Recognizing Drawings: A Human or Animal Ability?

[matthyaouw]

It occurred to me the other day that if someone draws a somewhat abstract picture of an object or animal (like a cartoon drawing), the majority of people are capable of recognising it as what it was intended to be, even though it bares little similarity.

Look at this picture: http://www.rainingdaisies.com/Cat cartoon.jpg
Most people I think would agree that it shows two cats, one with a mouse on its tail, even though cats look very little like that. Is this a learned response, or one inbuilt to the human brain? Would someone familiar with cats who had never seen any form of cartoon drawing be able to identify them? I think its safe to say this kind of recognition has been around in humans for a while, as early ones left cave paintings.

Furthermore, are animals capable of recognising abstract representations like we are?

I do wonder what implications the ability to recognise such things and make associations would have on the development of intelligence.

 

[Ouabache]

Fascinating questions! Wish I had an insightful reply.
Perhaps you may want to pose these questions over in topic
https://www.physicsforums.com/forumdisplay.php?f=85 as it sounds like ones I've read in psychology and anthropology.

 

[gerben]

It occurred to me the other day that if someone draws a somewhat abstract picture of an object or animal (like a cartoon drawing), the majority of people are capable of recognising it as what it was intended to be, even though it bares little similarity.

Look at this picture: http://www.rainingdaisies.com/Cat cartoon.jpg
Most people I think would agree that it shows two cats, one with a mouse on its tail, even though cats look very little like that. Is this a learned response, or one inbuilt to the human brain? Would someone familiar with cats who had never seen any form of cartoon drawing be able to identify them? I think its safe to say this kind of recognition has been around in humans for a while, as early ones left cave paintings.

I think it has something to do with categorizing things and with us being a very visual species. Things that belong to the same category can look different. There are very different looking kettles, houses, trees, cats etc. However, we still know something belongs to one category or the other, presumably because it has certain distinguishing characteristics. A drawing does not have to be similar in every detail. It just has to hint at these distinguishing characteristics. The cat drawing that you linked to displays a lot of the characteristics that a cat sitting in that pose would have: the global contour, the pointy ears, the whiskers, the contour of the upper “lips”, the front legs, and the tail. Also cave paintings look very much like the contours of the thing depicted, for example a bison (whether the size would fit depends on your distance from the painting and on your distance from the real thing, both of which should be seen as undetermined and unimportant in an abstract sense).

It may have given some impetus to us becoming a communicating species. When you inform someone else of something that you have seen you use an abstract symbol (a sound) to represent the thing you saw. This implies you are able to associate very different sensory qualities (i.e. sounds with visual impressions). It would probably not have been feasible to associate every small detail; it was better to stick to important distinguishing characteristics. The abstract symbols that were to be used needed to point out important differences. They should not point out the exact shape and colour nuances of the lion’s manes, the simple fact that it is a lion would be enough information. In any case the symbols should point out information that is similar to the information that one would use to guide ones own muscles.

Furthermore, are animals capable of recognising abstract representations like we are?

I guess most animals, like us, do not depend one every aspect of the things in their surroundings for identification. They probably also depend on certain important defining characteristics, but those may not always be the same characteristics that we use, for example many animals use smell much more then we do.

I do wonder what implications the ability to recognise such things and make associations would have on the development of intelligence.

I would assume that language and particularly being able to use written language depended a lot on such abilities.

 

[Ouabache]

extracting useful information

How our brains extract information from images and abstract drawings is also a primary focus of pattern recognition, one of the basic areas of study for AI (artificial intelligence) and ANN (artificial neural networks).

To give a feel for this... Try and imagine the number of statistical analyses of data that our brain runs, using a form of parallel processing, to compare features of abstract images, to all the memories stored our brain. If it finds a reasonably good match, we can tell what the image is. We do this quite well. I would guess other animals use pattern recognition. Interesting sidenote, psychiatrists & psychologists also attempts to use this interpretive sense of our brain (Rorschach Tests) to find subconscious psychoses.

 

[zoobyshoe]

Try and imagine the number of statistical analyses of data that our brain runs, using a form of parallel processing, to compare features of abstract images, to all the memories stored our brain. If it finds a reasonably good match, we can tell what the image is. We do this quite well.

If you think about it, the odds against seeing anything from the same angle a second time that you did the first time are astronomical.

In order to "recognise" anything, the brain has to be able to accurately do some fantastically complex extrapolations from the memories of previous viewings.

There is a lot more to the experience of recognition, though, than all this amazing computation: At each stage, the hippocampus and surrounding organs are constantly alerting us to the "familiarity" of a given image, with just the right dose of physiological response. This is, in effect, an emotion, but one which is usually so finely tuned to the object or experience being analyzed, that we don't percieve it as a response we are having to the phenomenon. We believe that "familiarity" is a property of the object itself.

In some neurological patients this function of the hippocampus breaks down, and it fails to alert them to the "familiarity" of things. Neurologist V.S. Ramchandran has studied and analyzed a couple cases like this. These patients come to believe that familiar people and objects have been replaced by "imposters". They look and act exactly the same, but they "feel" completely strange and unfamiliar: they fail to evoke the usually concommitant "familiarity", we take for granted.

Since we have come to believe "familiarity" is a quality that belongs to the object of our attention, instead of being something we add to it, people whose hippocampus fails to kick in are lead to the conclusion that the once familiar thing has been replaced by an exact duplicate. This seems to be the chronic way people who suffer from this failure of the hippocampus explain it to themselves.

So there is more to the human experience of recognition than the "data processing" that adds up to pattern recognition. The limbic system has to flag the results of all that processing with an essentially emotional spice or flavor. If it doesn't, people will not believe they are actually re-experience the same thing at all.