The Institute on Man and Science; Systems Interact, part II
National Educational radio in cooperation with the Institute on man and science presents a series of talks drawn from the institute's annual conference held recently in Rensselaer Vale New York. The Institute on man and science is a nonprofit educational institution chartered by the New York State Board of Regents. The annual assembly of the institute is designed to focus attention on 20th century technology and the human relationships resulting from its application. On this program VO Parsi Jin of Rensselaer Polytechnic Institute will give the second of three lectures on the topic. What systems analysis has taught us about ourselves and the natural world. Here now is Mr. Parsi Jhon. Now it's fine to talk about systems and so on but how does one utilize these in an actual illustration and actual demonstration that is useful for analytic purposes. And here
we come to attempts to develop systems diagram systems diagrams. But now what I've said is that no event emerges unrelated to earlier events and no event and without follow up of many other events. And I noted also that the thing gets all bollixed up with feedback all along the way. In the past in the future where do you start drawing a system's diagram a control diagram. When you've got these things feeding backward and forward interrelated in such complicated ways. Well it turns out that there is no absolutely right way or absolutely wrong way I guess either for drawing a system's diagram. I have never yet seen a system's diagram that tells the complete story on any situation but me warn you about
that. This is the word Cybernetics is not a cure all nor is the effort to use cybernetics a cure all. It is more an attitude an appreciation that interrelationships exist then it is a computer that gives you final answers with everything fed in properly and everything coming out neatly. However the realization that there is there are these interrelationships. It is very important is that this is the heart and soul of the concept of cybernetics. And this is the heart and soul of any effort that tries to utilize these in any particular situation. So what we've got to do is to try to begin somewhere. And usually it's a good idea to think of systems diagrams as being made up of input
and input. Well what do you call an input. Information is often the input. It may be a mechanical motion however it may be a position of that some a stat by metal It may be a variety of things. But it's possible to begin with something that you call and inflict this experiences a transformation transformation. It may be an amplification in the case of your phonograph recorder. You've got an input in the form of the needle motion and an amplifier converts that into sound. It comes out. So usually you can find something that you call an input and something you can call an output. And usually after you've tried a few of these you come back to something that seems to
represent the thing as effectively as you can. The input and output concept is significant and helpful in approaching a problem. It makes you think what shall I call the variable What is it that I'm really after. What is it that is the significant thing that I'm concerned with. In what way does it become revealed in its effect. I can think of maybe cause and effect and this is a beginning. However we find that usually the output does not go in the straight chain as we noted yesterday. But there is an influence back which is the feedback. Now usually there is a transformation from the input to something thats totally different from the input. For example the information that I got was a tap tap of my
finger or the stick on the desk. This is nothing at all like the motion of my feet going forward and yet that tap tap must be converted to become something that represents motion of the feet. So here I've got information going into the system amplified through my body and output my legs moving. Now I've got feedback and what's wrong with the diagrams I have drawn. I'll ask only our hot shots in front what's wrong with this. Yes exactly. Exactly I've got my feet feeding into my head which is you've observed in people but it is not good. So what we've got to do is to put in a transform function.
So we have a transformation here in the form of maybe an amplification the information in my head becoming transformed amplified into foot motion and the foot motion must then similarly become through the various media transformed back into the same kind of be it apples. It's got to feed back apples and so on. So that bit is this interrelationship. Now what's missing in this diagram. An energy source. So we've got usually a variable here or more often a variable and a sensor device. Detector or whatever. Something that could be measured. Experiencing a transformation into an output which
in most cases requires a feed back which must again experience the transformations to make it possible to have the thing correct. So that's an illustration of a simple diagram that we if we might look at. Now I will put all the details on this that we normally have on diagrams. However let's look at this point. The input in what is it that we can't consider to be input of the right kind. It may be able to. This may be a plus or minus.
Evolves and in the case of an amplifier this is exactly what happens when you get out of it and output in books. It may be able to teach which is not a certain of a certain absolute magnitude that you are measuring but very often it is a differential a differential of voltages where you have a standard. But you've got some kind of a standard and you've got a Delta be a differential B plus A-minus which is the thing that you're after. In other words it may be simply an error detector and error detector. In the case of the thermostat this is exactly what we have. The thermostat goes up and down a little bit. We don't have an absolute calibration of that the stat position. All we have is a position that we set for that temperature to be 70 degrees.
And it's the differential motion of that thing that goes up and down up and down and causes all the control action to be executed around that around and air around and a measure of measure of departure and increment of departure from the thing. So that's very often the kind of system that we must have. Well let's proceed then with the automobile driving that we mentioned yesterday. Here's a case where we are now driving the automobile. Our foot is on the accelerator and we proceed but the red light on the stop sign becomes a signal for us to take some action. And in this case the proper transformation is for that information to become negative feedback to not cause the foot to go further wrongly for down on the accelerator but rather to go
on the brake and take the proper action to make that system stable. Or as we noted it might be a light changing from red to green as we approach it in which case we get a positive feedback because we can go a little faster the light tells us we may within reason. A positive feedback can sometimes be just what you need and persistence to remain stable within certain control of it. Let's see how we might represent that motion driving of the automobile. What is the variable in this case. The variable is the speed and power and actually the position along the road. There are several variables. There are several variables that come in here. Well
now we can use that as the variable or we can simply utilize something we call external visual and other sensory signals. On instruction. At any rate we have a situation where the input is not just one driving down the road but it is actually information of many kinds that comes to us relative to what's happening as we drive. Then this comes in the form of sometimes lost sometimes minus and you ask me which is plus which is minus. And I cannot answer because this is a plus in relation to something else that you call plus or minus. The same must be self-consistent and just put it down at this point does not give the story but now this information is fed into the system. The body of the driver the brain of the driver is foot motion
and the mechanical gadgetry to be represented by the automobile. There is the. First the metabolism of the body and sensory. And motor phenomena of his nervous system. Which causes the action on the foot to be what it should be. There is the output of this the muscle action. Which went imposed on the mechanical system of the automobile. The results of the motion forward. And here they auto motion the conversion of the gasoline energy into the motion of the wheels. And
note that we've got an energy source at this point for the metabolism of the body. It is the energy source at this point for the conversion of gasoline power to the motion of the vehicle. And now let's see what we have as output of all this first. There is an immediate feedback. As one translates the information into this action there is an immediate awareness of what one has done. This feedback of a kind which again needs to transform function so we put a box there that is as soon as you have decided that you will do something with your foot. The rest of your body has some cognizance of what you've just done so that's immediate
feedback that's contained within the driver's own system. The automobile begins to move down the road. And here again that is feedback of a variety of kinds. The motion of the automobile actually begins to influence the automobile itself. The automobile also suffers input feedback devices again requiring it transform function because this may be the motion of the wheel at this point. And this may be the behavior of the carburetor so that there is some kind of a conversion thats needed so that in each case there is a local feedback that comes into play. Then finally the car moves down the road and we've got the automobile position and speed etc as the
output of this system. Not noticed that I have made this the output and I could have made this the input and it really doesn't matter because we've got a system that reflects back on itself and so whether you begin at one point or the other point is not absolutely determinable in any situation and you may do it one way. Another person do it the other way and there's no one to say that you're right and he's wrong or the other way about. But we do have this situation now. The position the motion or whatever then does give an overall feedback and is one of these inputs to the system. And this is again the wheels of the automobile. The position along the road and everything else and we've got to be careful not to feed the rolling wheels into the brain so
we've got to put in the transform function again for this feedback loop. So this is a simple representation of what might be the driving of the automobile in terms of the control system illustration. That is something else we may put it supposing there is a and imposed hazard that suddenly shows up. Sometimes it's possible to put in a disturbance at this point. Or if that is the sudden appearance of a car and the risk of an accident or it's possible to simply call that one of the input sayings that you make and lies so that you see what we have done is we've made an imperfect representation of the driving of an automobile. But in the process of trying even this imperfect representation there is a systematic approach to the thing. There
are these concepts that one we've had to keep in mind. All along the way and this is of course important in life in science and everything else. The other thing that you recall that we won't go into in detail but remember that the driving of the automobile has this one very highly sophisticated characteristic that it is a proportional control system and of a sophist of a degree of sophistication that very few control systems can approach. In fact the bodies and tire operation represents a control system which the physical control systems be it with complicated Electric an electronic computer or whatever. I cannot begin to approach. There's a sensitivity a variability and it adaptability which is simply beyond the machine
age. Supposing we go to a biologic system and illustrate another application of this in connection with the raising of an arm. Supposing that I raise this arm to the horizontal position and of course I can see the position to be horizontal and if I were to move it up and down I would see that the arm has moved up and down. But now if I close my eyes I find that I am still aware of the ups and downs of that arm and I am able to hold it out there until fatigue sets in. What permits me to know what is the position of that arm despite the lack of visual observation. The reason is the very reason that brought about the reasoning that brought about the concept of cybernetics by Norbert Wiener and his associates namely that there is within the body.
A system of control there are into relationships between the muscle action and sensory action that allow awareness and allow control without vision. The illustration of this as it is usually given involves having say the arm out here as at this and with a muscle that does the actuating and sensory nerves running from that to the spine spinal area where going through a Cynapsus there is an automatic or appears automatic in many cases. Motor connection to the muscle which causes that muscle to act in the way that we intended it
to act from our original volitional move. Now this is incidentally I think the bit where I depart from the usual illustrations that are given of this phenomena. Usually the illustrations carry the sensory nerve. From here to there and then the muscle is put somewhere here so that the efferent nerve is here the efferent nerves going to the muscle is here and this is called a reflex arc. I don't like that. I think it is absolutely necessary to move these things to be part of one in the same muscle system and in which case I prefer to call it a reflex loop a reflex loop. I've got to check with some biologist friends and find someone who'll argue with me on this. I haven't found anyone who is
prepared to argue yet. Now this exists. Of course it is not a single muscle or a single afferent nerve that comes into play. There are very many of them that come into play there is a gradation of things. But this information is there and in many cases much of it is purely automatic. Now what about the the withdrawal reflex that we sometimes referred to when we touch a hot object or a shark pit. Supposing there is a shark over here and we do get the. Knowledge of that and this is something that we've become aware of. Therefore from this the efferent nerves must run to the spine but also to the upper regions of the nervous system the higher regions maybe even the cortex. Undoubtedly the cortex I guess.
And from there there is a response which brings this thing back and causes the muscles of the arm to pull away from the pain stimulus. The thing causing the pain stimulus or the sharp point. But throughout all that this other automatic reflex loop continues its function. In other words you can impose many other sensory signals on top of what already exists as an automatic city and so we would represent this as this nice diagram shows but I can't draw on the board as the. Sensory receptor system functioning sending signals in and the receptor then causing either automatic action hero reflex action or the sensory signal requiring relational action of some sort
on the part of one or many groups of muscles. One way to illustrate this is the following. There is in the muscle system the receptor organs nerve endings which are sensitive to stretch of the muscles. This is then the the input and a sensory nerve may carry this thing to the central nervous system. And just to which levels of that central nervous system will depend on the nature of that stimulus and how much that commands our attention. This in turn causes feedback and at this point we have great difficulty in
identifying the feedback separate from the sensory receptor. So we've got to put the effect or the muscle system. Very close to the receptor nerves and in fact the relationship is so close that the feedback between the action the stretch of the muscle and the senses and the response of the sensory nerves is so direct so immediate so responsive that you can almost put these in and call the feedback and the sensory system as one in the same unit. You have to recognize that the functions are there and differentiated. So here's another way to illustrate one of the characteristics of the human body. If however we want to illustrate this other
situation with the other sensory information coming in we would make this watch maybe a little bit larger. We would come in with a sharp sensation from the arm again to this to the spinal regions and now however we've got to go up and then down our stay within the box and then down. And the feedback at this point may be feedback to a different set of muscles. Different set of muscles to pull the hand away. But this system persists in its functioning. The reflex loop continues to function. In this case what have we done. We have moved the muscle which has caused the which has ended the sharp pain stimulus
from the pain and feedback. It kind of terminates at that point. But with this although it has the continuing cycle that does not break the circuit. Let's try another illustration of of a man in his environment. Again we find that you can draw boxes and interconnect them. And very often it almost doesn't matter whether you choose the situation from the social from a social phenomenon social situation or biological situation. You can put new names in the boxes and this is the interesting thing. We can't put any kind of a name.
You can put Greek names into these boxes and the one who has studied these principles sees the interrelationship and you may not know what is being fed back but he knows that there is a feedback function and the function runs this way or that way and so on. So regardless of the names that I put in there the nature of the interrelationships can be realized at a glance. And this is one of the virtues of this approach. You heard VO Parsee Gena Rensselaer Polytechnic Institute as he gave the second of three lectures on the topic. What systems analysis has taught us about ourselves and the natural world. Mr. Parsi Jian spoke at the annual conference of the Institute on man and science held in Rensselaer bill New York. Join us for our next program when Mr. Parsi June will conclude his discussion of systems analysis. These lectures are recorded by the
- Systems Interact, part II
- Producing Organization
- Institute on Man and Science
- Contributing Organization
- University of Maryland (College Park, Maryland)
- AAPB ID
- For series info, see Item 3566. This prog.: V.L. Parsegian, part II
- Media type
Producing Organization: Institute on Man and Science
- AAPB Contributor Holdings
University of Maryland
Identifier: 68-33-25 (National Association of Educational Broadcasters)
Format: 1/4 inch audio tape
If you have a copy of this asset and would like us to add it to our catalog, please contact us.
- Chicago: “The Institute on Man and Science; Systems Interact, part II,” 1969-03-03, University of Maryland, American Archive of Public Broadcasting (GBH and the Library of Congress), Boston, MA and Washington, DC, accessed September 21, 2021, http://americanarchive.org/catalog/cpb-aacip-500-zg6g624w.
- MLA: “The Institute on Man and Science; Systems Interact, part II.” 1969-03-03. University of Maryland, American Archive of Public Broadcasting (GBH and the Library of Congress), Boston, MA and Washington, DC. Web. September 21, 2021. <http://americanarchive.org/catalog/cpb-aacip-500-zg6g624w>.
- APA: The Institute on Man and Science; Systems Interact, part II. Boston, MA: University of Maryland, American Archive of Public Broadcasting (GBH and the Library of Congress), Boston, MA and Washington, DC. Retrieved from http://americanarchive.org/catalog/cpb-aacip-500-zg6g624w