Nowadays we try to simulate each process by a computer program which hashes parallel tasks in a sequence of partial tasks. Do we lose in such a mode the dynamic of the entity? Artificial neuronal networks were also simulated as simple brain models.
Is there a risk in such a computer model that important phenomena would be lost which could result in an analog network interesting phenomena?
Wilfried ,
My personal opinion would be that it is not the computer simulation itself that is the problem but the model built into the simulation. Our models of neural networks are so simple and they are forced to do complex tasks done by so many of our neurons that it is unreal to expect good results in comparison to the real thing.
I would not even bother about the creep in calculations in a computer simulation(due to its digital nature) at this point in time since our models are so crude(lets not even go into a higher level than network were we are still in infancy).
Another point worth mentioning is that current trend in artificial neural network has focused so much in optimization rather than biological simulation.
I recommend the dialogue between Rosen and Pattee (and the consequent breakdown in their dialogue) regarding the "modeling relation" and the "epistemic cut."
Ultimately, my colleagues and I on both sides of the Rosen-Pattee fall-out end up finding more inspiration from analog simulations.
Pattee's commentary on his relationship with Rosen
https://www.researchgate.net/publication/5891221_Laws_constraints_and_the_modeling_relation--history_and_interpretations?ev=prf_pub
Arran Gare's commentary on Rosen and Pattee
http://cosmosandhistory.org/index.php/journal/article/view/109/218
Also, see Cariani's review of Gordon Pask's electrochemical ear as a classic case of what analog simulations can do:
http://www.cariani.com/CarianiNewWebsite/Cybernetics_files/PaskDevice93-SelfOrganizingSystems.pdf
Article Laws, Constraints, and the Modeling Relation – History and I...
Pentti Haikonen build a small robot XCR-1 without any microprocessor or computer software because he has the opinion that only a hard wired electronic machine and not a simulation of that machine would be able to have qualia.
This is an example to my question and I would like to discuss this question here.
Link:
http://www.conscious-robots.com/en/conscious-machines/conscious-robots/xcr-1-robot.-sound-direction-detection-demo.html.html
I am afraid that my question is in a dead end, so I like to re-ask it once more.
Are the digital simulation (algorithm) and a analogous electronic construction equivalent, not even in the results but also in behavior? Pentti Haikonen's opinion is against equivalence of these two approaches.
Happy New Year
to all who have contributed their valuable posts and to all who will do this soon.
- Badges
- Science topic
- Similar topics
- Psychology
- Cognitive Science