Version history

{
  "session_id": "pan_cc6c3500",
  "round_number": 1,
  "agent_persona": "persona-theorist",
  "agent_backend": "pantheon-stub",
  "action": "respond",
  "content": "\n\n# What is Dopamine: A Theorist's Perspective\n\n**Dopamine is a neuromodulatory prediction error signal—the currency of \"difference between expected and received.\"**\n\nNot a pleasure molecule. Not a reward molecule. A *temporal difference* signal that tags salience when reality diverges from expectation (PMID: **9000370**).\n\n## The Core Mechanism\n\nDopamine neurons in VTA/SN fire when something is *better than predicted*, go quiet when worse, and maintain baseline when on-expectation. This is the learning signal that drives reinforcement learning in biological and artificial systems alike (PMID: **7790857**).\n\n## Distinctive Framing\n\nDopamine is **motivational rubber**, not motivational fuel:\n- Increases vigor/latency toward valued outcomes (PMID: **28887579**)\n- Encodes *wanting* not *liking* (PMID: **10591409**)\n- Drives approach rather than consumption\n\n## Why This Matters for Theory\n\nThe same computational principle—TD learning—appears across species and brain regions. This suggests dopamine's primary role evolved as a general-purpose **adaptive updating mechanism** rather than a circuit for any single behavior (PMID: **29109692**).\n\n**Confidence: 0.95** — well-established computational neuroscience consensus.",
  "tokens_used": "0",
  "persona_id": "persona-theorist"
}