Local vs. Distributed Coding
Bibliographic Data
| ID | 14815704 |
|---|---|
| Authors | Simon Thorpe, Simon J Thorpe (0000-0003-4997-3367, corresponding author) |
| Year | 1989 |
| Volume | 8 |
| Issue | 2 |
| Pages | 3-40 |
| Publication date | 1989-01-01 |
| Peer Reviewed | Yes |
| Open Access | No |
| Type | ARTICLE |
| Venue | Intellectica Revue de l Association pour la Recherche Cognitive (JOURNAL) |
| Journal identifiers | ISSN: 0769-4113 • E-ISSN: 2273-9262 |
| Publisher | PERSEE Program (PUBLISHER) |
| DOI | 10.3406/intel.1989.873 |
| OpenAlex | W2900442071 |
| Language | EN |
| Citations received | 1 |
The question of how information can be represented in networks of neurons is central to cognitive science - especially so now that some researchers are turning to connectionist models in an attempt to understand human cognition. A major distinction can be made between local coding schemes, in which each unit is assigned a particular meaning, and distributed coding, in which information is encoded by large numbers of units, each of which is involved in representing many different concepts. Many brain scientists dislike local coding schemes, but it is argued here that the arguments generally invoked against local coding are far from convincing. There are situations where it has been shown that the brain does use a form of distributed coding, often referred to as coarse coding, which is particularly useful for representing continuous variables during sensory and motor processing. However, it may be appropriate to think of the brain as a hybrid system, using more local codes for higher level representation. Such a view has the advantage of allowing more conventional symbol manipulation based artificial intelligence approaches to be integrated into a connectionist framework
Artificial neural network · Cognition · Cognitive psychology · Cognitive science · Connectionism · Information processing · Neural coding · Cognitive Science and Education Research · Computer Science · Mathematics · Neural dynamics and brain function · Neural Networks and Applications · Neuroscience · Psychology · Artificial Intelligence · Theoretical Computer Science
| Unique citing works | 1 |
|---|---|
| Citations per year | 1 |
| Citation span | 2025 - 2025 (1) |
| Citation velocity | recent |
| Highly cited | No |