A Neural Theory of Mind

Paul Bush paul at phy.ucsf.edu
Thu Apr 25 15:40:06 EST 1996


In article <4ljiob$4de at globe.indirect.com>, marty at indirect.com (Marty Stoneman) writes:

|> Two questions hack at me: how do you distinguish, if at all, between 
|> learning of mammalian kinds resulting in new "memory" -- and episodic 
|> "memory" of humans for specific chains of events?

I would say they are the same process, episodic memory being a specific
spatio-temporal pattern of the relevant concepts.

|> and, since (I believe) 
|> other mammals have the same kinds of brain tissue and parts that we 
|> humans have, what gives us abstract stuff like "ideas", "thoughts", "deep 
|> conceptual learning", etc.?

I am proposing that all mental activity is produced by the same basic process
(neural firing pattern matching and induction, in space or space and time).

I believe that an idea/thought is neurally equivalent to a perception or motor
act - the firing of a pattern of neurons driven by a specific collection of
simpler concepts/features in the level below (actually the inverse for motor
acts). A new thought or inference is a pattern induced by neighboring patterns
(what we would regonise as similar concepts). I think new patterns induced at
high levels are constrained by the objective structure of knowledge space, which
really just means that the neural pattern dynamics capture the dynamics of the
real world system they are modeling.

Thanks for your comments. I still want to hear about your work. (My email is down
for the moment if you have mailed me).

Paul



More information about the Neur-sci mailing list