Is Biological Intelligence Different From Machine Intelligence?
> Analogies have helped progress in all domains , they aren't only stuff of American TV preachers
, they are ofcourse abused often, depending on the accuracy required, and the level of the audience metaphors may be appropriately chosen

Is an analogy by itself ever serious science?
> Can anybody clear this for me if it is possible to have parallel computing on computers?, i mean really parallel not pseudo-parallel as with many-core computers where os switches from one to other fast enough to give illusion of simultaneous processing, brain performs 'parallel computing', different parts of brain work simultaneously

My guess is that true parallel computers have multiple processors. When source code (in say parallel Fortran) is compiled/linked, it produces executables for running on the different processors. (This is vastly different from time sharing.) This is only a very vague recall. I haven't used these myself.
> Cognitive Science today has a character that is Darwinian, behaviorist, probabilistic and empirical

What does this mean? How is Cognitive Science behaviorist? Is it a natural science, a branch of biology? Does it take into account the law of effect? Does it treat utterances as verbal operant behavior, namely a function of stimuli that precede and follow? Is it functional as opposed to structural? Does it consider motivational and emotional variables? Behavior analysis does all of these.

And how is Cognitive Science empirical? Does it study humans and other animals?

And how is it Darwinian?
This link has the entire Skinner (1977) paper that I have been referring to.
Responses here will be brief as an elaboration of the same has been provided at length previously in this very thread.

(11-Aug-2011, 05:23 PM)sojourner Wrote: Is an analogy by itself ever serious science?

Far from being 'an analogy by itself', the endeavors of the Cognitive Sciences have been to come up with neurophysiologically plausible theories and the field has had its fair share of spinoffs. Examples of all of these can be found in this series of talks the IBM Almaden Institute on Cognitive Computing, 2006.

(11-Aug-2011, 05:23 PM)sojourner Wrote: Is it a natural science, a branch of biology?

It draws on and is constrained by biology and also draws upon a number of other sciences. In the Marr Hierarchy, Level 3 is where biological realism is emphasized.

(11-Aug-2011, 05:23 PM)sojourner Wrote: Does it take into account the law of effect? ... Is it functional as opposed to structural?

The mechanisms underlying the operation of the Law of Effect in the first place can be studied computationally in the framework of Reinforcement Learning and explained neurophysiologically in terms of dopamine pathways and is another triumph of the interdisciplinary approach.

Also, 'functional' need not be irreconcilably opposed to 'structural' as both of these lines of investigations can be pursued at different levels in the Marr hierarchy eg. a reinforcement learning approach is 'functional' while the dopamine pathway investigations are 'structural'.

(11-Aug-2011, 05:23 PM)sojourner Wrote: Does it consider motivational and emotional variables? ... And how is Cognitive Science empirical?

For example, the Iowa Gambling Task pioneered by Damasio and colleagues is just one of many experimental paradigms developed by cognitive scientists to examine the role of factors (say motivation and emotion) which ought to be explained, when in the words of the investigators "adequate knowledge of a situation does not guarantee correct decisions". Here is a sample publication from the group.
Such publications have testable hypotheses and report results statistically based on empirical data.

(11-Aug-2011, 05:23 PM)sojourner Wrote: Does it study humans and other animals?

I wonder if the investigator of this primate experiment (and hence addressing the 'humans' and 'other animals' demand) would self-identify as a 'behaviorist' or 'cognitive scientist' or even would bother at all to belabour such a distinction. However the investigator does choose to present her results on Relativity and Loss Aversion as 'Cognitive Biases'.

(11-Aug-2011, 05:23 PM)sojourner Wrote: And how is it Darwinian?

Darwinian explanations can be brought to bear at almost any level in the Marr hierarchy. At the 'implementation level', Prof. Gerald Edelman proposes a theory of Neuronal Group Selection and for interested readers a lecture by him on this topic can be heard here.

(11-Aug-2011, 05:23 PM)sojourner Wrote: You guys are freethinkers and should at least be willing to consider what I have to say.
And this is not a hasty flaky interest. My interest is a 40 year one.

Speaking of a 'freethinking ethic', our replies so far have been elaborate and based on a careful reading of the preceding posts; and in fact we hope that this will be reciprocated!
Ours too isn't a 'hasty flaky' interest and though admittedly we haven't yet had the chance to spend '40 years' on this interest (or any other interest of ours!), our replies are at least not systematically ignoring the advances of recent decades. Skinner's views while worthy of inclusion in any historical review of the field, do not by any means represent the 'current scientific consensus' and it would be a disservice to readers to simplistically present it as such. Also, once and for all, we are not belittling Skinner but only bringing to bear on the discussion recent advances in related fields since his demise, which are pioneering and paradigm-shifting.

[+] 4 users Like arvindiyer's post
This is, again, mostly a rehash of much of what has already been said in this thread.

(16-Aug-2011, 05:48 AM)sojourner Wrote: > note that you have conveniently disregarded the scientific study I pointed to that demonstrated an advanced computational process in a plant system

Not at all. I had clearly mentioned while responding to that one post that I had seriously considered only the first paragraph. I had not gotten around to the others.

I made that statement in a later comment in response to a post that came AFTER the post in which you mentioned you were responding to the first paragraph alone, in a context where it is apparent you had read not only just the experiment I said you were disingenuously omitting to make your point, but are actually talking about the experiment that came AFTER it. Go back and look at the context of the statement you are quoting. This is terribly disingenuous of you.

Even in quoting me you left out the part that clearly shows the context, which demonstrates you are resorting to diversionary tactics. I've been trying very hard to avoid attributing intentional deception behind your diversionary tactics, but now there is no choice. Here is my full statement, in context:

"Also, note that you have conveniently disregarded the scientific study I pointed to that demonstrated an advanced computational process in a plant system, and are selectively quoting the example in which they are doing a simulation to make the point that it "does not make the plant a computing (sic) system"."

Is it possible that your confirmation bias is so extreme that you were unaware that you were deceptively editing my statement to exclude the part that shows you not only read past the first paragraph, but had also previously selectively picked one experiment that came after the one you ignored, to make a point that the ignored one clearly demonstrates as false? At this point I think you are just not interested in the truth.

Quote:5. I didn't take up your above question (which follows my ">") up until now for a simple reason: my major point in this thread is something else. (See 6 below.)

You said a lot of stuff about my question, but didn't answer it. I think you missed the point. The point of the question was to see if you thought if there is something to biological intelligence that cannot ever be replicated in machines, theoretically. You did everything in your capacity to avoid that question, even to the point of dismissing the value of hypothetical questions altogether. It is not about proof (in the pudding or elsewhere). It is about pointing out that you are simply refusing to see human intelligence as nothing but the function of physical materials that can theoretically be replicated by a machine.

Quote:7. When a person calculates the 15% tip on a restaurant bill, he is engaged in behavior that may be called computing. Likewise when he works on calculations. Other than this obvious sense, he is not involved in computing at all.

You are completely unable to see anything other than your own point of view. If you could, you'd see by now why your view of "computing" (you continue to use the term despite Arvind pointing out earlier why it is not what we are referring to) is myopic. If you refuse to even open your mind to the possibility that you are limiting yourself by viewing computation as simple mathematical calculations, despite the vast literature on the subject, there is no point to this conversation. We are going around in circles because you are unable to accept the basic premises that have been made explicit.

Quote:8. Neurons fire to produce different behavior in different circumstances. Nothing is gained by calling this computing.

Neurons send electrochemical signals. There is nothing to be gained by saying neurons "fire". Fire is a specific physical phenomenon and the analogy is wrong and NOT USEFUL.
See how that works?

I have a background in biochemistry. You are consistently implying that we are ignoring basic mechanistic biological processes when we see intelligence as involving computation on a scale that we are far from understanding. This is simply not true.

It is also simply not true that looking at the computational aspects of intelligence is not useful. Not only have we already pointed out multiple applications of such an approach that are already in use, but we have also pointed out evidence for why scientists believe that much of the benefit is yet to come because this is a new science. Most importantly, I have already explicitly addressed this specific point, saying that fundamental science does not follow some pre-conceived plan to apply future scientific discoveries in specific service to humanity.

Quote:11. You guys are freethinkers and should at least be willing to consider what I have to say.

And now we are in troll territory. The minute you refer to the entire group as a monolithic entity, dismissing everyone else as unwilling to see your point of view, we must take a stand. Again, let me repeat, NO ONE here has ignored any of what you have said. It is you alone who is doing the ignoring. I have not refuted but have agreed with every single thing you have said about behavioral aspects of humans. It is you who is not willing to consider more recent scientific advances that have gained favor since the time of your favorite scientist.

If anyone should be accused of not acting like a freethinker here, it's you. Do give this some thought.

I think it is time to wrap up this thread, as we are wasting our time in distractions and diversions. I request that you, Sojourner, refrain from such tactics. Let us stick to the topic under discussion. To end on a good note, here are two articles by professor Wadhawan from his series on Nirmukta, talking about how the evolution of complexity relates to intelligence.

Quote:"An ant colony has swarm intelligence, just like a beehive. It is instructive to describe its basic (even though highly simplified) features here, the more so because ANT LOGIC has already found several applications in artificial evolution and in computational science. Dorigo and coworkers did some pioneering work in this regard."
"An ant colony is a remarkable parallel processing machine. To quote Kevin Kelly again: “Ants are the history of social organization and the future of computers.”"

And this next article is even more relevant, with more than a dozen references to computation:

"Fossil rabbits in the Precambrian"
~ J.B.S.Haldane, on being asked to falsify evolution.
[+] 2 users Like Ajita Kamal's post

Users browsing this thread: 1 Guest(s)