HOT-1: Generative, top-down or noisy perception modules

Indicator: multiple sources of perceptual activity including top-down and noise required for reality monitoring.

Neighborhood — ranked by edge-count

Papers (2)

paper

Frameworks (1)

framework

Perceptual Reality Monitoring Theory (PRM)
associated_with
A computational HOT claiming that consciousness depends on metacognitive monitoring distinguishing reliable perceptual representations from noise.

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

HOT-2: Metacognitive monitoring distinguishing reliable perceptual representations from noiseconcept0.806
Indicator: reality monitoring mechanism assigning veracity probabilities.
HOT-3: Agency guided by a general belief-formation system, with strong disposition to update beliefs per monitoring outputsconcept0.752
Indicator: consumer system that relies on 'real' tagged perceptions for rational action.
HOT-3: Agency guided by a general belief-formation and action-selection system, and strong disposition to update beliefs in accordance with metacognitive monitoring outputsconcept0.752
Third HOT indicator; builds on HOT-2; first clause entails AE-1 and connects to predictive processing
HOT-4: Sparse and smooth coding generating a 'quality space'concept0.738
Indicator: representation format providing qualitative character via similarity/discriminability.
RPT-2: Input modules generating organized, integrated perceptual representationsconcept0.721
Indicator derived from recurrent processing theory; requires interpretability methods or behavioral tests like Kanizsa illusion
Models with 1-hot activation sparsity still have polysemantic neurons; single neuron trained on 4 mutually exclusive features prefers polysemantic representation with loss ~0.7 vs 0.8finding0.712
Counter-example disproving that architectural sparsity alone can prevent polysemanticity
There is a many-to-many mapping between neurons and concepts, meaning multiple high-level causal variables might be encoded in overlapping groups of neuronsclaim0.706
Fundamental theoretical claim motivating DAS, attributed to Smolensky/Rumelhart/McClelland.
It's tricky, because for a typical language model the entity is sort of tricameral: the base simulator, the simulated simulator, and the simulated awareness.quote0.706
Antra's earlier definitive statement of the tricameral model.