Fish Can Count to Four

Khemist writes "Fish can count, according to scientists, who have found that North American mosquito fish have the ability to count up to four. Previously it was known that fish could tell big shoals from small ones, but researchers have now found that they have a limited ability to count how many other fish are nearby. This means that they have similar counting abilities to those observed in apes, monkeys and dolphins and humans with very limited mathematical ability."

Makes sense

[clonan]

They are in School!!!

(Yes I know, I know! It is a stupid joke)

Clever Hans

[suso]

Have they ruled out the Clever Hans [wikipedia.org] effect? Doesn't look like it.

Re:

[aproposofwhat]

Well if there is a Clever Hans effect, it's even more astounding, because it would imply that these fish are able to pick up on the experimenters' body language!

Re:Clever Hans

[esocid]

Have you seen a mosquitofish? Those things are most likely not able to discern body language and cues from humans nearby. Even if they were, I doubt that the fish had any interaction with humans during this experimentation. In TFA the females were observed being harassed by male fish and given a choice of either a 3 fish or 4 fish school, and chose 4 over 3. Those characteristic decisions were on par with human infants 6-12 months old, which is pretty impressive for a type of minnow.

Re:

[suso]

Did you ask them? Humans can notice things on a large scale, like weather changes, a supernova, etc. Why couldn't fish see at our scale. Maybe its not even that, perhaps starting the experiment triggers something that we are not aware of.

I Can count to Four...

[Jeremiah Cornelius]

...therefore; I am a Fish!
Q.E.D.

Now, back to this fascinating "Dianetics" you were talking about....

Re:

[rs79]

"which is pretty impressive for a type of minnow."

They're not a "minnow" they're a killifish you insensitive clod.

I saw a documentary with a prawn...

[Joce640k]

I saw a documentary with a prawn who could recognize its owner when he came into the room - despite the distortions of the glass tank, etc.

And monkeys/apes can count more then 4. Watch this TED presentation - they've got bonobos who figured out how to write all by themselves.

http://www.ted.com/index.php/talks/view/id/76 [ted.com]

Re:

[immcintosh]

Given the setup of this experiment, I don't see how this reference is even remotely applicable. I would say that the "Clever Hans effect" may be ruled out by default.

Re:

[Poltras]

My name is Hans, you insensitive clod.

Re:Clever Hans

[spun]

I doubt researcher in the field of animal intelligence would know about this completely obscure effect. I mean, it's not like the clever Hans case was a seminal event in the field of comparative psychology which completely changed the way everyone did animal experiments for the past 100 years. Thanks for bringing up this little known incident.

Sorry, sorry, it's a pet peave of mine when some /. poster thinks that scientists don't consider the most obvious and well known aspects of their field. If you, someone not in their field, have considered it, chances are pretty damn good they have too.

Re:

[iminplaya]

Sorry, sorry, it's a pet peave of mine when some /. poster thinks that scientists don't consider the most obvious and well known aspects of their field. If you, someone not in their field, have considered it, chances are pretty damn good they have too.

Sometimes the most obvious is the most overlooked. If I had a thousand bucks for every time I stumbled into the shop and asked, *Is it plugged in?*, I would have at least 2000 dollars.

Re:

[spun]

Well, you have a point there. But you know to ask partly because of those 2000 times, and all the times you heard someone else go through the same thing. I'm guessing mind you, but I think the clever Hans incident is to animal psychology as 'is it plugged in?' is to the field of computer science: basic.

Re:

[iminplaya]

...those 2000 times...

:-) Actually I doubt it was more than twice...

Re:

[spun]

Oh, heh. Me can read good, for reals!

Re:

[GargamelSpaceman]

Even if they had ruled it out, I don't think the counting would be the same as what we consider counting.
If you are alone, it's obviously different than if there is a fish swimming by. If one fish is swimming by, then adding another fish is doubling the number of fish nearby, adding the next fish increases the number of fish nearby by fifty percent, and adding the fourth fish increases the number of nearby fish by 33 percent. Adding a fifth fish only increases the amount of nearby fish by twenty five perc

Re:

[hullabalucination]

Could you please speak up? I'm hard of herring.

Re:

[dgatwood]

My clownfish, you're gonna have a halibut time making people laugh with that old joke. That said, something does seem fishy about the whole situation.

Hmm. Something tells me you just got schooled.

maybe

[Anonymous Coward]

maybe they should help Taco learn how to count so I stop seeing that there are -1 replies?

Or not..

[Brian Gordon]

Neurons fire when a certain threshhold of other neurons are firing around them, does that mean they can "count"? It's an electrochemical reaction, not intelligence.

Re:

[JK_the_Slacker]

You're an electromechanical... err...

...pretty lights... Wait, where am I? Who are... 4!

Re:Or not..

[clonan]

By that deffinition please explain how "intelligence" is any different...it is bigger, more complex but still the same general idea.

BTW they never said that fish were "intelligent" only that they could descern 4 from 3.

That could...

[jd]

...be hard to do, without a working definition of intelligence. (This is the bane of all studies on intelligence in animals, computers, etc. Until someone can determine what it actually is that people are trying to look for, nobody can be certain whether or not they've found it. All they can be sure of is that they've found something, where the something has properties in common with another something that is believed associated with what they're really interested in.)

Re:

[Anonymous Coward]

In other news, Nemo, the Einstein of North American mosquito fish, has just invented the concept of negative numbers. He discovered it while computing his income taxes.

Re:

[esocid]

TFA didn't mention any brain monitoring. It simply observed whether a fish preferred to enter a school of 4 or 3 fish when being harassed. So it can discern the difference between those two, either because it can tell 3 from 4, or the school looked bigger. Either way, it can tell the difference between a small number of fish and a larger number of fish. Can your dog do that?

Re:

[Brian Gordon]

I'm not talking about brain monitoring, I'm just using it as an example of other "counting" in nature that doesn't have any sort of thought behind it. And the reason I say thought is because the article summary says counting like fish can think.

Re:

[Bloke down the pub]

Either way, it can tell the difference between a small number of fish and a larger number of fish. Can your dog do that?

I doubt it, my dog isn't interested in fish at all.

Is not... Is so?

[spacey]

Apparently it's been demonstrated that our basic ability to count to around 4 is controlled by specific neurons firing upon recognition of the specific quantity. That's why the OP spoke of fish having the same basic ability to count as apes, people with learning problems, etc. See http://www.newyorker.com/reporting/2008/03/03/080303fa_fact_holt [newyorker.com] on page 3 for this:

"Dehaene has been able to bring together the experimental and the theoretical sides of his quest, and, on at least one occasion, he has even theor

Re:

[2names]

I wonder if this subitizing (on a much larger scale of course) is what we see in savants when they are able to seemingly instantly count high numbers of objects, e.g. the spilled toothpicks in Rain Man.

Re:Or not..

[Ed Avis]

"The question of whether a computer can think is no more interesting than the question of whether a submarine can swim." -- Dijkstra

I think a similar principle applies to this experiment. They showed that goldfish can count to four. Whether this signifies 'intelligence' in some abstract sense is a different question, and not really relevant here.

Re:

[exp(pi*sqrt(163))]

> They showed that goldfish can count to four.

But they didn't even do this. They can compare two schools of fish, up to 4 fish, to see which is larger. When you say someone can "count to 4" you imply a whole bunch of other skills too. I'm not talking about the kind of metaphysical nonsense that Dijkstra is ruling out. I'm talking about practical, measurable behaviour. By my reckoning, fish are nowhere near able to count. You might as well claim that water can count to 6 because it always forms snowflak

Re:

[Ed Avis]

Heh, I read that studies had shown some plants can count up to two... the word is certainly used more loosely than it should be. Let's just say that they are able to distinguish different numbers of other fish up to four. Presumably though they don't have a significant preference between a school of four and one of five. They can tell the difference between 0, 1, 2, 3 and 'any number above 3'.

Re:

[exp(pi*sqrt(163))]

But it's not clear that they can count anything other than fish. Can they tell the difference between 2 and 3 seaweeds? Or 2 and 3 rocks? Or 2 and 3 predators? Or even 3 and 4 'things' (that might be friends, foes or rocks). The whole point of numbers is that they are abstracted from the particular thing being counted. For example, we know that from a young age, human babies notice when the number of objects in a small collection have changed for a wide variety of objects. That's not exactly counting, but a

Re:

[soren100]

Neurons fire when a certain threshhold of other neurons are firing around them, does that mean they can "count"? It's an electrochemical reaction, not intelligence.

All humans have going for them is electrochemical reactions in neurons as well (in terms of intelligence). If your statement was at all correct (which it isn't) you would be "disproving" intelligence in humans as well.

Research on fish intelligence is showing some interesting results -- here is an impressive video of fish swimming in unison in response to hand signals . Science is busily proving that fish are smarter [nootropics.com] than most people realize [telegraph.co.uk].

If you believe in the theory of evolution, which most Slash

Interesting fish video

[soren100]

Sorry for replying to my own post, but here's the fish video [youtube.com] that I was referring to in the previous post.

The trainer gets four goldfish to swim in some very interesting synchronized patterns that he appears to communicate to them through hand signals.

Re:

[Brian Gordon]

Oh come on, quit questioning "but what does intelligence really mean?" this is not even close to intelligence. Nor is the trainer communicating with his goldfish, they've just recognized through manipulation of food/pain what they're supposed to do when they notice a particular pattern. Don't even think of saying "but isn't that communication?".. that's so far from real symbolic communication that it's almost as aburd a comparison as neurons firing to counting.

5? Interesting? IDTS!

[FreakWent]

You are the first person to use intelligence on this page. You are arguing against a claim that was never made.

The claim of intelligence doesn't even appear in TFA, so you're assuming that someone somewhere is claiming something they are not.

Maybe you should fire a few of your own neurons...

Re:

[mstahl]

Did you even RTFA? The experiment demonstrates that multiple fish exhibit more precisely-defined social behaviour than we'd seen before. This is not an electrochemical response; it's a reaction of the fish to the world around them. That's kinda neat.

Now What?

[arizwebfoot]

Are we gonna be strapping up to four mini-torpedoes to fish now?

Re:

[apdyck]

Only once we learn that they can count backwards from four...that way they know how many mini-torpedoes they have left based on how many they've fired.

in other news

[Lt.Hawkins]

Scientists discovered that fish are excellent at mod 4 arithmetic.

Re:

[gardyloo]

They keep making fencepost errors, though.

Hmm...

[mckinnsb]

"Limited mathematical abilities?" I wonder if they can shift left one bit...

Re:

[nschubach]

I'm pretty sure they can only shift right half the time.

one fish, two fish

[stoolpigeon]

red fish, blue fish

Re:

[jd]

But only blue fish can speak whale.

Does that make it "1 in 4"?

[Loibisch]

Does that mean the child molestion rate among fish is even higher than among humans [slashdot.org].

if you're a fish politician

[thegnu]

you can probably only count to 3. which makes for an alarming 1 in 3.

Bad for cardassians...

[geminidomino]

Gul Madred: There are five lights.
Jean-Luc Salmon: I only count four. ... ...
*6 hours later*
Gul Madred clicks on a fifth light.
Madred: How many lights do you see?
Salmon: There are FOUR lights!
Madred: *facepalms*

Re:

[EMeta]

I'm sorry I haven't mod points today. That was awesome.

Re:

[Anonymous Coward]

Nice, except it should have been Jean-Luc Pilchard [wikipedia.org]. (And Gill Madred).

Re:

[geminidomino]

Nice. If I was more studied in oceanic biology, I could have inflicted far more pun-ishment.

I;m not sure

[techpawn]

Not sure if the can count but I think my betta fish is learning how to lift the gate that separates it from the other betta in the dual betta tank [animalworldnetwork.com]. A few minutes after I close the gate he's there at the bottom trying to lift it. I know fish are smarter than most give them credit for, thank god there not a reverse scuba suit...

Re:

[esocid]

Too late [technovelgy.com] my friend, better keep an eye on your betta. Or sleep with one eye open.

Re:

[techpawn]

Okay, when it comes with a machine gun and lotus notes then I'll worry.

Re:I;m not sure

[Rogue Haggis Landing]

The betta I had as a kid could recognize me. When I opened the lid to his little tank he would come to the surface on the right side, where I would drop food. He wouldn't come when my brother (who never fed him) did the same thing. We were both about the same height, same hair and skin coloring, etc., so that's pretty clever, at least for a fish. Hell, my own father still can't tell us apart. I suppose I should have fed him more often.

Anyway, the point is that fish aren't mindless automatons like everyone thinks. They have a (limited) ability to learn (simple) things.

Re:

[internetcommie]

I observed a similar thing with the angelfishes I had some years ago. If somebody walked in front of the tank they would swim towards the front/top of the tank. If it was me (who fed them), they would then swim to the end where the feeding funnel was, but if it was my ex husband (who did not feed them), they would lose interest and go back to what they were doing previously.

Re:I;m not sure

[matria]

One summer an orb spinner spider build her web above the porch stairs. I began feeding her when she and the bugs were small; I'd toss them into her web. Towards the end of summer, she was huge, no longer maintained her web, and she would come down from under the eves to the bottom of the matted mess that was left whenever I stepped on the stairs, waiting for me to toss her something, but not for anyone else.

Re:

[tomthegeek]

Did you feed the fish at the same time everyday? If so the fish might have been correlating human approaches + time of day = feeding. When your brother came up to the tank it was probably the wrong time of day.

Re:

[Ihlosi]

I know fish are smarter than most give them credit for, thank god there not a reverse scuba suit...

They use *GO!GO!* for the *dancing* with *silly cows*.

(Turn in your geek card if you don't get the reference.)

Just like rabbits.

[Stavr0]

Rabbits: One, Two, Three, Four, Hrair Fish: One, Two, Three, Four, School

Re:

[dontthink]

Wow, first Watership Down reference I've seen on slashdot. Accurate too - right on.

Re:

[Hognoxious]

I dunno, seems the Ron Paul supporters are always going on about wog dogs [sparknotes.com]

But the real question is. . . .

[smooth wombat]

are they smarter than a Fifth Grader?

Based on at least one person I've seen on that show [stereogum.com], they very well could be.

Not good enough

[davidwr]

Let me know when they can count to 5 [slashdot.org].

humans with very limited mathematical ability ...

[thrillseeker]

politicians?

Re:humans with very limited mathematical ability .

[AHumbleOpinion]

humans with very limited mathematical ability ... politicians?

No. Donations, polls, and votes all use mathematics quite heavily. Perhaps you are confused by how they handle "our" money, to get a better evaluation of their capabilities note how they handle "their" money. :-)

Re:

[ROMRIX]

to get a better evaluation of their capabilities note how they handle "their" money.

Ya, maybe that's why Hillary still thinks she is in the lead and refuses to make public her tax statements... although, her accountant could be a fish, I know her husband was a shark.

Perhaps Best Not to Publish This Research

[hardburn]

This will only encourage PETA to make loud press releases about how fish are "intelligent, sensitive creatures" and how the Inuit diet is a source of great evil in the world.

Re:

[Khemisty]

I was thinking of adding a PETA reference to the story when I submitted it, but considering they already have a Fish Amnesty Day [peta.org] I'd hate to be forced into a Pledge [peta.org] if they ever found out who I was!

Re:

[corbettw]

Of course fish are intelligent, that's what makes them delicious!

Mmmm, brain food....

Re:

[K. S. Kyosuke]

Oh, come on. That would be such a fishy press release, everybody would immediately understand that all the fish is just a red herring.

Dr. Seuss knew!!!

[jollyreaper]

One fish, two fish, red fish, blue screen...oh, fuck!

so off the mark

[hesaigo999ca]

It's sad how they test to see if fish can count, or even assume that they are in fact counting, when they could be looking at the overall size ratio of the school itself and not the fish collective individually. This is one of those studies we pay for with our tax dollars, that I would highly decree to be a waste of money.

If I look over at a group of 2 diff. gangs and by instinct I will go for the bigger gang, then chances are that I will go for the visibly larger gang. Now to test if they were actually cou

Re:

[Skreems]

Well, they did test whether they'd prefer groups of 12 or 8, which is a larger ratio than 3 vs 4, and found no response in that case. So they did check for the size thing.

Re:

[blueg3]

The fact that you didn't read the article is showing. :-)

I'm going to assume from that that you also didn't investigate the source of their funding.

Re:

[Anonymous Coward]

I think you're misunderstanding the article. They ARE testing instinctive behaviour. Nobody is claiming that these fish are deliberately making any conscious decisions, or that they understand why they're choosing one group over another. The term "counting" might be misleading you, as it does imply intelligent thought, but in this case they simply mean that they can tell the difference between 3 and 4, in certain situations. They probably shouldn't use the word "count", but it's easier to explain to the

Re:

[corbettw]

From the second line of the article:

Previously it was known that fish could tell big shoals from small ones, but researchers have now found that they have a limited ability to count how many other fish are nearby.

So, yeah, they've already tested for size differences, this was set up to test something else entirely.

Fourth method

[Babu 'God' Hoover]

Adult humans use a third counting mechanism, in which they verbally count much larger numbers.

and a fourth method in which they assign the task to a machine

Not counting

[N7DR]

Although TFM insists (like the summary) on using the emotive word "count", there appears to no evidence of this ability cited. TFM does describe a situation in which the fish appear to be able to distinguish among small numbers. But that is a far cry from any concept of counting.

(Which is not to say that fish can't count; I'm merely pointing out that I see no evidence for it in TFM.)

Big Deal

[trongey]

I can count way higher than 4, and nobody's writing research papers about me.
Stupid fish.

We finally have proof!

[Gregour]

My goldfish is smarter than my president.

Re:

[MrMacman2u]

My goldfish is smarter than my president.

Dang funny comment, but, how did you not know that? My pet rock is a MENSA candidate compared to our current president. Though the horror will become less all encompassing soon... I hope... Pet Rock For President!

Very Limited Mathematical Ability

[ill stew dottied ewe]

"This means that they have similar counting abilities to those observed in apes, monkeys and dolphins and humans with very limited mathematical ability." -from the _obvious_ department

What can't count, then?

[wattrlz]

The potted plants on my desk grow towards the side with two visible lamps instead of the side with one visible lamp. Does that mean they can count? Bacteria will migrate towards the side of the petri dish with a lower concentration of toxins, does that mean they can count? Honey bees often fly towards the garden with more flowers.. etc. ad absurdum. So, what makes the mosquitofish special?

Re:

[armareum]

You've purposely chosen experiments/tests with confounding factors.

You fail at experimental design.

Subtle flaw, FEMALE fish can count to four

[SmallFurryCreature]

They tested the behaviour of FEMALE fish, not male ones. We know that in humans women are not as good at the hard sciences as men are. Offcourse counting to four barely classifies as hard science, except maybe in Utah, but still.

Still the article makes me wonder, how do we count. I am currently looking at the number of open tasks on my taskbar, there are 5, I counted four instantly. I am now trying to convince myself that this is because the fifth "icon" is where in previous KDE versions part of the start menu was so that I overlooked it and not that I have the brains of a fish. A female fish. Some hard liqour maybe inorder, although sadly I then also frequently have a problem counting past the fourth drink.

counting? or recognizing?

[largejunglecat]

Are the fish really "counting" to four? I think this is misleading. Just because they can recognize that 4 things is more than 3 things does not mean they are counting them; counting would imply a concerted, intelligent effort on the part of the fish.

Oh You Mean The Animal

[pokerdad]

As an avid poker player I was going to come in here and bring up the enormous amount of evidence I have seen that fish [pokertips.org] almost certainly cannot count to four, then I read a little further and realised its about the animal. Never mind.

Fish approved for...

[lpangelrob]

Well, I suppose this means they're more than eligible to receive their own Holy Hand Grenades then. Science is always causing trouble...

Ants

[FunkyELF]

Reminds me of a story I read on here about ants being able to count.
Some guy put stilts on ant legs when the ants got to some food source. He then hid their 'home' and on the way back the ants overshot where their home was.
They also did it the other way around, put stilts on the ants on their way to food and took them off on the way back. The ants fell short.

Anyway, apparently ants can count higher than 4

So I guess this puts them...

[Criceratops]

.. a couple of steps smarter than Gully Dwarves? Ack.

H*R

[magus_melchior]

Ah, then that means Homestar is dumber than a fish [homestarrunner.com].

"Then what's two plus two?"
"Well, the force between any two charges is equal to the absolute value of the multiple of the charges divided by 4 pi times the vacuum permittivity times the distance squared between the two charges."

The fish are counting in binary

[wildzeke]

The fish are counting on there front two fins, so of course you get 00, 01, 10 and 11.

quantity != size

[Walenzack]

So, the female fish couldn't see the difference between a group of 4 and a group of 5, BUT was able to see it when the size difference between the groups was bigger: say, 4 to 6.
What I don't understand, is why this study concludes that fish "can count up to four" rather than "fish can tell apart objects with a size ratio greater than 3/4".

It's like saying I actually count the number of sand particles in a beach when I decide one beach is greater than another...

Am I missing something?

Shoals??

[Cur8or]

Is this groundbreaking research suggesting that all fish might be of some Hebrew descent?