How do animals form categories?

[coberst]

How do animals form categories?

 

Quickie from Wiki: “In metaphysics (in particular, ontology), the different kinds or ways of being are called categories of being or simply categories. According to the Aristotelian tradition, a being is anything that can be said to be in the various senses of this word. Hence, to investigate the categories of being is to determine the most fundamental senses in which things can be said to be. A category, more precisely, is any of the broadest classes of things - 'thing' here meaning anything whatever that can be discussed and cannot be reduced to any other class.”

 

Donald Schon, a researcher in the cognitive sciences, tells us of a group working on a difficult task of designing a satisfactory paintbrush made of synthetic bristles. In the middle of a discussion among the technicians designing the brush one of the group had that eureka moment and shouted “You know, a paintbrush is a kind of pump”. This insight Schon explains caused the designers “to notice new features of the brush and of the painting process”.

 

Cognitive science has introduced a new way of viewing the world and our self by declaring a new paradigm which is called the embodied mind. The primary focus is upon the fact that there is no mind/body duality but that there is indeed an integrated mind and body. The mind and body are as integrated as is the heart and the cardiovascular system. Mind and body form a gestalt (a structure so integrated as to constitute a functional unit with properties not derivable by summation of its parts).

 

The human thought process is dominated by the characteristic of our integrated body. The sensorimotor neural network is an integral part of mind. The neural network that makes movement and perception possible is the same network that processes our thinking.

 

The unconscious categories that guide our human response to the world are constructed in the same way as are the categories that make it possible of other animals to survive in the world. We form categories both consciously and unconsciously.

 

Why do we feel that both our consciously created and unconsciously created categories fit the world?

 

Our consciously formed concepts fit the world, more or less, because we consciously examine the world with our senses and our reason and classify that world into these concepts we call categories.

 

Our unconsciously formed categories are a different matter. Our unconsciously formed categories fit our world because these basic-level categories “have evolved to form at least one important class of categories that optimally fit our bodily experiences of entities and certain extremely important differences in the natural environment”.

 

Our perceptual system has little difficulty distinguishing between dogs and cows or rats and squirrels. Investigation of this matter makes clear that we distinguish most readily those folk versions of biological genera, i.e. those “that have evolved significantly distinct shapes so as to take advantage of different features of their environment.”

 

If we move down to subordinate levels of the biological hierarchy we find the distinguishing ability deteriorates quickly. It is more difficult to distinguish one species of elephant from another than from distinguishing an elephant from a buffalo. It is easy to distinguish a boat from a car but more difficult distinguishing one type of car from another.

 

“Consider the categories chair and car which are in the middle of the category hierarchies furniture—chair—rocking chair and vehicle—car—sports car. In the mid-1970s, Brent Berlin, Eleanor Rosch, Carolyn Mervis, and their coworkers discovered that such mid-level categories are cogently “basic”—i.e. they have a kind of cognitive priority, as contrasted with “superordinate” categories like furniture and vehicle and with “subordinate” categories like rocking chair and sports car” (Berlin et al 1974 “Principles of Tzeltal Plant Classification”; Mervis and Rosch 1981 Categorization of Natural Objects, “Annual Review of Psychology” 32: 89-115))

 

The differences between basic-level and non basic-level categories is based upon bodily characteristics. The basic-level categories are dependent upon gestalt perception, sensorimotor programs, and mental images. We can easily see that these facts make it the case that classical metaphysical realism cannot be true; the properties of many categories are mediated by the body rather than determined directly by a mind-independent reality.

 

“Try the following thought experiment: Close your eyes and picture a chair. Now, close your eyes and try to picture a furniture. You cannot—at least not one that isn’t a basic level object such as a lamp, table, or chair. The reasons are, first, that one can perceive lamps, tables, or chairs in terms of a single overall shape, but there is no overall shape for pieces of furniture in general…Second we have special motor programs for interacting with basic-level objects such as lamps, tables, and chairs but no motor program for pieces of furniture in general.”

 

In humans basic level categories are developed primarily based upon our bodily configuration and its interrelationship with the environment. For other animals almost all, if not all, categories are basic-level categories.

 

Quotes from A Clearing in the Forest: Law, Life, and Mind by Steven L. Winter

[Rade]

....“Try the following thought experiment: Close your eyes and picture a chair. Now, close your eyes and try to picture a furniture. You cannot—at least not one that isn’t a basic level object such as a lamp, table, or chair

?? Sure I can, I can very easy picture a furniture, it is because the word 'furniture' is based on a specific 'definition' of a 'concept'. Try the following thought experiment: close your eyes and try to picture a flying machine--now try to picture a furniture--see how easy it is for your mind to so distinguish the two. Now, why is this so ? It is because there is an "overall shape for pieces of furniture in general" that your mind can interact with that differs from the "overall shape for types of flying machines in general". That is, the human mind has the ability to form concepts of things for both primary (species level) and secondary (genus level) categories of substance.

 

Thus I find as false this comment you make:

.. we have special motor programs for interacting with basic-level objects such as lamps, tables, and chairs but no motor program for pieces of furniture in general.”..

Whatever "special motor programs" you refer to, the human species does so have them to interact at multiple levels of categories (1) species or primary level: chair vs jet, (2) genus or secondary level: furniture vs flying machine.

[Boerseun]

Rade, I think you're missing his point a bit.

 

But then again, I fail to see the mystery here.

 

Of course you can't picture a "furniture". "Furniture" refers to a category, and not to an individual item. It could be a chair, a table, a book case, a lamp. Anything. You cannot interact with a category. I fail to see the mystery, here.

 

Animals, on the other hand, probably also categorize - albeit at a very low level.

 

Consider dogs, man's very best buddy:

 

Anecdotal evidence abound about dogs who know how to "fetch" when a stick is thrown. Don't throw a stick and still say "fetch!", and the dog will be off and find something that's got roughly the same dimensions as the stick he's used to you throwing. He won't come back with a rope, for instance. Also, they categorize things into what's eatable or not - male dogs will mark their territory on just about anything that's nose-high (fellow dog's noses, of course). But not on edible stuff. Even when a dog just ate and has a full belly, he will not urinate on food. There must be at least some rudimentary categorization taking place there.

[lawcat]

Boerseun,

 

coberst says that you can not picture "furniture" because we have motor program for picturing only simple shapes, thus only lamp, or chair, or table. But that is patently untrue to prove some desired conclusion.

 

Furniture is defined as any one of the objects of that class or category, or a grouping of objects of that category.

So if the conclusion rests on inability to imagine single furniture, then it is incorrect because it is based on the premise that one can imagine a chair or a table.

 

If the conclusion rests on the inability to picture a group of objects within the category of furniture, it is also untrue, because I can certainly picture my room which contains a group of objects within the category of furniture.

 

But the problem is one of contextualization. I contextualize my images in accordance with my memory of them. Everything is always contextualized. It would be hard to picture a "frusthok" which I have no perception, knowledge, or memory of. But if I have perception of something, it is usually in the context because that is the nature of the world we live in. However, in addition to context, with a little hard work and focus I can isolate the object in my mind and think of it in abstract. And if I am a bit more brilliant, I could isolate groupings of objects within or without the same category.

 

The conclusion above is of no interest. The conclusions simply states that our mental images are contrained by our sensory ability and their configuration on the body, as well as the context that we perceive them in. Big deal. I probably knew that in grade school, in 3d geometry.

 

But the problem is more interesting if the focus of the post above is animals, non-humans. However, the post departs from anything informative about animals.

[coberst]

I suspect most animals categorize but the human animal has the ability to create abstract ideas from which we have the ability to create meaning. We are meaning creating creatures and thus have filled our world with categories that are about abstract concepts that we are willing to live, die, and kill for.

[Rade]

I would agree with you Coberst that animals do not have the ability to form abstractions. But, I may error as I am not well informed of research in this area. Here is one way to test the hypothesis--present an animal (say a Bonobo, most closely related genetically to humans) with say two apples and two grapes (could be anything). If anyone can provide a published paper where the animal was able to understand the concept "two" --then we would have evidence that not only humans have ability to use abstraction as a form of mental focus.

[Miranda]

I have to agree with Boerseun on this one.

 

Here is one way to test the hypothesis--present an animal (say a Bonobo, most closely related genetically to humans) with say two apples and two grapes (could be anything). If anyone can provide a published paper where the animal was able to understand the concept "two" --then we would have evidence that not only humans have ability to use abstraction as a form of mental focus.

Ask and you shall receive. There are many studies done on primates to invesigate their ability to symbolize, categorize, and communicate overall. I suggest reading about Kanzi, Koko (gorilla), Washoe (chimpanzee), or Nim Chimpsky, to name a few. (I don't usually condone citing only Wiki for research, but I'm going to do so for the sake of simplicity.)

 

Actually, the research for embodied cognition is quite interesting. George Lakoff is a good start if you want to know more about the subject.

[Buffy]

I would agree with you Coberst that animals do not have the ability to form abstractions.

As I've said many times, I think this sort of anthropocentrism is absurd, unsupported, and continues to lead many down a path that not only ignores the abilities of animals, but leads to major misunderstandings of human cognition.

 

I'd ask you all to be a little bit open minded.

 

Especially because:

Here is one way to test the hypothesis....

In addition to the links Miranda gives above, run down to the store and pick up the September 2009 issue of Scientific American and look at the leader "Animals by the Numbers":

Recent studies...have uncovered new instances of a counting skill in different species, suggesting that mathematical abilities could be more fundamental in biology than previously thought. Under certain conditions, monkeys could outperform college students...

The article cites not only experiments with monkeys, but robins, baby chickens and even bees!

 

Never underestimate your own ignorance, :)

Buffy

[Rade]

I'd ask you all to be a little bit open minded.

Thank you for information about ability of abstraction in animals, but recall I did post "but I may error", thus your call for me to be "open minded" seems misplaced. I enter these forums to learn and offer my thoughts on issues--many times I am incorrect in my thinking.

 

To this topic --what exactly is an "abstraction" ? The definition is of great importance. How do we know that any of the experiments you cite in reality have evidence of "abstract" thought in animals ? I find these papers a topic to discuss, not take as truth handed down from above. So, until I read these papers I will have an open mind, open to agree or disagree with the data presented.

[coberst]

Thank you for information about ability of abstraction in animals, but recall I did post "but I may error", thus your call for me to be "open minded" seems misplaced. I enter these forums to learn and offer my thoughts on issues--many times I am incorrect in my thinking.

 

To this topic --what exactly is an "abstraction" ? The definition is of great importance. How do we know that any of the experiments you cite in reality have evidence of "abstract" thought in animals ? I find these papers a topic to discuss, not take as truth handed down from above. So, until I read these papers I will have an open mind, open to agree or disagree with the data presented.

 

 

“Subject” is a word defining the human locus of consciousness, experience, reason, and will; “subject” is our human “essence” and everything that makes us who we uniquely are.

 

“Object” is a word defining something material that may be perceived by the senses.

 

I define these two words because there are many different definitions for these words but this is the definition I am using for these words in this discourse.

 

The qualities acquired by objects through association are what are normally called their expression. In the form or material of objects “there is one object with its emotional effect, in expression there are two, and the emotional effect belongs to the character of the second or suggested one.”

 

Many or perhaps most “things” that we find to be meaningful are objectified abstract concepts.

 

Examples of concrete concepts: the infant feeling warm and secure while being held following birth; being repelled by foul smelling stuff; the burden of carrying heavy stuff; observing the rise of milk in the measuring cub while watching mother make corn bread; noting that Grandma needs support while walking; getting knowledge while examining a tree.

 

Examples of abstract concepts: feeling warm when around my best fiend; telling a friend that the movie stinks; the feeling of being weighed down by troubles; the sense that stock prices are too high; I feel good when I support the troops; seeing a distant problem that might result.

 

Most of our concepts are abstract concepts and they are constructed from our concrete concepts. The more complex, broad, and sophisticated that our life style is the larger, more complex and sophisticated are our abstract concepts. Most of our concepts dealing with freedom, morality, politics, religion, justice, and history are abstract concepts.

 

Quotes from The Sense of Beauty: Being The Outlines of Aesthetic Theory by George Santayana

[Buffy]

...thus your call for me to be "open minded" seems misplaced.

Please do note I said "you all" and my admonition was mostly directed at those whose statements on the topic are much less precisely qualified than yours was. Thank you for doing so! :hyper:

To this topic --what exactly is an "abstraction" ? The definition is of great importance. How do we know that any of the experiments you cite in reality have evidence of "abstract" thought in animals ? I find these papers a topic to discuss, not take as truth handed down from above. So, until I read these papers I will have an open mind, open to agree or disagree with the data presented.

I would point you to the article and others on the topic, but the specific answer is that it has to do with the relational concepts of math that go beyond simply counting: that is, it can be argued that two is obviously greater than one, but these animals have been observed making rational decisions about much larger numbers. For the robin, 12 worms in one hole versus 11 in another results in a rational decision to prefer the hole with 12. With the chicks, they were trained to associate sets of objects with "mom" and "dad" where "mom" had just one or two more objects in the set. Even when several of the objects were removed from both sets, the chicks continued to associate "one or two more" with "mom" even though the counts were completely different.

 

With monkeys and animals with larger brains, this sort of mathematical reasoning gets far more sophisticated.

 

Cats are smarter than dogs. You can't get eight cats to pull a sled through snow,

Buffy

[Kriminal99]

I suspect most animals categorize but the human animal has the ability to create abstract ideas from which we have the ability to create meaning. We are meaning creating creatures and thus have filled our world with categories that are about abstract concepts that we are willing to live, die, and kill for.

 

Given the nature of neural networks, and observation of how this system operates, I see little room to differentiate between the ability to form your basic categories and the ability to form parent categories.

 

An object in the physical world is less basic than you make it sound, and is itself an abstraction. The various angles it must be recognized from, it's physical properties (which animals often are and must be aware of) etc must be contended with.

 

From the neural network perspective, the issue is the ability to relate an entire pattern or grouping of connected neurons with another entire pattern. This ability must be realized in animals for them to be able recognize objects, as described above. This ability is unable to distinguish between what information is stored in those patterns, which can be composed of other patterns.

 

From a functional perspective, many abstract concepts can be created using the same functions needed to create your basic categories, operated on the basic categories.

 

I honestly think animals don't do it because they don't want to. I think intelligence evolves from a greater perceived need for intelligence.

[coberst]

What are Abstract Ideas?

 

Is “soul” the reification of an abstract idea?

 

“Abstract ideas, for the most part, arise via conceptual metaphor—a cognitive mechanism that derives abstract thinking from the way we function in the everyday physical world. Conceptual metaphor plays a central and defining role in the formation of mathematical ideas within the cognitive unconscious—from arithmetic and algebra to sets and logic to infinity in all of its forms.”

 

An example of the abstract concept ‘A purposeful life is a journey’ is constructed from several primary metaphors. These primary metaphors are concepts developed from living experience.

 

We are acculturated to recognize that a useful life is a life with purpose. The complex metaphor ‘A Purposeful Life Is a Journey’ is constructed from primary metaphors: ‘purpose is destination’ and ‘action is motion’; and a cultural belief that ‘people should have a purpose’.

 

’A Purposeful Life Is A Journey’ Metaphor

A purposeful life is a journey.

A person living a life is a traveler.

Life goals are destinations

A life plan is an itinerary.

 

This metaphor has strong influence on how we conduct our lives. This influence arises from the complex metaphor’s entailments: A journey, with its accompanying complications, requires planning, and the necessary means.

 

Primary metaphors ‘ground’ concepts to sensorimotor experience. Is this grounding lost in a complex metaphor? “Not at all.” Complex metaphors are composed of primary metaphors and the whole is grounded by its parts. “The grounding of ‘A Purposeful Life Is A Journey’ is given by individual groundings of each component primary metaphor.”

 

The ideas for this post come from “Philosophy in the Flesh”. The quotes are from “Where Mathematics Comes From”

[Kriminal99]

I don't see you making a distinction between weak inductive reasoning and strong or even mathematical induction. You seem to be lumping them all together under the moniker "metaphor". I feel like only human beings even make use of weak inductive reasoning, much to our detriment, and that the label metaphor is much more attuned to this than stronger types of inductive reasoning.

 

Arguments like "A purposeful life is a journey" would be completely foreign to an animal, because such statements are an unfortunate by-product of intelligent social interaction. When we discover that we can manipulate other people's emotions with weak inductive reasoning, and we are not immediately shown any negative consequences of doing so (like an inability to solve real world problems due to poor reasoning skills), we do so to satisfy our social needs.

 

Statements like the above are not directly useful, practical or applicable to real life. At best they can serve as an attention grabbing pointer to a more formal model - One that that explains how something as complex as a life can be like something that is capable of being purposeful (like a simple tool). Often times the formal model that should be pointed to by such a statement doesn't really exist or is incomplete. I see your attempts to provide some formalization of the statement, but they seem far from complete. In fact the statement embodies many things that philosophers have argued over for centuries and without resolving those disputes it would not be possible to create such a complete formal model. Example: Must a purposeful life have goals? If you arrive at a question like this that there is not a mechanically obtainable answer to, it's not a formal model.

 

Such formal models are created through a different process, making use of what is the same as mathematical induction - a small subset of your metaphors. I am sure animals do this all the time. It simply involves observing different things and noticing similarities. Then you assign a category to hold all those things, and make those similarities a property of the whole category. Then you reason based on the properties of the category. Find something that doesn't behave as it should? Then it's not a member of that category, look closer and find something to base a new category on. Animals demonstrate an ability to create categories in this manner all the time. Does your dog suddenly not know where to get its food when you switch it's food bowl for one of a different color?

 

I think the clearest difference between the two is that there is always a group of real contexts from which formal ideas are created from. With thought, I could break those formal ideas back down into a function of perceptual experiences.

 

An idea like "purposeful life" seems to be the result of just chaotically throwing one word in front of another and contemplating what the result means.

 

Maybe we have to be able to do this in order to find those formal models. It sounds like this might be what you are saying. Maybe doing so should prompt us to try and find a compatibility between the context based definition of each word and create a formal model from it. I think this particular pairing would fail any animals test of compatibility however. Only a human could be so stupid.

[coberst]

One of the things that set humans apart from the other animals is our ability to create abstract ideas. We create abstract concepts which we will live, die, and kill for. These abstract ideas are created from concrete concepts. I use the words idea and concept as being interchangeable.

 

The great truth of human nature is that wo/man strives for meaning. S/he imposes on raw experience symbolic categories of thought, and does so with conceptual structures of thought. “All human problems are, in the last resort, problems of the soul.”—Otto Rank

 

In the nineteenth century, after two hundred years of opposition paradigms, science faced the dilemma: if we make wo/man to be totally an object of science, to be as this object merely a conglomeration of atoms and wheels then where is there a place for freedom? How can such a collection of mere atoms be happy, and fashion the Good Life?

 

The best thinkers of the Enlightenment followed by the best of the nineteenth century were caught in the dilemma of a materialistic psychology. Does not the inner wo/man disappear when humans are made into an object of science? On the other hand if we succumb to the mode of the middle Ages, when the Church kept man firmly under the wraps of medieval superstitions, do we not give up all hope for self-determined man?

 

“Yet, we want man to be the embodiment of free, undetermined subjectivity, because this is the only thing that keeps him interesting in all of nature…It sums up the whole tragedy of the Enlightenment vision of science.” There are still those who would willingly surrender wo/man to Science because of their fear of an ever encroaching superstitious enemy.

 

Kant broke open this frustrating dilemma. By showing that sapiens could not know nature in its stark reality, that sapiens had no intellectual access to the thing-in-itself, that humans could never know a nature that transcended their epistemology, Kant “defeated materialistic psychology, even while keeping its gains. He centered nature on man, and so made psychology subjective; but he also showed the limitations of human perceptions in nature, and so he could be objective about them, and about man himself. In a word man was at once, limited creature, and bottomless mystery, object and subject…Thus it kept the best of materialism, and guaranteed more than materialism ever could: the protection of man’s freedom, and the preservation of his inner mystery.”

 

After Kant, Schilling illuminated the uniqueness of man’s ideas, and the limitations from any ideal within nature. Schilling gave us modern wo/man. Materialism and idealism was conjoined. Wo/man functioned under the aegis of whole ideas, just as the idealists wanted, and thus man became an object of science while maintaining freedom of self-determination.

 

The great truth of the nineteenth century was that produced by William Dilthey, which was what wo/man constantly strived for. “It was “meaning” said Dilthey, meaning is the great truth about human nature. Everything that lives, lives by drawing together strands of experience as a basis for its action; to live is to act, to move forward into the world of experience…Meaning is the relationship between parts of experience.” Man does not do this drawing together on the basis of simple experience but on the basis of concepts. Man imposes symbolic categories of thought on raw experience. His conception of life determines the manner in which s/he values all of its parts.

 

Concludes Dilthey, meaning “is the comprehensive category through which life becomes comprehensible…Man is the meaning-creating animal.”

 

Does it make sense to you that “All human problems are, in the last resort, problems of the soul”??

 

Quotes and ideas from Beyond Alienation by Ernest Becker

[Kriminal99]

When you have a formal model and definition for an abstract idea like honor, you can see that social animals often demonstrate knowledge of such ideas.

 

It seems to me that the only abstract ideas animals do not use are ones that are so complex (and yet formal) that they could only be motivated by communication, and ones that have no formal representation because they are formed from weak inductive reasoning.

 

I may not have direct access to the inner workings of my mind, but they do realize my consciousness and thus a person trying to determine the workings of their own mind is in a far better position than someone trying to understand someone else's mind. With some deductive reasoning, it is an accomplish-able task to discern how one's own mind is working. This understanding of myself that I have created does not need to transcend my epistemology. It is my epistemology.

 

Recognizing the determination of man is a moral imperative. Denying this fact is the cause of all kinds of social injustices - people taking credit for things that were really outside of their control and blaming others under the same circumstances. If I inherit a million dollars from my family, and another man inherits 5 cents, and that man buys a stick of bubble gum, and I invest 75% of my funds and spend the rest on gum, it is human nature for me to claim that I am wealthy because of good financial sense and choices. In reality such a claim is absurd.

 

Free will is a perfect example of an idea that has no formal model behind it. It was created, once again, by putting two other words together without understanding of the consequence. In this case we did so to try and support a naive understanding of our behavior - that feeling as though we can do whatever we want means we are not determined. But everything we want to do has some reason behind it, and even some reason why we want to do it more than something else even if we try to hide that reason.

 

Is free will choosing without a reason? If not then isn't it determined by the reason, which is the opposite of what free will is supposed to be? Choosing between reasons brings about the same issue - based on the reason for our choice between them. If free will is choosing without reason, how would you even make such a choice? Randomly? Would a person ever do this except for to try and prove he has free will (which is itself a reason)? Is choosing randomly not the same as choosing for a reason, just for a smaller reason among many? etc.

 

Without the ability to construct a formal model of free will, I am forced to conclude there is no such thing.

[gyll]

@Kriminal99: "Without the ability to construct a formal model of free will, I am forced to conclude there is no such thing." I'd rather sa you'd be forced to conclude that "there is/you don't know of any way to construct a formal model, and that's about it. don't get me wrong, i like your way of thinking, at least as i see it in your replied to coberst, it's just that in this particular case i think you jumped at a bit too far-fetched a conclusion.