Overview
This thesis investigates how temporal and layered interface design can help college women athletes interpret their own performance fluctuations — specifically how training load and menstrual cycle variability interact over time to shape their physical state.
The project responds to that gap by designing a system that makes AI reasoning interpretable for athletes rather than coaches or technicians — centering the athlete as the primary decision-maker in understanding her own body.
Problem
Research continues to reveal strong links between menstrual-cycle phases and women's athletic performance. However, most existing systems still provide limited support for interpreting how training load and physiological variability interact. At the same time, interactive systems are often designed as gender-neutral, overlooking women's lived physiological experiences.
Research
Talking to athletes
- Former D2 softball athlete
- Non-college ski trainer
- Active university basketball athlete
- Former university basketball athlete
What they shared
- Athletes navigate a tension between data and bodily experience.
- Wearable metrics are widely collected but rarely explained, leaving athletes confused when data doesn't match how they feel.
- Cycle fluctuations shape training capacity, yet athletes often feel pressure to ignore them — so generic advice erodes trust.
- Athletes prefer explanations over automated decisions, to stay in control.
Precedents
Current training platforms surface large amounts of data, but rarely explain why it matters or how it connects to the athlete's lived experience.
Research Question
How can a layered, temporal interface for an AI-supported training system help college women athletes interpret their own performance — by showing how training load and physiological data interact?
Make the link between objective training load and internal experience perceptible through multi-modal visualization.
Reveal why a training state occurred, through layered explanations of training and cycle phase.
Show how difficulties emerge across days by surfacing temporal patterns in load and cycle variability.
Concept
The system connects external training load with internal bodily experience over time. Through integrated visualization and layered explanations, it helps athletes understand fluctuations, uncover underlying causes, and navigate future readiness.
External training load
Objectively measurable demands — duration, distance, volume — applied during training or competition.
Internal experience
Subjective responses — perceived effort, stress, strain. Body language is a key signal of these states.
Design
Layered explanation — daily context
Unpacks why today's session felt different. Selecting a score such as Training Load or Readiness opens the factors behind it.
Layered explanation — pattern over time
Extends the explanation across days: filters surface flagged patterns, and a chat assistant explains why those states occurred and how to recover.
Multi-scale temporal navigation
Moves from a single session to 7, 14 and 30-day patterns, with a bird's-eye view and time filters for quick exploration.
