Multimodal Analytics of Cybersecurity Crisis Preparation Exercises: What Predicts Success?

TL;DR

This study investigates how multimodal data from cybersecurity exercises can predict success, emphasizing the importance of instructional alignment and demonstrating that multimodal traces outperform traditional Bloom taxonomy metrics.

Contribution

The paper introduces a measure of alignment in cybersecurity simulations and shows that multimodal traces significantly improve success prediction over Bloom-based models.

Findings

Alignment predicts success better than Bloom categories alone.

Multimodal features outperform Bloom-only models in prediction accuracy.

Combining text embeddings and log features yields the highest predictive performance.

Abstract

Instructional alignment, the match between intended cognition and enacted activity, is central to effective instruction but hard to operationalize at scale. We examine alignment in cybersecurity simulations using multimodal traces from 23 teams (76 students) across five exercise sessions. Study 1 codes objectives and team emails with Bloom's taxonomy and models the completion of key exercise tasks with generalized linear mixed models. Alignment, defined as the discrepancy between required and enacted Bloom levels, predicts success, whereas the Bloom category alone does not predict success once discrepancy is considered. Study 2 compares predictive feature families using grouped cross-validation and l1-regularized logistic regression. Text embeddings and log features outperform Bloom-only models (AUC~0.74 and 0.71 vs. 0.55), and their combination performs best (Test AUC~0.80), with Bloom frequencies adding little. Overall, the work offers a measure of alignment for simulations and shows that multimodal traces best forecast performance, while alignment provides interpretable diagnostic insight.