Pushing Boundaries with Dr. Thomas R Verny

Prof. Gordana Dodik-Crnkovic PhD, Recent Advances in Cognition, Computation and Robotics

February 07, 2023 Thomas
Pushing Boundaries with Dr. Thomas R Verny
Prof. Gordana Dodik-Crnkovic PhD, Recent Advances in Cognition, Computation and Robotics
Show Notes Transcript

We face a challenge of understanding information. Because information is considered as something that is written in a book, in a message in a newspaper, it's not understood as something that two bacteria can exchange. And then there is the social challenge, because we think typically of an individual,  an individual human, but not about the distributed cognition in a social setting, both for humans, but also for smaller organisms like cells. So, the idea of cognition on the basic level of the smallest living organism, which is the cell has been very difficult for many people to accept and to understand that everything starts on the level of cell. And some people would say that even parts of the cell have some cognitive properties.

There is much  controversy in cognitive science, between people who consider cognition only as symbol manipulation and language, human language, and the people who claim that we have to take into account how much of processes that  underlie cognition are done within the body and the environment. So it's embodied embedded, an active combination. It's not hanging in a vacuum. It's, it's, it's connected with the world. And we were very much interested to get this discussion between colleagues who believe in different views, to get them together and to get this discussion showing where the positions are, and is their common ground and how we go further.

Presently, Prof. Dodik-Crnkovic work focuses on intelligent computation and cognition as very important processes found in nature that we can use to inspire new technologies.

"Being human to me means the privilege of sharing the past and present, and anticipating the future with other humans. Of course, most importantly to be able to share feelings, and thoughts and be in the world with the closest people like family and friends."

If you liked this podcast

  • please tell your friends about it,
  • subscribe to this podcast wherever you listen to podcasts and/or write a brief note on apple podcasts,
  • check out my blogs on Psychology Today at

https://www.psychologytoday.com/intl/contributors/thomas-r-verny-md





SUMMARY KEYWORDS
cognition, computation, cell, people, understand, processes, cognitive science, thinking,  consciousness, problems,  life, agent, human, world, physics, 
nature, robotics
SPEAKERS
Speaker 2 (87%), Speaker 1 (13%) 
1
Speaker 1
0:03
Good morning. This is Pushing Boundaries, the podcast about pioneering research, breakthrough discoveries and unconventional ideas. I'm your host, Dr. Thomas R. Verny. My guest today is Dr. Gordana, Dodik-Crnkovic, Professor of Computer Science Mälardalen University and Professor of Interaction Design, Chalmers University of Technology.  Both located in Sweden. Her current research is in morphological computing and the connection between computation information and cognition by way of interacting agents and different levels of organization from physics to biology. Welcome Dr. Gordana, Dodik Crnkovic, may I call you Gordana.

1:00
Yes, absolutely.
1
Speaker 1
1:02
Thank you. As I said to you, before we started, I was fascinated by your paper in entropy, which is a bit of a mouthful in Entropy, Cognition as Morphological/Morphogenetic Embodied Computation In Vivo.
You have written that the present day human centric view of cognition has a variety of problems. A better understanding of cognition is centrally important for future artificial intelligence, robotics, medicine, and related fields. I totally agree. But I would like to hear more from you on that score. What do you mean by cognition? Let's start there.
2
Speaker 2
1:58
Okay, I would like to give you a background to this article. Absolutely. This is part of a special issue of entropy, which was dedicated to embodied cognition. And that's the third part, we had already two parts with different colleagues collaborating on this topic. And this is controversy we have in cognitive science, between people who think that it is quite enough to think cognition in terms of symbol manipulation, and language on the level of human language, and the other people who claim that we have to take into account embodiment of human agents and take into account how much of processes that  underlie cognition are done within the body and the environment. So it's embodied embedded, an active combination. It's not hanging in a vacuum. It's, it's, it's connected with the world. And we were very much interested to get this discussion between colleagues who believe in different views, to get them together and to get this discussion showing where the positions are, and is their common ground and how we go further. 
So my idea here, in this article was to explain what is the conventional use of cognition, if the open encyclopedia or Wikipedia or whatever which is considered the common knowledge, shared knowledge about cognition, we will see that it is about human cognition, it is always on the level of language and symbol manipulation. So I started the article with this definition from Stanford, Stanford Encyclopedia of Philosophy were written by Paul Taggart. He divides the cognition in the following way. 
In its weakest form, cognitive science is just the sum of its constituent fields, psychology, artificial intelligence, linguistic neuroscience, anthropology, and philosophy. Interdisciplinary work becomes much more interesting, when there is a theoretical and experimental convergence on conclusions about the nature of the mind. And he elaborates further about our need to understand mind as a process of knowledge generation and cognition. And all those terms are quite vague, and he actually ologists that our present understanding of cognition is connected with a number of challenges, questions that we cannot answer from this framework, which is quite narrow and quite rigid. So Tagger says that we have emotion challenge, how to understand emotion in this framework, then consciousness challenge, the World Challenge, how to understand the connection between the agent and the world and the agent is always a human, the Body Challenge, how the body interacts with this cognitive process and dynamical system challenge, because it is assumed and it is classical historical assumption that computation is symbol manipulation. So, it is impossible to understand dynamic systems as computational systems because computation is not considered to be a dynamic process. So that many, many things meant many notions that are imminent in this approach are traditional historical, and they have been changing within their specific fields. But the encyclopedia articles are not following the development in all those fields which are contributing. So, the idea of computation is has evolved during the years since, since 40s, or 50s, when it when it all started. 
And we have a challenge of understanding information. Because information is considered as something that is written in a book, in a message in a newspaper, it's not understood as something that two bacteria can exchange. And then the social challenge, we are thinking typically of an individual, individual human, but not about the distributed cognition in a social setting, both for humans, but also for smaller organisms like cells. So, the idea of cognition on the basic level of the smallest living organism, which is the cell was very difficult for many people still to accept to understand that everything starts, at least on the level of cell. And some people would say that even the parts of the cell have some cognitive properties within this context of cell and within this cognitive process. And so the last was the mathematics challenge. And that was something that already Barry Cooper has been writing about in terms of mathematicians bias. And Matthias show, it's in the book on computational ism, New Directions, where he says that our classical understanding of both computing and mathematics is not is not sufficient. Mine is not that kind of symbol manipulation. It's not entirely only the symbol manipulation, it has layers below, which are signal manipulation, data manipulation, that physical computing or physical processes that we can understand as computing, which which I termed morphological computing, that's the term that is coming from robotics, where people understood robot is understood that they can replace part of brain control over the robot with the body shape and the relationships between body parts. So for example, if you want to have a robot which grasps an object with the hand, you just need right form of the hand right? proportions of things, right right materials in it. So and if you put an object in the hand, it will automatically close because of the morphology. So it will produce behavior, just based on morphology. And similar with the passive Walker that they have in, in robotics as they have a robot very, very primitive one, which is going down the slope without any control just based on the body, which it has arms and legs and everything. And then it moves down the slope without any control. So, they saw the possibility and the chance to use the body instead of motors, motor control and computers and so on in part of the control of the of the robot. So, those thoughts converge towards an idea that we can study natural systems and try to learn how are they using their bodies in order to solve their problems? And I think you had as a guest, Michael Lydon

10:45
was that yes,
2
Speaker 2
10:48
yeah. And he was working a lot about the detail mechanisms be beyond the biological behavior on the level of the cell, how the cell is capable of solving its own problems in its own space and efficiently learn and change its morphology, so that it survives and thrives in in in the environment. So, those are very exciting developments, and they have a lot of application seven practical applications. So, for us, it is really very important to understand not only the language on the level of human, but also the language on the on the level of bacteria, quorum sensing it is called when when they talk to each other with chemical molecules exchanged. So, there is a lot we can learn from nature.
1
Speaker 1
11:51
Now, you said a minute ago that even parts of cells not just cells, but even some parts like the mike mitochondria, for example, would some people think may have may be able to have some combination? Could you just explain that a little bit,
2
Speaker 2
12:11
I will tell you, I think we were talking in this article, I mentioned the continuum between evolution of biological systems and evolution of physical system evolution of the universe for example. So, physics already has mechanisms for changing the structures depending on the past history. So, we have a galaxies that have History and development and evolution and so on as far so, people have been writing about evolution in physics, evolution in chemistry, evolution in in biology, so, it is a continuum. And when we look at physical systems, they have ability to self organize, and they did this pure physical process based on physical forces between parts and it is spontaneously developing in nature. So, the similar processes go on on a chemical level with molecules and so on, they self organize, then we have okay on the on the physics level, we have we have self assembly then we have self organization and then we have autoplays is when we come to the to the level of of the cell which is already a complex system communicating with the environment through the membrane and this makes this difference between in inside and outside world and we have this boundary which which which is defining an agent, an agent which which is a self in relationship to the other that is outside and it is solving its own problems. So, it's a complex system which is made up of many parts and some of those
1
Speaker 1
14:40
parts per second. What is our autopoiesis
2
Speaker 2
14:45
Okay, outpaces is a term that Mathurin and perilla Santiago, School of cognitive science, introduced to describe how a cell is able to self maintain and continuously have a kind of browser support the processes, which will retain its essential structures, and it is done with exchange of energy and the matter with the environment. So, it takes the energy from from the environment, because it is the thermo dynamical system far from equilibrium. So, in order to support it's order orderly structures, which retry inside the cell, it needs constant flow of energy inside and then it has a rest products, which are eliminated through the, this boundary. So, the sale is a very, very complex system which which, which is made of parts, and they are interacting all the time. And some of those parts are taught to be integrated by symbiosis, or in some way, through the history of the development of cell from the outside, maybe they were their own autonomous agents in the beginning, but in some processes, they became part of the cell. So, some of them I think, I'm not a biologist, but I think mitochondria have their own DNA, or there are some parts in the cell which have, has had their own DNA, they are they, they have some kind of agency, and they interact and optimize and the, you know, if they have some sort of agenda. So, I think that's, that's the part that I do not really know, we should ask biologists and people who are really studying this, the My point is that whatever we have in the cell, those parts are interacting all the time, and they are interacting with the outside world cell has definitely goals, it is trying to optimize its own inner state endogenous state, which which is like a thermostat or whatever it is trying to keep its parameter within parameters within within some ranges. And it is avoiding dangerous from the outside world, if it is acid or something in the vicinity, it moves from this dangerous area, and so on. So cells are really trying to do something in the world, they are not passing them, and they are not just moving at random, they have some goal. And I think Michael Levin explained very nicely, that we have some sort of fear of, of goal directedness in science, because when Aristotle was talking about this final goal and things are are moving towards this goal, people were criticizing his his way of explaining especially physics in terms of final goals. So, somehow with Galileo, we had this feeling that we will stop to talk about why things are happening, we will just say how what is happening. So but now today, we have normal scientific explanations for goal directedness. So it is completely possible to understand what the cell doing, based on experiences from from its past, its learning process, because the cells are able to learn they are learning we can see that. And we can also say that it's evolutionary past is also sort of learning on on the long time scape
1
Speaker 1
19:27
or would you say that cognition can occur without consciousness? Or would consciousness enter any sort of being that has cognition automatically? What is the relationship between cognition and consciousness? I guess it's the question
2
Speaker 2
19:50
is a very good question. I have never been studying consciousness really. My first allergic reaction was When people say the whole universe is consciousness, then I think you postulate the notion that you want to explain, you just say okay, it is everywhere. So it is. But what I would like to understand if we can derive consciousness from processes happening in nature, to understand somehow consciousness in terms of something more primitive, more basic. So I really, I will tell you how I'm thinking about it, even though I am by no means an expert on that. But I'm interested in cognition, which means ability of an agent to take my perspective is information, computation, embodied information and computation. So I have an agent, and it could be living system, it could be a robot, or whatever cognitive system, and this cognitive system learns about the world, through the signals through the data coming from the outside, it's always some sort of interchange interaction between the cognitive agent and the environment. And so what would be cognition on the level of a cell? Is it possible to imagine that cell behaves the way it does, without postulating any cognition, just thinking of it like an automaton. So is that think in terms of physics being automatic process, you cannot make so much choices about what two electrons will do when you collide them, or when you it seems pretty much predefined, that it repeats itself, always, chemical processes have a little bit more combinations and possibilities, but also happen quite automatically. So when you see something coming into picture, which seems to be meaningful behavior, it's on on the on the level of biology, when you have biological cell, you can clearly distinguish the behavior which is called directed, where the cell is solving the problems and increasing probability to survive, and so on for. So I would, I would say that it is possible to understand current cognitive behavior of basic organisms like like a cell, without postulating any consciousness as some extra in the model. And we had a discussion some years ago, on this topic. And my claim was that consciousness would be possible first, when we have some kind of nervous system, where the agent has a possibility for self reflection for self self representation. But you can approach it in in many different ways. But at the moment, I think it's, it's not very useful to assume that consciousness is in everything, because it doesn't explain anything. So I'm not pan psychist. But I'm Pan computational is. A lot of discussions about that. So my framework is computing nature. And I assume that everything that is physics, that moves that is changing all the time, is computing, and it's computing its next state. So it's not my own idea. It's an old tradition from from many generations of physicists. Thinking about expressing what is happening in natural sciences, with help of a computation and information and those are very basic processes. And then you use them to understand how the knowledge is generated in agents, in living agents and in artificial agents and so on. So consciousness is really a tricky question when it arrives. Yeah, I think you are much better at had that far?
1
Speaker 1
25:02
No, you're quite right. So what's in store for you next? What are you working on right now?

25:10
Right now I'm planning for

25:12
a book. Yes,
2
Speaker 2
25:15
it was run scientific, who proposed that? I would write the book on or morphological computing or on this computing nature information computation, how it relates to living systems to evolution to, to, in a bigger scale, what I'm talking about in the article.
1
Speaker 1
25:41
So where do you? Is it robotics that you're most interested in? Or? Or is it cognition of robotics? Or both? Or what?
2
Speaker 2
25:54
Yeah, well, I think my first ideas come from attempt to understand natural sciences up to biology and cognitive science in the same framework. So I started as a physicist, you know, I have I have education in physics, and spend 10 years as a physicist, and then worked 10 years for industry, and then worked 10 years as as a computer scientist, and then Teddy horses, as a cognitive scientist. And just recently, I started to apply this to design interaction, design, information design, and so but the majority of my thinking through through those years, was based on physics. And I was always trying to go back to physics and understand how other fields relate to basic embodiment. And now in cognitive science, I'm interested in understanding those parts which are self organization, self assembly, self generation, like outpaces. And then evolution in new terms. That's a very, very exciting topic, new views of evolution based on cells, which possess agency, which which show goal directed behavior,
1
Speaker 1
27:30
where do you? Where do you think that will take us? What do you see the future for that kind of research?
2
Speaker 2
27:39
I think we will learn much more when we understand how those pieces work together. One thing is to understand how organisms work on the level of mechanisms, physical, chemical, biological, cognitive mechanisms, so that we see those levels of organization, how they emerge from each other, and how we get to the level where we understand when we see a complex behavior in an agent, we can learn from from biological agents, both how they work and that can help us to cure illness to understand the psychological makeup of a human on some level and then we can use it in artificial life, robotics, artificial intelligence, for example, a study of computation, I mentioned that all cognitive science was thinking in terms of Turing machine, let's say a symbol processing that in the form of Turing machine. However, we have learned that there are many more types of models of computation that could be interesting for modeling of for example, cognition and intelligence. We would like to have efficient computation, dynamic computation, adaptive, intelligent cognitive and everything like that. So, Turing machine has no boundaries to resources, it has limited resources, and every living organism, which is solving their problems have very scarce resources and they are saving their energy and space and everything. So they are computing in a completely different way. So we would like to learn to compute with small resources were a big because we are using more and more can computing in everyday life everywhere, and it costs a lot of energy and then emits a lot of energy in the environment. We would like to have much better computers And we hope to learn from nature how to compute more effectively and more intelligent and more cognitive, and so on. So that's one thing. Also, information processes, information storage, memory, all those mechanisms that we have in nature, we would like to learn and apply for our own purposes. So one part of the idea to learn from nature is to learn how to better compute how to better manage information, but also and that is, what what other people are doing in medicine and psychology and other fields, to understand the better processes that are going on, in our bodies and in our brains. So it's both both applications and really understanding what's going on. But what is happening, you're not we are changing the basic definitions. So not now, cognition, that's not only human cognition on the level of language, even when we talk about the human, we think about all the cells of the organism, not only nervous cells, in the brain, and nervous system, but also all the cells which can be like, like immune system, it is, it is computer it is communicating with, with the brain constantly, and and also all this connection with the rest of the body, you know, so we would like to really understand how all those parts work together and what is happening. And certainly, we will know better to help people to cure problems of all kinds. If we understand how those processes work, and how body connects to mind, or thinking or psyche, those very complicated terms.
1
Speaker 1
32:14
Right? So and a little more personal level, who could you tell me who were sort of the three most important influences in your life, people or theories, particularly people where they're sort of, could you point to three people in your life, who, in some ways really, really influenced you?
2
Speaker 2
32:40
I can tell that my parents were the first people who influenced me considerably because my mother was natural. Or she was a chemist, oh, and she was a typical, prototypical, natural thinker. So for her, all the explanations were somewhere in nature and nature was the idea. And my father was economist, and he, he was looking at, at the world through the lens of relationships between people, and social relationships, in the social groups, and so on. So those were two different polls. In my development, I never really decided to be only on the on the natural sciences, but but I was mostly during my whole life. A lot of my interest was on natural sciences. But the people are also important that I'm interested in ethics. So I am teaching ethics. And then I'm thinking ethics. And I think without ethics, things don't make sense. So we need to have this value system and the ongoing discussion about what we are doing what what we want to solve, what what we want to have in the future, and so on. So those are those natural sciences, human poles, two of them. And then I had an uncle who was an engineer. And he was my third very important influence in life. He could he was in electrical engineering. And at that time, all times when I was a child, he would get TV sets, old, old TV sets made with the vacuum tubes, from our neighbors and other people. They were always having problems with TV sets, they didn't work properly. So he will take a TV set open it and they tell me how he was reasoning where the problem must be. So it was the power of logic. And he was able it He was guiding me through his own steps of reasoning how to pinpoint which of the tubes must be wrong. And I was totally fascinated by his ability to change this to and then it works. So I started, that was my fascination for engineering. I still have it with me. It's like a magic, engineering when it works at best. Of course, yeah.
1
Speaker 1
35:33
You mentioned ethics. Yeah. So in this work that you're doing with computations, and building better and more clever robotics, what sort of ethical issues have come up for you?
2
Speaker 2
35:51
There is a lot of ethics connected, especially with AI. And that's a very, very interesting development. So I remember when I started teaching ethics, around 2000, I introduced this course at the university and people were asking me, why should we need this course, we have so many other important things to teach students like computing heat computing in this way or that way. And I was telling that, trying to argue that this makes perfect sense. Because computing is becoming so pervasive, so central in our lives, it is changing everything. And it has shown with time, that it really affects us so much as civilization and as individuals, using internet using mobile phones, texting, everything changes our relationships, and our understanding of ourselves and communication, everything. So I think it was really meaningful to start to start discussing with classes of computer scientists or future computer scientists. What is the roll up of science? What What should we do? What are the problems? Where do we ship see challenges and so on. And with AI, becoming stronger and stronger, many people started to understand how powerful this technology is. And other people don't really know what is in those boxes that our students are building. So it's really important that they know and they think about it in the first step. So that was my motivation to start to start and continue with with teaching ethics to computer scientists, in generations. And I see that many, many of my students from from before now professors at the university are using this thinking about they are AI professionals thinking about ethical, ethical aspects of their work. And I think it's really very, very good. It makes very good sense. For example, we have now a huge discussion, discussion about chat to GPT. How to manage and how to relate to chat GPT, which generates perfectly reasonable text, it sounds very reasonable, and could be wrong as well.
1
Speaker 1
38:38
Yeah. But by the way, you mentioned Michael Levin, he has used that app to do some incredible art. And he has got some beautiful, beautiful pictures, which are like, right out of the alley, or, you know, all kinds of like, all you have to do is talk into that app and tell them what you wanted to do. Right. And it will create that, whether it's art, or as you mentioned, essays or anything like that. So there are some huge ethical problems involved in that. Huge.
2
Speaker 2
39:22
Yeah. And it's very, very good and interesting that everybody's talking about it. We had already workshops at the university, where the teachers were discussing and telling how to relate to this, because we have got from students, our students send the mail and said, this mail is generated by chapter 50. And by the way, my essay is also generated, or in some way, and
1
Speaker 1
39:50
yeah, so are you still teaching? Yes. What is it that you enjoy about teaching?
2
Speaker 2
39:59
Oh, For example, in those courses in ethics, we have a lot of discussions and we work together. It's like a workshop. I had one lecture this year, last year about my idea of teaching is like Leonardo da Vinci's workshop with his colleagues, painters painting together. And we were really working during those years, together, publishing together at conferences, and other publications, journals, and so on. And together students with me, so then groups, and so on, we have a lot of publications them together. And I think I learned so much, as much as they learn, we learn from each other. And they are so bright, and they can be new ideas, and they are quick, and they are full of energy. So it's really very rewarding.
1
Speaker 1
41:01
Very good. Very good. So you have been in this, you have been teaching and you have been in academia now for a long time. How has it influenced your own life? Like, what difference has this kind of work had on you as a person?
2
Speaker 2
41:25
I think when I was very young, I was thinking about and an ideal life in a monastery. Really? Yeah. reading the literature and so okay, but it's quite close. So somehow, you are very much devoted to what you are doing. And it is through many, many, at least in my case, for many, many years, there is a red line from the first models of atomic nuclei in shell model in quantum mechanics, until today, when I think about physics itself, organization and the information processing everything. So I think it's, it's all somehow connected, and you are always connected with it. So that there is no free time actually, it's always there. And my daughter remind me is the other day, she said, Do you remember when I was little that I asked you? Mommy, where are you gonna? Sometimes I would lose her. I was just thinking, Oh, something very important. So my biggest fear was that I would lose this connection with the real world with with with my closest ones. So I was always thinking very intensely not to those. That was important.

43:04
How many children do you have?
2
Speaker 2
43:06
I have two, I have Luca who is 43. He is computer scientist and big AI enthusiast. And Thea she is 31 and she is bioinformatics. Titian, and very much interested in helping people in curing diseases and so that's her dream working for Swedish Health Authority right now. So they are in this epidemics and such things.
1
Speaker 1
43:44
So do you do you ever do you ever find time to relax?
2
Speaker 2
43:50
Yes, yes. Yeah. Nowadays I have three girls girl grant -child children. Yeah. So why do you five years and eight years and I relaxed with them playing with them. It helps me forget everything. So they are totally absorbing me. And I used to have yoga classes to yoga classes. Sometimes I was painting that was a long period in my life I was painting in oil and you know when you paint in oil, it is a very slow slow process and you must relax and you must focus on it. So I was playing piano some when I was a young person. In my youngest years I was playing piano and that was also a fantastic way to to relax. So it's it was different periods in my life where in a relaxed in different ways, right now it's mostly yoga and my grandchildren playing with them.
1
Speaker 1
45:12
So my final question, taking that into consideration, and all the work that you have done, what does being human mean to you?
2
Speaker 2
45:27
Well, I think being a human is a real privilege. It gave me the possibility to meet people, exciting people, and to connect to people from past and even think of our common future for the humanity. This feeling is a part of a big, big humanity. I think there are so many exciting people in the world. And you can read about what they are doing, you can watch their works in art, you can find really amazing, amazing human results. And so I think it's so it feels like a big privilege to be able to connect to the rest of the world as a human. I don't know how it will feel if I was be elephants, but I think it's good to be human.
1
Speaker 1
46:42
You're enjoying being human? Yeah. Good. Good. Very good. Well, I enjoyed meeting you. It was a real privilege. Thank you for spending some time with me and our listeners. I think you have given us a lot to think about. And I will promise I will read read your paper and hope to be in contact with you in the future, if I may.
2
Speaker 2
47:11
Absolutely. Thank you so much. It was so nice to talk with you and the privilege to be part of this podcast.
1
Speaker 1
47:19
Thank you. May I just say that my guest in two weeks will be Professor David Peters, director of the Westminster center for resilience, Faculty of Science and Technology, University of Westminster, London. Our subject will be the neurobiology of resistance. Thank you once again, Dr. Dottie Cranko edge, and thank you to all the listeners. See you next time. Bye bye.