Semantic Pointer Architecture. Cognitive architecture created with the Neural Engineering Framework to map low level information (i.e. vision) to high level symbols (i.e. objects).
The Semantic Pointer Architecture is a cognitive architecture created with the Neural Engineering Framework. In the most general sense, it defines how low level sensory information (for example, vision) can be mapped to higher order symbols (for example, an object) and then back down again (for example, motor actions). In reality, it also helps define how this information is routed in the brain using the thalamus and basal ganglia, as well as how symbols can be combined together for knowledge representation.
It's name comes from the fact that it's functionality is similar to the "pointer" of computer science and that this pointer points to the senses. Thus, we get the term "semantic pointer".
For a high level view of the philosophical and functional motivation behind this architecture, please see "How to Build a Brain" by Chris Eliasmith or more briefly this overview.
This tag should be used for questions about the use of the Semantic Pointer Architecture in cognitive models, including Spaun.