworks, researchers can develop more sophisticated interventions that help students navigate the inevitable fluctuations of attention during learning.
Summary updated