Human beings seem to have a predisposition to form categories and to categorise the world around tham. This goes back to Aristotle’s writings and probably before, and has been expanded upon since that time in many disciplines, including psychology, psychometrics, philosophy, sociology,….. I believe that there is a neuroscience literature which identifies a neurological basis for category formation.
I would be interested to know readers' opinions as to whether they believe that behaviorism and reinforcement theories in general operate upon categories of events or on a more individualistic basis? Is reinforcement a point in human (and perhaps animal) understanding and experience of the world at which stimuli are generalised or differentiated? Are certain stimuli differentiated and others generalised? If so, what are the factors associated with a stimuli being generalised or individuated?
On the question of "factors associated," you may find at the link a very, very pop-sci version of a fascinating new stream in the neuroscience research that has been finding how random twitches during REM sleep may be crucial for the differentiated form of the nervous system which is key for the differentiation of the types of sensations picked up and used for adaptive behaviors.
http://www.apa.org/monitor/jan06/twitch.aspx
Here is a more technical exemplar of the research:
http://www.nature.com/nature/journal/v424/n6944/abs/nature01719.html
There is a paper claiming that generalisation of stimuli is more pronounced in associations with negative valence compared to positive.
http://www.ncbi.nlm.nih.gov/pubmed/20685988
You ask: "I would be interested to know readers' opinions as to whether they believe that behaviorism and reinforcement theories in general operate upon categories of events or on a more individualistic basis?"
Assuming I understand the question, I can say that it does both. If one reinforces operant responses in the presence of a particular stimulus, say, a picture of a tree (and this experiment has been done with pigeons), you will get high rates of responding only to that picture. There will be some generalization, of course, if you slightly tweak the picture, for example, or happen to have another picture of a very similar tree. However, as you begin to add more and more exemplars, reinforce, responding in their presence, AND withhold reinforcement for responding in the presence of non-exemplars, you will eventually produce a very general functional stimulus class and even novel exemplars will lead to high rates of key-pecking and novel non-exemplars will lead to low rates of key-pecking. That is, in ordinary language, you will have produced a "category." In this sense, all birds and mammals (at least) "have a predisposition to form categories" - the category "trees," in this example. This is simply the way operant multiple-exemplar training works. Presumably, in humans, the process is also important. It may be that humans require fewer exemplars etc. In addition, and in closing, let me say that currently in behavior analysis (the natural science of behavior founded by Skinner, who also founded the philosophy called radical behaviorism) there are many who think that reinforcement contingencies can lead to certain phenomena only in humans in which stimuli can form "equivalence classes" and relations between stimuli can emerge that were never directly trained (though they are still the product of exposure to contingencies of reinforcement). This is a very complicated literature, however, and I will, therefore, merely mention it now.
Cordially,
Glen
- Badges
- Science topic
- Similar topics
- Psychology