Thoughts on Long Term Memory

I'm not going to bore my readers with a simple overview of my text's introductory chapter on long term memory (LTM from here on out). But, I'm going to write about the more provocative theories, and I am also going to talk about my own experiences with LTM: the latter should be interesting....

Since cognitive psychology uses the computer metaphor for cognitive processes, I have an easy time identifying with it! As a computer geek (well, a self-professed geek), how our own memory works, according to cognitive theory, "makes sense." However, in this chapter, we're introduced to alternative theories: luckily, they still use the lovely digital domain as their metaphor. One theory uses a computer network as its metaphor while another uses a distributed processing model metaphor.

The network theory has three basic concepts: nodes, links, and spreading activation. In a network, a node can be seen as a "router:" a unit that sorts traffic (a computer network router makes sure information gets to the right place, or at least, makes sure it goes in the right direction). A link is a physical or cognitive connection between concepts: computers having a physical connection to other computers and routers while cognitive connections are more abstract in that they link to concepts. Although, I do think that there could be analogous neurological structures in terms of nodes and links: it may not be as abstract as we think. For example, when it comes to brain injury, the brain reroutes information as best as it can to recover lost abilities or at least provide new neurological pathways to relearn lost abilities. Spreading activation simply refers to the brain's (or computer networks) ability to find information in a systematic way: when a node is activated, it can activate other nodes that are related to it (conceptually and physically), and send that information back to working memory or the computer requesting the information. I would not' t be surprise that there's a lot of collaboration, or at least there should be, between internet search engines and scientists working on cognitive function.

The distributed model metaphor for cognitive function also seems like an apt explanation. Our text makes a distinction between the networked model and connectionist model, but I fail to see the fuss. It is apparent that the connectionist model evolved from the network model: the emphasis is on processing instead of searching information. In the connectionist model there are two other concepts: parallel/serial processing and distributed representation. Like a computer with multiple cpu's is "faster" compared to one with a single cpu, the brain works better when more processes work at the same time: information storage and processing occurs at the same time or "in parallel." When searching memory, we search and explore multiple concepts in parallel, BUT, it isn't the unit of information but the strength of the connections between multiple nodes of concepts (it's "distributed") that creates meaning. Even though there is a distinction between this model and the last, it just seems that the connectionist model is just a more refined theory and the two really are not incompatible: in fact, I think they're basically the same thing.

Well, I was going to get into my own experiences with LTM, but I'll save that for my next post!