Bioadaptive Interfaces - Designing for Physiological Co-Regulation in Human-AI systems
ABSTRACT
The HFE industry has seen a rise in human-AI systems leveraging biometric data for personalization. However, these systems predominantly treat the human body as a static input, overlooking the cyclical nature of physiological rhythms including stress states, hormonal variation, and respiratory patterns. This paper presents a speculative exploration of bioadaptive interfaces: leveraging biometric sensing to modulate ambient environmental cues. This supports co-regulation between user and system with an aim to support somatic awareness and reduce discomfort during varying stages of fatigue, focus, or pain. Drawing on frameworks from affective computing, somatic interaction design, and human-data interaction, we propose a future-facing design and governance framework for emotionally responsive systems.
We raise critical questions around agency, transparency, and embodied consent in next-generation AI environments.
INTRODUCTION
In recent years, the consumer health tech market has exploded with biometric products like Oura, Whoop, and Viome. These systems promise deep personalization, but often treat the human body as an input to optimize. This mechanistic framing fails to reflect the lived, fluctuating, and context-dependent experience of being in a body. Emotions, pain, hormonal cycles, and fatigue cannot be fully reduced to metrics. And yet these metrics increasingly shape how AI systems respond to us.
This paper speculates on an alternative paradigm- one where AI systems co-regulate with users in ways that acknowledge the body's cyclical patterns and aim to enhance somatic comfort, not just performance. We draw on somatic design theory (Loke & Robertson), affective loop models (Höök), and feminist HCI to outline a model for adaptive interaction that supports agency, care, and bodily literacy.
MOTIVATION AND BACKGROUND
Advances in physiological computing have made it feasible to detect stress, anticipate discomfort, and adapt environments accordingly. Yet these capabilities are rarely connected to somatic design practices, which emphasize interoception, emotional awareness, and user agency.
At the same time, emotionally adaptive technologies raise ethical concerns around manipulation, surveillance, and informed consent. Work in human-data interaction (Mortier et al.) and data governance (Selbst et al.) highlights the need for new interaction paradigms that foreground transparency, refusability, and data rights. This research bridges speculative design with these governance perspectives to ask: What should emotionally responsive AI systems feel like? And how might we govern them before they become ubiquitous?
SPECULATIVE SYSTEM CONCEPT
The proposed speculative system integrates:
– Biometric Sensing: Continuous, unobstrusive monitoring of physiological signals (e.g., HRV, breathing, cycle phase).
– Ambient Modulation: Real-time adjustment of environmental cues, including: lighting, localized warmth or cooling, posture-based haptic feedback.
- Co-regulation Mechanism: User and system mutually influence one another, forming an adaptive loop that prioritizes somatic awareness over behavioral optimization.
EXAMPLE USE CASES
– Fatigue Management: Lighting and warmth adjust to support alertness or relaxation based on detected fatigue.
– Pain Mitigation: Haptic or thermal feedback and environmental cues shift to reduce discomfort during pain episodes.
– Focus Enhancement: Subtle environmental changes help maintain or restore focus during cognitive tasks.
METHODOLOGY
This project will use a hybrid, design-led research methodology combining speculative prototyping, autoethnographic exploration, and early-stage user walkthroughs to investigate how bioadaptive systems might support human–AI co-regulation.
Autoethnographic Reflection
The researcher will document personal physiological state across stress, fatigue, hormonal cycles, and recovery periods using biometric sensing tools. These reflections will help identify recurring bodily thresholds and environments where adaptive support may be most meaningful.
Speculative Prototyping
Low-fidelity simulations will be developed to model environmental modulation such as lighting, warmth, vibration, or texture in response to physiological inputs. These will serve as testbeds for imagined interactions and user-system dynamics.
User Walkthroughs and Scenario Testing
A small group of participants will be invited to experience these speculative prototypes during staged activities, such as focused work or recovery. Their responses will be used to surface questions around control, emotional interpretation, and comfort.
Analytical Framework
Thematic analysis will be conducted on participant responses using grounded theory, focusing on three axes:
- Cognitive and emotional resonance
- Perceived agency and control
- Willingness to share physiological data
Ethical safeguards will include anonymized data capture, informed consent protocols, and participant control over data interpretation.
Why Speculative Design
Because bioadaptive systems of this kind are not yet widely deployed, speculative design provides a critical method for exploring their future implications. Rather than testing existing systems, this approach enables anticipatory thinking around emerging risks, interaction norms, and governance blind spots. By simulating and staging imagined interactions, we can surface latent assumptions, emotional reactions, and ethical concerns that might otherwise go unnoticed until after deployment.
DESIGN FRAMEWORK: PHYSIOLOGICAL CO-ADAPTATION
Dynamic Responsiveness - Adaptation occurs in sync with rhythms, not momentary inputs.
Empathetic Modulation - System aims to soothe or support, not correct.
Mutual Agency - Both users and system can initiate or veto adaptions.
Opacity and Trust - Users are informed about what is sensed and how adaptations are triggered.
Privacy and Ethics - Data is stored locally with explicit, continuous consent.
Design tensions to be explored include:
- Seamlessness vs. legibility
- Comfort vs. autonomy
- Personalization vs. standardization
KEY QUESTIONS AND GOVERNANCE IMPLICATIONS
– Agency: What forms of refusal and override should exist in emotionally adaptive systems?
– Transparency: How can system responses be made legible without disrupting user experience?
– Oversight: Should emotionally responsive AI require affective or somatic design audits?
– Equity: How do we prevent exclusion of bodies that fall outside normative biometric baselines?
– Context of Use: What rules should govern deployment of bioadaptive systems in public settings like schools or transit hubs?
CONCLUSION
This project will investigate the emerging terrain of bioadaptive interfaces - AI systems that adapt to physiological rhythms to support somatic comfort, cognitive focus, or emotional regulation. While these systems offer new opportunities for care-centered technology, they also raise urgent questions around privacy, manipulation, and bodily autonomy.
As AI becomes more ambient and intimate, governance must expand to include the dynamics of human–machine interaction itself. This research will explore how design, consent, and policy can intersect to guide the development of emotionally adaptive systems in ways that preserve agency and promote trust.
Rather than assuming optimization as the goal, this work will imagine co-regulation as a shared, situated process—one that positions AI not as a controller, but as a partner in navigating human experience.
