Affect state
A persistent vector representing the system's current emotional context across dopamine-like (reward sensitivity), norepinephrine-like (urgency and arousal), and serotonin-like (stability) channels. Affect state modulates attention salience and reinforcement weighting in real time.
Affect state is not a simulation of emotion for its own sake - it is a practical mechanism for making salience and reinforcement context-sensitive. A flat salience model treats every incoming signal identically regardless of the system's recent history; affect state breaks that flatness. When the dopamine-like channel is elevated after a string of successful outcomes, novel high-reward signals get amplified attention. When the norepinephrine-like channel spikes in response to detected risk, urgency dimensions dominate salience scoring and the budget engine compresses deliberation. When the serotonin-like channel is high after sustained stability, the system favors established patterns over exploration. The three channels are updated continuously from environmental signals and decay toward baseline without stimulation. Affect state is stored as part of the cognitive session and is auditable - operators can trace how the system's emotional context shifted during a session and how that shift influenced the attention and reinforcement decisions made in that window.