Goal Alignment
In a fully cooperative interaction, the agents’ goals are completely aligned: both evaluate outcomes using the same success criteria and seek the same end state. Any improvement in the system’s outcome is beneficial to both agents simultaneously, and no action that advances one agent’s objective can undermine the other’s. There is no incentive to withhold effort, deceive, or defect on motivational grounds, because success is jointly defined. At the level of goals, the agents function as a single evaluative unit rather than competing decision-makers.
Structural Symmetry
In a fully cooperative interaction, structural symmetry means that both agents occupy equivalent positions within the system. Each has access to the same set of actions, operates under the same constraints, and possesses comparable capacity to influence outcomes. No agent is assigned a privileged role or a dependent function; in principle, either could substitute for the other without altering the structure of the interaction. Any differences in behavior arise from choice or coordination, not from built-in asymmetries of role or capability. Structure does not impose hierarchy, delegation, or dependence.
In a fully cooperative dyad, both agents pursue the same objective and evaluate outcomes identically, while operating from structurally equivalent positions within the system. Their goals are aligned, and neither agent possesses a role-based advantage over the other: both have access to the same actions, information, and strategic possibilities. Because incentives are shared and structure is symmetric, the interaction contains no inherent conflict, hierarchy, or bargaining pressure. Strategic behavior in this category is therefore centered on coordination and joint optimization rather than persuasion, competition, or control. Success or failure emerges from how effectively the agents synchronize their actions, not from opposing interests or unequal footing.
Information Relationship
Both agents pursue the same objective and occupy equivalent strategic positions. Neither benefits from the other’s failure, and neither holds inherent structural advantage. All failure in this scenario is unintentional: it arises from information, timing, or design—not adversarial strategy.
What changes across cases is how coordination succeeds or fails depending on the information regime.
1) Pure Coordination
Fully Cooperative × Perfect + Symmetric
Pure Coordination describes interactions in which agents share identical goals, occupy equivalent structural positions, and possess complete, mutually known information about the relevant state of the world. Nothing is hidden, nothing is unequal, and nothing is adversarial. All agents see the same situation, know that the other sees it, and understand that success or failure is shared. In this regime, disagreement is not strategic—it is purely operational.
Both agents see all decision-relevant state, and both know that both see it.
Nothing important is hidden, delayed, or privately held.
- Dominant dynamic: pure coordination
- What matters: timing, synchronization, control topology
- Failure mode: miscoordination despite full agreement
- Not possible: mistrust, deception, informational leverage
Examples:
- Two people lifting a couch in full view of each other
- Paired climbers on a visible route with shared conditions
- Rowers in the same boat with full sensory feedback
- Two surgeons performing a synchronized procedure with full visibility
- Cooperative puzzle games with a fully visible board
If failure occurs here, it is a coordination or execution error, not a strategic one.
2) Guided Coordination
Fully Cooperative × Perfect + Asymmetric
Guided Coordination occurs when agents share goals and equivalent formal standing, but differ in their access to information. The world itself is fully knowable, yet one agent observes more clearly, sooner, or from a superior vantage point. Cooperation remains intact, but coordination is no longer purely mutual; it becomes mediated through the better-informed agent.
All relevant state is observable in principle, but access is unequal.
One agent sees earlier, clearer, or more directly.
- Dominant dynamic: informational dependency
- What matters: relay speed, accuracy, trust in reporting
- Failure mode: bottlenecks, delay, accidental hierarchy
- Risk: one agent becomes silently indispensable
Examples:
- A driver and navigator where only the navigator sees the map
- A rescue team where one member has drone or aerial visibility
- A pilot coordinating with crew while only the pilot sees instrument readouts
- Two engineers where only one has live diagnostics
Cooperation succeeds only if asymmetry is actively integrated rather than allowed to harden into control.
3) Joint Sensemaking
Fully Cooperative × Imperfect + Symmetric
Joint Sensemaking describes cooperative interactions in which agents share goals and structural standing but face a world that is partially unknown, noisy, or unfolding. Neither agent has privileged access to the truth; uncertainty is shared, and both know that it is shared. Cooperation is intact, but certainty is not.
Some relevant state is hidden or uncertain, and both agents face the same uncertainty.
Neither has privileged access.
- Dominant dynamic: joint inference under uncertainty
- What matters: shared models, risk tolerance, adaptability
- Failure mode: hesitation or divergent interpretation
- Risk: coordination collapse without disagreement
Examples:
- Search-and-rescue in fog or smoke
- Two scientists investigating an unknown phenomenon
- Emergency response during an unfolding natural disaster
- Joint exploration of an unmapped environment
Trust exists, but certainty does not. Success depends on how well uncertainty is managed together.
4) Trust-Mediated Cooperation
Fully Cooperative × Imperfect + Asymmetric
Trust-Mediated Cooperation occurs when agents share goals and formal standing, but uncertainty is unevenly distributed. Some relevant state is hidden or unclear, and one agent possesses superior knowledge, expertise, or insight into that uncertainty. Cooperation remains genuine, but it is filtered through reliance rather than mutual visibility.
Relevant state is uncertain, and one agent knows more or knows sooner.
Uncertainty exists, but it is unevenly distributed.
- Dominant dynamic: trust vs dependency
- What matters: disclosure discipline, verification, calibration of reliance
- Failure mode: blind obedience or quiet exploitation
- Risk: cooperation becomes control
Examples:
- A senior expert guiding a novice through a complex task
- A doctor diagnosing a patient
- A manager with full financial data directing a team
- A systems architect coordinating developers without full system context
This is the most dangerous cooperative regime: shared goals coexist with the capacity for unintentional dominance.
Structural takeaway (for Fully Cooperative)
- Perfect + Symmetric: coordination problem only
- Perfect + Asymmetric: dependency problem
- Imperfect + Symmetric: uncertainty management problem
- Imperfect + Asymmetric: trust and power problem
All four are cooperative.
Only the last two can feel adversarial—without actually being so.