Epistemology of Automatic Categorizations

[Christopher]

Rand makes very clear that concepts are a matter of isolating and integrating attributes from the local environment.

Isolation at the concrete level takes place automatically as a function of perception.

Prior to conceptual formation there is awareness to entities and identities.

Entities and identities are also founded on perceptual concretes (Rand does not explicitly say this, but I see no other explanation).

What Objectivist Epistemology does not address is the evolution of awareness to perceptual concretes.

Since survival is dependent upon action, logically perception evolved to support successful interaction with the environment.

Therefore, it seems to follow that perception is organized to understand attributes in the environment related to behavior and causations that influence interaction.

For example: a thornbush is identified as a single entity/identity because interaction with the thornbush leads to negative consequences. Perception of the boundaries of the thornbush are established in order to recognize the boundaries where interaction with existing stimuli produces identical consequences (touching thorns and hurting). In other words, the thornbush is organized wherein it has an active and connected influence with an organism's behavior. Similarly, if I shake a part of a bush, another part of the bush vibrates. Automatically, my mind connects both the part being shook and the vibrating part into one category based on how the two stimuli are connected through behavioral interaction.

Within this approach, there must also be automatic categorizations of causation. Since perception ultimately facilitates successful behavior, part of automatic perception must be to identify sources of causation within actions. In other words, categorization based on interactions are essentially categorizations based on causalities of stimuli to observed effects.

Does this sound logical? The extension of this is that changes to behavioral goals (i.e. directions of interactions) should also guide perception towards specific forms of categorizations. For example: Animal needs fruit. Fruit hangs on tree. Animal interacts with tree by shaking in order to get fruit. Therefore, animal's category of tree is formed based on how the interaction with tree demonstrates common effects from tree (base shakes, branch shakes, fruit falls). Conversely, shaking a tree does not shake tree's exposed root about 3 feet away from the base, so it is unlikely animal recognizes tree and root are the same category.

[merjet]

Christopher, I suggest a look into Aristotle's ideas about perception. I wrote a short article about it here.

There are also a couple fairly recent books on the topic, both of which I have read and recommend.

Aristotle on the Common Sense, by Pavel Gregoric

Aristotle: The Power of Perception, by Deborah Modrak

Edited July 31, 2009 by Merlin Jetton

[Christopher]

Interesting. So Aristotle is basically hypothesizing associative memory, which is exactly how lesser evolved learned brain functions work.

If we think via associative memory, then basically a single stimulus arouses a single neuron; when two separate stimuli arouse two separate neurons at the relatively-same moment in time, those neurons become associated together. Later, when either of those neurons is stimulated independently, the network of all neurons tied to that neuron are also stimulated (like seeing a ball partially hidden by a blanket, but actually understanding immediately the entire shape of the ball, including the hidden part). Eventually, these simultaneous arousals form a network which becomes the "implicit" concept of the entity.

So all perception as a mechanism does is funnel electrical impulses from sensory nerves to the brain in such a manner that the stimuli excite neurons. The rest happens internal to the brain. It's interesting that the neuron-bonding which creates an entity actually happens. Rather miraculous, actually.

I wonder to what degree electrical impulses that reach the brain are already pre-structured via pathways to the brain.

[Guyau]

Christopher,

You might like to check out some of the Objectivity work indexed below.

Also, there is incorporation of categorical perception and preattentive perception in this one, followed by this one.

Association and Logical Inference

V1N2 37–41, 55–56, V1N3 10–11, 21–23, 26–27, 74–75, V2N1 134, V2N4 128–33, V2N6 96–97;

Association and Perceptual Integration

V1N2 38–40, V1N3 21–22, 61–65, V2N2 75, 86–88, 100, V2N4 149–55, V2N6 112–13;

Association v. Representation

V2N4 128–29, 132–33, V2N5 39–40, V2N6 83

Neuronal Mediation of Perception

V1N1 20–21, V1N3 7, V1N5 20, V2N1 113, 115–17, V2N2 86–87, V2N3 126–27, V2N4 110–11,

123–26, 134–35, 142, 145–46, 149–57, 166, V2N6 13–18, 22, 24, 31–32

Evolution of Consciousness

V1N2 54, 69, V1N5 22, 36–42, 46, V2N2 88–96, V2N6 4–5, 36

Edited August 1, 2009 by Stephen Boydstun