Transcription
What does it actually mean to have intelligence? How do you recognize intelligence? We evolved with a very specific cognitive system and set of senses. We are obsessed with this three-dimensional world. And so we are oblivious to all kinds of minds that are not like ours: intelligence in plants, in cells, in unicellular organisms, and even in minimal matter, which are materials that are far simpler than a cell or anything like that.
Intelligence is not about neurons or what brains do. What matters is the ability to pursue certain goals. Whatever type of being you are, one of the things you're going to have to do is reach those goals even when things change. Solve a new problem using the tools you already have. In my lab, we focus on this problem-solving aspect to try to go beyond our native prejudices about what kinds of things can be cognitive and try to recognize those same behaviors in very um unfamiliar novel contexts. The field of diverse intelligence seeks to understand embodied minds, specifically what kind of physical bodies can host different types of cognitive architectures. You can't just paint beautiful complex minds onto everything you see in nature. And so fundamental to diverse intelligence is setting up experimental situations where we can determine how much ingenuity does a particular system, no matter what it's made of, no matter how it got here, what kind of skills does it have to reach the outcome.
Intelligence has many definitions. We focus on William James' definition and James said: intelligence is a degree of competency to reach the same goal by different means. So from that perspective, I think the majority of the field feels strongly that intelligence goes all the way to the cell level. Individual cells are very good at pursuing tiny little agendas. Very very basic forms of goal-directedness in many different spaces. Maybe physiological-state based, metabolic states, gene expression states. You know, a parramium-like thing swimming around the ocean has goals, preferences, competencies, sensing, decision-making, processing on a very small scale. But life is really good at scaling that up.
From single cells to groups of cells, what's important to track is what I call the 'cognitive light cone'. The cognitive light cone represents an attempt to delineate the size of your goals. If you tell me that all of your goals are around maximizing the concentration of sugar in a tiny little, you know, micron-sized area and you have a memory that goes back 20 minutes and maybe some predictive capacity going forward, I'm going to guess you're a bacterium and you have a tiny little cognitive light cone. And if you tell me that you have goals around the global financial markets across the whole earth, I'm going to say you're at least a human. By comparing the size of the cognitive light cone, you can make a map of diverse intelligences and how they fit relative to each other.
When you started life as a group of cells, there were many, many possibilities about what shape they should make. In order to complete your journey to be whatever you are, the cells create a system that has goals that its individual parts don't have. The collective intelligence of a group of cells, can pursue much larger goals and, in fact, project them into new spaces like anatomical space. For example, a set of cells building an amphibian limb knows things that no individual cell knows. It knows exactly how many fingers it's supposed to have. And and you know it does because if you get between it and its goals, for example, you amputate the limb, it will work really hard and it will regenerate to the same pattern. So, it's a homeostatic set-point, like in your thermostat, it's a homeostatic set-point that all the cells work to. When they get there, they stop. And the way that happens in in embryogenesis is simply that the cells connect with these electrical synapses that that allow signals to to propagate back and forth. There are all kinds of interesting dynamics as the voltage changes throughout the network that allows groups of cells to make decisions about what shape they should they should make. And those decisions are mediated electrically using exactly the same processes as go on in the brain. Exactly the same. Bioelectricity isn't just more physics that you have to keep track of. It's actually the gateway to the mind of the body to the primitive cognition of of of cells and tissues.
Because of our emphasis on goal-directedness, a lot of experiments in our lab are trying to understand how cells find solutions to novel problems. This is every IQ test that that anybody's ever taken. They say solve this problem creatively using this this set of objects, right, problem you haven't seen before. What a lot of people think is that the growth of your body through anatomical space is a clockwork progression. That there are sets of rules determined by the DNA and just by sort of cranking through those rules step by step something complex emerges and you go from from here to there. Yeah, under normal situations it always solves it the same way. But if you change the environment, you change the parts themselves, you find an amazing capacity to reach those goals even when things change. For example, we made tadpoles with eyes on their tails. So the only eye in this animal is on the tail. Those animals, it turns out, can see. You would think that if you're going to radically change the sensory motor architecture of a creature, you need many rounds of new evolutionary selection. You need mutation, you need adaptation, you need selection to make this thing work. Those animals see the first time you make them. Because you're already starting with a problem-solving system that's very prepared for novelty. This is a part of the wisdom of the cells and tissues, this problem-solving that we still don't understand.
As humans, we try to simplify everything. Cast this binary kind of distinction between intelligent and non-intelligent or cognitive and non-cognitive. I think these binary categories are really doing a lot of harm. I think what's much more useful is a continuum view where you ask what kind and how much. And so to gauge intelligence, you have to make a guess, a hypothesis as to what problem space is this thing even navigating. That requires a lot of visualizing what intelligence would look like at other scales of time and space. You almost always win and discover something new by skewing high-intelligence than you do by skewing low. Meaning assume higher levels until proven otherwise. Based on where the science sits and where the experiments have have led. There is a demonstrated degree of learning and some degree of problem-solving that go down at least as far as molecular networks. As far as I'm concerned, even you know humble particles, they can do interesting things on that cognitive spectrum. On that I think I'm I'm somewhat more radical than than many people in the field.
As far as how far up it goes, sometimes people ask me, they say, "Then you would even say the weather is intelligent." And I would say I I wouldn't just say that, but have you ever tried to train a hurricane? That's an empirical question. It's difficult to do the research, but it's not impossible. I'm certainly not saying that any of the things I'm talking about are inconsistent with molecular biology or with physics. I'm saying it's a research program to see which level of description and what tools are giving you the most bang for your buck. You can't do that from a philosophical arm chair, you have to do experiments. That that is the scientific method and and I'm proposing it to be judged by the same exact criteria that everybody else is judged by. What what did you find using that method? ] The biggest limiting factor in this field, I think, is is imagination, our imagination. And I think it's the same for all sciences, but in particular here, because we're just not prepared for this with our cognitive system. Our ability to to recognize and communicate with novel intelligences depends on us getting beyond our our evolutionary biases.