I doubt you could do what this 7-year-old chimp does. Certainly he can’t “count,” but he can memorize an order of numbers, and faster than you can! (Go here if you want to take the test yourself.)
I had no idea that chimps had such a good short-term memory. Or maybe only this chimp does.
h/t: Peter
Amazing.
I notice that a note pops up at the start disputing the accuracy of the study. Following the link to this abstract: http://link.springer.com/article/10.1007%2Fs10071-008-0206-8 it seems that humans can do this too with an equivalent amount of practice. That doesn’t take away from the fact that it is an amazing feat for a chimp. But it would certainly be harder to explain if chimps turned out to have *better* short term memory skills than humans.
For similar reasons, this is why I’m skeptical of claims like “Humans are the most efficient long-distance runners…” Members of other species don’t spend years training for distance running, so it’s not a fair comparison.
Humans aren’t the most efficient endurance runners. Horses are very good at long distances, so are wolves and some dogs and some antelope. Re. training: what do you think chasing/being chased around the landscape for much of your lifespan is?
But we are certainly better at running than the rest of the apes. Upright stance is one thing, but our ability to put water into cartons is also important.
On the other hand, don’t try to wrestle a chimp.
Don’t forget pronghorn sheep. I have some memory that they can run at 30 mph for an hour. They have a very high VO2(max).
Don’t forget weasles. They can whip their weight in wild cats (like my dad said he could).
Speaking of weasels, I saw one today! On the bank of the Milwaukee River. Way cool!
It seems to me that there are two problems with that argument: 1) Are you sure that other animals don’t spend as much or more time travelling long distances or engaging in various other physical activities that, if a human were doing them, we would call “training”? 2) Why don’t other animals spend time training for distance running? It seems like the only answer here that would make it an unfair comparison is to say that other animals could train for distance running, but choose not to. That’s a tenuous claim for a variety of reasons. Otherwise, if other animals don’t train for distance running because they are not capable of it… well, if we’re comparing the capabilities of various species, that’s just part of the comparison. At best, you can just push the comparison back a level–from comparing distance running ability between humans and other species to comparing training abilities between humans and other species–but I’m not sure that really accomplishes anything (AFAICT, humans still win at both levels of comparison).
The word “eidetic” springs to mind, or as they used to say “a photographic memory”. The chimp is certainly fast, but it would help greatly if the numbers were still there as the kind of almost-after-images we ordinary people have of memories only more so, or even literal after-images like those we see after prolonged exposure to bright images. I wonder how this could be tested?
That’s my assessment as well. I don’t think the chimp is remembering at all. It can literally still see the numbers, just as a human with eidetic memory can. The fact that it can sort numerically is impressive enough for me.
I recall reading somewhere that humans in hunter-gatherer societies tend to have a high incidence of eidetic memory as well. On the other hand, humans in a modern society who have eidetic memory usually have serious mental deficits in other areas. It may be that growing up in a modern society tends to either brush aside eidetic ability or suppress its development.
The individual, as it is displayed, never perform a sort or numerical comparison though. It could have memorized the point order as a list, for all we know from the video.
Interesting claim you make in the end. Is there any statistics on eidetic memories and “deficits”? AFAIK there are quite a few with eidetic memories out there.
Anecdotically, the majority I have been told and shown such memory function with (well, 2 out of 3) were high functioning individuals in all areas I could judge them. And they used it to advantage (learning languages respectively individuals as their studies respectively work had them to.)
The remainder was only deficient in some social interaction areas, and interestingly he used his ability only for party tricks that I know. (I.e. continuing discussions several months apart where he left them off.)
In fact, I believe that is how I sort (today, after years of practice).
When I note that “2 < 3" most often I don't visualize the sets and/or count, I take it from the list order for 0 – 9 because it is less effort.
Agreed. The chimp is pointing to where it still sees the numbers.
It light of this, I found it odd that the narrator seems so impressed with the chimp’s speed — as if retaining that image for longer would be somehow less impressive.
One way to test this hypothesis might be as follows: instead of replacing the digits with white squares, replace them with a new set of digits in the same spatial pattern but in different order. Then reward the chimp for touching them in the first displayed order rather than the second.
The chimp cannot still see the numbers. Part of the purpose of the high-contrast checker-pattern mask is to disrupt the formation of retinal afterimages. It also takes either many seconds of staring with a completely stable gaze or an extremely bright flash to form afterimages in the first place, neither of which is in evidence in the video. Even under these conditions, discerning fine detail in an afterimage typically requires a fluctuating background light level, which is why blinking repeatedly helps afterimages stand out. Finally, afterimages of numbers would only be legible near the fovea. I’m not sure what the angular size of the screen is, but it’s clear the positions of the numbers can extend tens of degrees into the periphery, where they would be subject to neural blurring at the retinal level and the effects of crowding at the perceptual level, rendering them illegible.
Put “see” in quotes then. I wasn’t referring to retinal afterimages but to images retained in visual short-term memory and then mentally overlaid on the screen — “eidetic” images, as Shuggy suggested.
I don’t see that using the term “eidetic” clarifies very much. Probably, we can all remember spatial configurations to some extent and it’s likely something that can be vastly improved by training when the brain is presented with lots of problems of the same type.
Similarly, chess masters can learn to remember the positions on many boards at once, the record for playing blindfold being close to 50 simultaneous games. Exactly how this works doesn’t become any more clear with the use of a term such as “eidetic”, that’s a description at best, not an explanation.
Maybe the chimp has palinopsia.
He can see Russia? How would that help?
+2
Certainly he can’t “count,”
Well, maybe he can. When I was an undergrad, I worked in Irene Pepperberg’s parrot lab at the U of Arizona. We had pretty convincing evidence that African Grey parrots can count. If you present them with a collection of objects, say 5 blue and 4 green and 4 red ones, and then ask how many green there are, they’ll answer “4”, and they’re able to do it with surprising accuracy. I haven’t kept up with the literature on the subject since then, but I’d imagine that chimps can do much the same thing.
(…not that that’s what’s going on in the video, of course, but I don’t think chimp counting should so quickly be dismissed.)
As I remember it, birds can “count” by judging an impressive set of objects correctly. Parrots do something like 7-11, humans can do 3-4 before needing to actually count which is about where fishes land [sic] I think.
Something is improved in birds compared to the vertebrate norm, especially the more behaviorally versatile, but I don’t think they know what it is.
Why is perhaps easier, don’t many birds stash seeds? It would be handy to know if they have recovered their local stashes. (Some small birds don’t do very well there, I hear.)
Maybe we should compare squirrels with birds on this ability.
Three is easy.
But, if you can do that, it is only a small step to “seeing” groups of three in a larger collection of objects, which should increase the total number you can “see” to nine.
(For the numbers that are not multiples of three, you can simply add or subtract one)
Try it.
I beat the chimp (on my 4th try)
The trick was guessing
As I recall, this task which chimps do so well is a skill our lineage “gave up” when we acquired other cognitive abilities. Sort of like how upright walking made us clumsier in the trees.
That seems plausible. And it might be nurture as well as nature. Both our evolution and our upbringing may have emphasised verbal skills at the expense of some more intuitional skills.
Sort of what Thanny described above.
Is this a generic idea out there? Does it have any source? It is IMHO apriori unlikely, since we are tasked to do these sorts of things daily. (I.e. remember fleeting visuals.)
Remember also that those numbers are just shapes to the chimp, while to us they are symbols, and symbols which to us have a meaning, and are in an order which we have known since childhood. What if the numbers were Chinese characters? Would non-speakers find it so easy then?
Just So…
I guess we’d have to know what brain regions are involved, whether and how they differ in humans and chimps, and whether regions have been recruited differently for other types of processing. Perhaps we could infer the ancestral state if we could test this in gorillas and orangutans.
There was an earlier post by Jerry where his new creationist pen pal in Israel implored him to explain why human abilities are so much more advanced than that of other primates. And I immediately thought of chimps who are demonstrably superior in short term memory. Any chimp can handily beat any university student in a test devised by Japanese researchers that measures the ability to remember a sequence (of numbers or objects). Chimps also seem to have a very robust immune system and can easily ward off many “bugs” that afflict us. I don’t think the Jerry’s new buddy had any of this in mind when posing his question.
Evolution serves no ego. It only confers an advantage in the proximate frame. Chimps are truly amazing for being able to remember each of their food trees and when they fruit. And because we have faced pathogen X in our history, certain mutations advantageous back then rendered us more vulnerable to pathogen Y today. Evolution is bereft of a foresight or sense of glory. Extend the comparison of humans to other beasts and there is one that smells much better than us, hears much keener, runs faster, and sees much further and across a broader visible bandwidth. Heck, there’s even one enjoys recreational sex more than us. On the other hand, there are startling similarities between the abilities of chimps and us. For one, chimps and a troublingly large number of humans both share a deficit in Abstract and Hypothetical thinking abilities. Like not being able to grasp evolution.
Claim in need of ref… Tell you what though. As we can see in the link provided by Roq Marish in the 2nd comment, in the test shown in the video humans can train to be as proficient.
The Japanese researchers had an erroneous experiment regime for making the claim they did. (Eg. “this performance difference was evidence of a memorial capacity in young chimpanzees that was superior to that seen in adult humans”.)
“Ayumu had many sessions of practice on their task before terminal performances were measured; their human subjects had none. The present report shows that when two humans are given practice in the Inoue and Matsuzawa (2007) memory task, their accuracy levels match those of Ayumu.”
As that test belongs to the group you describe, I doubt any of that is the case.
[Unless you have other references, that is.]
*** Insert gratuitous joke here about human inability to remember comments just above in a thread. =D ***
Claim in need of reference.
Conversely, there are pathogens that affect them but not us. This is not terribly surprising given that we have evolved in different habitats for the last few million years.
If in fact there are more human-specific pathogens than chimp-specific pathogens (and I don’t know that there are), I suspect that would have less to do with our “inferior” immune systems than with the fact that we live in dense concentrations of population where pathogens can thrive, rather than in dispersed tribal groups as our ancestors did (and chimps still do).
More on that chimp here: http://www.bbc.co.uk/nature/16832379
And from an article in Nature here (where it indicates that he was 7 in 2007) http://www.nature.com/news/2007/071203/full/news.2007.317.html
I heard his trainer on some podcast hypothesizing that it might have to do with language… we can’t help but say the numbers in our mind– sort of like we can’t help but read the colors on the stroop test– the chimpanzee doesn’t have the interference of a narrator in his head.
Right now I’m reading Temple Grandin’s book about living with autism, and her understandings of how autistic brains function differently than others. Temple herself is an entirely visual thinker, and she remembers full images in her head that she can refer back to whenever needed. We’re trying to remember 9 separate positions of numbers, but she would remember a visual snapshot. She hypothesizes that animal memory works that way too, which is why she’s so good with understanding them.
So I’d like to see an autistic visually oriented human go up against a chimp for this challenge.
I beat the chimp on my 5th try. Woohoo! I’m going out for a banana smoothie.
Re the Silberberg and Kearns study, I may be missing something, but they talk very breezily about being able to improve their skill in the “limited hold” test (the one performed by Ayumu in the BBC clip linked to by Articulett), yet if that’s the case they must have very remarkable abilities themselves. When the numbers are displayed for a quarter of a second before being masked, *my* success rate in reproducing the sequence of 9 digit positions would literally be zero, every time, without ever improving. Why? Because I simply don’t have the ability to look over the numbers and note the position of each one in the time available, let alone *remember* that sequence later. To succeed, one simply *must* have an eidetic or photographic memory, since the whole image of the screen must be taken in as a single operation — and then be available indefinitely for recall — *before* its individual components have been analysed or understood. I think people are underestimating how unusual an ability that is. In humans, it’s equivalent to the kind of feat Stephen Wiltshire does by repdroducing a complex cityscape view after briefly seeing it (for instance http://www.stephenwiltshire.co.uk/art_gallery.aspx?Id=5863).
Jerry doesn’t think Ayumu can count, but I don’t know on what basis he assumes this: Ayumu clearly understands that these 9 symbols represent a *sequence* — a sequence he is able to reproduce no matter what order the symbols are displayed in, or whether some are missing from the sequence. I may be wrong, but it seems to me that this means he understands that these are numbers representing quantities. In addition, the beginning of the BBC Liz Bonnin clip shows him reproducing the feat with numbers up to, I think, 20, which suggests that he may also understand the concept of positional notation. i.e. that putting a 1 in front of a 5 makes a total that is 10 larger. If he is able to manipulate numbers at such extraordinary speed, what other arithmetic, or perhaps mathematical, operations might he be able to understand?
This is unlikely. As you say, he’s been trained to recognize that the symbols have a certain sequential order, not that they represent a certain quantity. They could have used the roman alphabet or kanji (either ordered by the way its typically taught to school children) or even arbitrary shapes. Pure speculation here, but I’ll suggest that they chose arabic numerals because they are used world-wide and the proper sequence will be recognized by basically everyone, while the ordering of any particular alphabet would not be accessible to everyone they might want to reach.
Exactly. Imagine dumping a box of alphabet blocks out on the floor and asking a five-year-old to pick them up in ABC order. No arithmetic required for that.
Agreed — I guess what we would need to know is the circumstances in which he was taught the numbers and their sequence, i.e. was he taught simply that they were nine symbols that must always appear in a certain sequence, or was he taught that they each have an intrinsic meaning relating to quantity? Can anyone who knows more about Inoue and Matsuzawa’s research comment? It’s worth pointing out that he could not have learnt this sequence from the game itself — there must have been a separate instructional phase about the number symbols beforehand.
I got hooked on that game last summer so thought you might be amused by my entirely non-scientific personal anecdote.
The game can be very addictive, especially as a way of avoiding other work that needs doing… Playing it for a good amount of time does increase my success rate, so practice seems to be a factor. I find certain strategies useful: To avoid fretting when I don’t succeed seems crucial. The faster I play, the better it gets. Also I look at the image of the numbers with a broad view, like looking at a panorama, rather than trying to follow the numbers with the eyes. It is only after the numbers dissapear that the right sequence pops into my head. Sometimes I know the sequence without being sure what the numbers actually were, especially the last ones.
Does anyone know of a version of the game online with more numbers? The chimp gets to play with 1-9 after all. Just to feed the addiction… 🙂
The warning at the beginning of the video, does anyone know why this research is under dispute? Sorry if this has already been dealt with. Haven’t read comments yet…
I would think that this type of memory capability would be very helpful in an animal that is rapidly swinging tree branch to tree branch (or whatever.)