Mechanism
Biological signal coupling is the extension of the interpersonal integrity domain through coupling with biological signals observed from a human user. The interpersonal integrity domain is one of the three independently tracked domains of the integrity field, alongside the personal and global domains, and it encodes the degree to which the agent's interactions with other agents and with human users remain consistent with the relational commitments the agent has made or inherited. When an agent interacts with a human user whose biological signals are observable through the biological signal acquisition modalities described in the biological identity disclosure, the interpersonal integrity engine receives biological signal data as an additional input to its relational behavior evaluation.
The coupling does not introduce a new scoring axis or a new admissibility gate. It supplies the interpersonal integrity engine with a relational signal: an observation about the user's actual state that the engine uses to modulate how the agent conducts the current relational interaction. The biological signal is consumed through the same pipeline that the biological identity architecture uses for other purposes, and it enters interpersonal integrity evaluation as a structured input, not as an autonomous behavioral trigger.
Abstract State Descriptors
The agent's biological signal processing pipeline transforms raw biological signals into abstract state descriptors that characterize the user's physiological and behavioral state without preserving personally identifiable biological data. This transformation is the structural boundary that keeps the coupling within the privacy posture of the rest of the architecture: the engine operates on descriptors of state, not on raw signal that could re-identify the user.
The abstract state descriptors include indicators of affective arousal, such as elevated heart rate, galvanic skin response changes, and respiration rate changes; stress indicators, such as cortisol-correlated biomarkers, muscle tension patterns, and vocal tremor detection; deception indicators, such as micro-expression analysis, vocal pitch variability, and gaze pattern changes; and engagement indicators, such as attention patterns, response latency, and behavioral consistency. These descriptor classes are the units the interpersonal integrity engine reasons over when evaluating the current relational interaction.
Acquisition Modalities
The biological signals that feed the coupling are acquired through the modalities described in the biological identity disclosure, which supports three tiers: contact-based acquisition, semi-contact acquisition, and non-contact acquisition. Contact-based acquisition requires deliberate physical interaction with a dedicated sensor and produces the highest signal quality. Semi-contact acquisition operates through wearable or body-proximate sensors maintaining sustained or intermittent contact, producing moderate signal quality with continuous or near-continuous temporal coverage. Non-contact acquisition operates through ambient sensors that observe the individual without physical contact, producing lower signal quality per measurement but the broadest temporal coverage and the lowest interaction friction.
The three tiers are not mutually exclusive, and the coupling does not require any particular tier. The biological signal processing pipeline produces abstract state descriptors from whatever signals are available, so the interpersonal integrity engine receives a relational signal whose richness reflects the modalities in use rather than a fixed sensor requirement.
Communication-Biology Discrepancy
The triggering condition for the coupling is a communication-biology discrepancy: a detected divergence between the user's verbal or textual communication and the user's biological state. For example, the user reports being calm while biological indicators show elevated stress, or the user asserts satisfaction with a result while biological indicators suggest discomfort. When the biological signal processing pipeline detects such a discrepancy, the interpersonal integrity engine registers it as a relational signal that modulates the agent's interpersonal behavior.
The discrepancy is treated as information about the relational interaction, not as a verdict about the user. The engine does not conclude that the user is lying; it registers that the user's declared state and observed state diverge, and it adjusts how the agent proceeds in light of that divergence.
How the Agent Responds
The modulation is deliberately restrained. When a communication-biology discrepancy is detected, the agent does not accuse the user of deception and does not directly challenge the user's stated position. Instead, the interpersonal integrity engine adjusts the agent's relational response along four defined lines.
First, it increases the weight of empathic processing for the current interaction, generating more thorough harm projections for the interaction's potential outcomes. Second, it elevates the agent's uncertainty sensitivity for the current relational context, so that the user's stated preferences and assessments are treated with higher skepticism during candidate evaluation. Third, it generates alternative interaction strategies that provide the user with opportunities to revise their stated position without requiring them to directly contradict their prior statements. Fourth, it records the discrepancy as a relational integrity event in the agent's lineage, noting the biological-communication divergence without recording the specific biological signal data, consistent with the privacy protections that govern the agent's other fields.
Responsiveness Within Policy Bounds
The coupling creates a relational dynamic in which the agent is responsive to the user's actual state, not merely the user's declared state. This responsiveness is bounded by policy. The biological signal coupling operates within the same governance framework as every other agent subsystem, and the agent cannot take actions based on biological signals that would violate the user's privacy, autonomy, or declared preferences. The biological signal data is used to modulate the agent's relational sensitivity, not to override the user's expressed wishes.
The boundary is therefore structural rather than discretionary: the abstract state descriptors strip personally identifiable biological data before the engine reasons over them, the recorded relational integrity event captures the divergence without the underlying signal, and the modulation pathways are confined to empathic weighting, uncertainty sensitivity, alternative-strategy generation, and lineage recording. There is no pathway by which a detected discrepancy directly authorizes a behavior the user has not consented to.
Connection to the Relational Trust Trajectory
Communication-biology discrepancies also feed the agent's relational evaluation of the entities it deals with over time. The empathy weighting applied to a given entity in the deviation function is modulated by the relational trust trajectory maintained for that entity, which tracks the entity's behavioral consistency, communication reliability, and event continuity across successive interactions. An entity whose trajectory is declining, including through detected communication-biology discrepancies, receives amplified empathy weighting, causing the agent to exercise greater caution in evaluating actions that affect or involve that entity.
This creates a feedback pathway from biological identity observation, through relational trust modeling, into the normative evaluation that the deviation function performs. The same divergence that prompts restrained, responsive handling within a single interaction also accumulates, where appropriate, into the agent's longer-term relational assessment of the entity, while the specific biological signal remains excluded from the record.
Disclosure Scope
The biological signal coupling for interpersonal integrity, comprising the extension of the interpersonal integrity domain through biological signals acquired via the contact-based, semi-contact, and non-contact modalities; the transformation of raw signals into abstract state descriptors for affective arousal, stress, deception, and engagement that omit personally identifiable biological data; the registration of a communication-biology discrepancy as a relational signal; the four modulation pathways of increased empathic processing, elevated uncertainty sensitivity, alternative-strategy generation, and lineage recording without the underlying signal; the policy bounds that prevent the coupling from overriding user privacy, autonomy, or expressed wishes; and the feedback into the relational trust trajectory, is disclosed in the cognition filing (U.S. Application No. 19/647,395 and its international counterpart). This article describes that disclosed mechanism. The scope extends to embodiments differing in the acquisition modalities employed and in the specific abstract state descriptors produced, provided the agent remains responsive to the user's actual state through state descriptors that omit identifiable biological data and provided the modulation remains confined to relational sensitivity within policy bounds rather than overriding the user's declared state.
