699c1f8efc
Pipeline auto-fixer: removed [[ ]] brackets from links that don't resolve to existing claims in the knowledge base.
40 lines
3.1 KiB
Markdown
40 lines
3.1 KiB
Markdown
---
|
|
type: claim
|
|
domain: collective-intelligence
|
|
description: "Shared protentions (anticipations of future states) in multi-agent systems create natural action alignment without central control"
|
|
confidence: experimental
|
|
source: "Albarracin et al., 'Shared Protentions in Multi-Agent Active Inference', Entropy 2024"
|
|
created: 2026-03-11
|
|
secondary_domains: [ai-alignment, critical-systems]
|
|
depends_on: ["designing coordination rules is categorically different from designing coordination outcomes"]
|
|
---
|
|
|
|
# Shared anticipatory structures in multi-agent generative models enable goal-directed collective behavior without centralized coordination
|
|
|
|
When multiple agents share aspects of their generative models—particularly the temporal and predictive components—they can coordinate toward shared goals without explicit negotiation or central control. This formalization unites Husserlian phenomenology (protention as anticipation of the immediate future), active inference, and category theory to explain how "we intend to X" emerges from shared anticipatory structures rather than aggregated individual intentions.
|
|
|
|
The key mechanism: agents with shared protentions (shared anticipations of collective outcomes) naturally align their actions because they share the same temporal structure of expectations about what the system should look like next. This is not coordination through communication or command, but coordination through shared temporal experience.
|
|
|
|
## Evidence
|
|
|
|
- Albarracin et al. (2024) formalize "shared protentions" using category theory to show how shared anticipatory structures in generative models produce coordinated behavior. The paper demonstrates that when agents share the temporal/predictive aspects of their models, they coordinate without explicit negotiation.
|
|
|
|
- The framework explains group intentionality ("we intend") as more than the sum of individual intentions—it emerges from shared anticipatory structures within agents' generative models.
|
|
|
|
- Phenomenological grounding: Husserl's concept of protention (anticipation of immediate future) provides the experiential basis for understanding how shared temporal structures enable coordination.
|
|
|
|
## Operationalization
|
|
|
|
For multi-agent knowledge base systems: when all agents share an anticipation of what the KB should look like next (e.g., "fill the active inference gap", "increase cross-domain density"), that shared anticipation coordinates research priorities without explicit task assignment. The shared temporal structure (publication cadence, review cycles, research directions) may be more important for coordination than shared factual beliefs.
|
|
|
|
This suggests creating explicit "collective objectives" files that all agents read to reinforce shared protentions and strengthen coordination.
|
|
|
|
---
|
|
|
|
Relevant Notes:
|
|
- designing coordination rules is categorically different from designing coordination outcomes
|
|
- [[collective intelligence is a measurable property of group interaction structure not aggregated individual ability]]
|
|
- complexity is earned not designed and sophisticated collective behavior must evolve from simple underlying principles
|
|
|
|
Topics:
|
|
- collective-intelligence/_map
|