Connectionism
8 important questions on Connectionism
Propositional modularity (Ramsey, Stich, Garon, 1991)
Distributed connectionist processing does not support such propositional modularity and hence that if human minds work like such devices, then the folk vision is fundamentally inaccurate.
The most famous argument against connectionist models of human thought:
So internal representations are structured;
Connectionist models lack structured internal representations;
So connectionist models are not good models of human thought
-> Fofor & Pylyshyn
Recursive autoassociative memory (RAAM) Chalmers, 1990
- Hierarchical structures for representing information.
- Recursive processing of data to learn patterns.
- Efficient storage and retrieval of sequences.
- Applications in language understanding and cognitive science.
- Higher grades + faster learning
- Never study anything twice
- 100% sure, 100% understanding
Connectionisme vs Folk Psychology kort:
- Folk psychology: verklaart gedrag via beliefs en desires.
- Connectionisme: verklaart gedrag via activatiepatronen in neurale netwerken (subsymbolisch niveau).
- Relatie: connectionisme kan laten zien hoe folk-psychologische toestanden emergent ontstaan, maar stelt ook de vraag of die begrippen wel nodig zijn.
→ Alternatief verklaringskader naast of in plaats van folk psychology.
Connectionisme en mental causation (Clark, H4)
- Folk psychology ziet beliefs en desires als oorzaken van gedrag.
- Connectionisme laat zien dat deze oorzaken corresponderen met gedistribueerde activatiepatronen in neurale netwerken.
- Mental causation = legitiem, maar op een hoger verklaringsniveau: beliefs en desires zijn emergente oorzakelijke beschrijvingen, geen aparte “dingen” in het hoofd.
2 varieties of the computational view of mind:
2. Connectionism, parallel distributed processing and artificial neural networks
Principal Component Analysis (PCA)
- Kind of posttraining analysis Elman used in 2nd gen connectionism to determine what the network learned
- Cluster analysis stresses relations of silimilarity and difference between static states (snapshots), PCA reflects the ways in which being in one state (in a recurrent network) can promote or impede movement into future states
Dynamic representations (Elman)
The question on the page originate from the summary of the following study material:
- A unique study and practice tool
- Never study anything twice again
- Get the grades you hope for
- 100% sure, 100% understanding

















