----- Original Message ----- From: "Ben Goertzel" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Monday, February 24, 2003 7:36 PM Subject: RE: [agi] the Place system of the rodent hippocampus
> > > I agree that physicists, myself included, might need some sort of a > > neural-plugin to supplement the physical-map-building hippocampus for > > mapping "quantum environments." > > > > I disagree that physicists, myself included, with this sort of a > > neural-plugin would have any enhanced cognition of external "quantum > > environments.". > > > > In brief, I'd never be able to get the darned plugin, plugged in and/or if > > you did, there would always be a certain 'uncertainty' vis-a-vis the state > > of communication between the systems, i.e. remember the > > Heisenberg bracket. > > Obviously, my opinion of cognition is somewhere between non > > 'Penrosian' in a > > modern context and 'Parmenidean' in an ancient context. > > > > But, to be sure Ben, you've got a phenomenal imagination and an articulate > > way of expressing it ;-). > > Ed, > > I see two options > > 1) the quantum neural module is purely classical in its operation, merely > *simulating* quantum dynamics via implementing equational models of them. > In that case, problems like the ones you describe don't exist. The question > is whether this kind of simulation can be done in an adequately efficient > way to provide helpful quantum-environment-modeling. > > 2) the quantum neural module is actually a macroscopic quantum system, > allowing modeling of quantum environments in a more direct way by *being* a > quantum environment. The problems that might come up with the two proposed options: Option 1. Problems of implementing equational models...jeez, Ben, I'd be out of my 'quantum mind' to claim that I could say anything definitive about adequately determining wave functions to correctly distinguish and describe a cognitive environment...let alone a physical one. To do this correctly, there are some very serious quantum concerns that are particularly exacerbated when applied to a presumably an 'internal' cognitive environment. These include implementing 'the' correct model, 'distinguishing' the correct elements within the model, and performing the act of modeling. Where does 'quantum cognitive model' draw the line in the sand to begin ascribing wavefunctions and what wavefunction formalism should it begin using? From the simple, how does the system 'internally' differentiate elements of the system that it is modeling, to the more complex, how does the system 'internally' differentiate itself from the modeling process and correctly model the process? It strikes me as what I might characterize as an 'internally recursive' quantum problem. And I know you understand the implications of the intimate relationship of the quasi-imaginary mathematics that is inherent in quantum modeling...to say the least it is counter-intuitive...so I'll stop there. And I certainly don't mean to denigrate anything that has been stated or implied about real AGI models so far, because there's been a lot of excellent and constructive discussion, but my best guess is that the real AGI models would look like 'tinker-toys' juxtaposed to a correct quantum AGI model, if such could be accomplished. Option 2. Problems of the classic 'Parmenidean' dilemma of 'being v. becoming'. Even in philosophical treatments, in my humble opinion, solutions for the problem of 'being v. becoming' leave a lot to be desired. Your suggestion does provoke thought...and it might be interesting to consider ideas of 'quantum cognition', or plugins therein, along side ideas of 'quantum computation . I'll think about it a bit more, while I take another sip of my coffee. And at that, imagine having to deal with the harsh realities of a microscopic quantum world...I might only be able to get that next sip of my proverbial quantum coffee, if I could only get the quantum coffee in a stable enough quantum 'on' state for a period 'long' enough to allow me to appreciate it ;-) Just my $0.02 worth. EGHeflin > > In this case, couldn't it happen that, via the close coupling with the > quantum plugin, the dynamics of the brain as a whole would be transformed > into those of a "macroscopic quantum system" ? > > In other words, even if those who call the brain a macroscopic quantum > system NOW are wrong, isn't it possible that the introduction of the > hypothetical quantum-environment-modeling neural module would cause this to > become the case. > > And presto, a new species, Quantum Man, is born .... > > -- Ben > > > ------- > To unsubscribe, change your address, or temporarily deactivate your subscription, > please go to http://v2.listbox.com/member/[EMAIL PROTECTED] > ------- To unsubscribe, change your address, or temporarily deactivate your subscription, please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
