Reference Number: ARL-SRP-00059
Project Description
How can we measure whether people and intelligent systems are adapting effectively to one another? This project will investigate Complexity Synchronization, a data-driven framework for quantifying coordinated changes across human teammates and artificial intelligence (AI) agents. The goal is to determine whether patterns of synchronization can provide useful indicators of team coordination, trust, workload, stress, or impending performance problems.
The student will contribute to a pilot study using either a human–agent or multi-human teaming paradigm. Activities may include helping design experimental tasks; preparing, synchronizing, and quality-checking multimodal time-series data; analyzing human behavioral and physiological signals; and examining AI behavior, decision variables, or adaptation parameters. Depending on the study design, data may include EEG or fNIRS, cardiovascular and respiratory measures, eye tracking, movement or posture, surface electromyography, keystrokes, button presses, and task-performance measures.
Students may use Python or MATLAB; signal-processing, statistical, and machine-learning methods; time-series analysis; visualization tools; and techniques such as detrended fluctuation analysis, diffusion entropy analysis, cross-correlation, and dynamic network analysis. Students from computer science, electrical or biomedical engineering, physiology, neuroscience, psychology, human–computer interaction, data science, applied mathematics, or related fields are encouraged to apply.
The expected outcome is a documented analysis pipeline and preliminary results showing how complexity-based measures relate to human and team behavior. Deliverables may include software, visualizations, a technical report, and a presentation of findings. Prior experience with physiological data, human-subjects research, or AI systems is helpful but not required; training will be provided.
Technical Skills
- Artificial Intelligence/Machine Learning
- Biological Sciences/Engineering
- Computer Science
- Data Analysis
- Human Factors Engineering/Human Systems Integration
- Statistics
