Bees Can Solve Math Problems With Addition and Subtraction

According to a new study published in the journal Science Advances, researchers from Australia and France have shown that bees can perform simple arithmetic, adding and subtracting small numbers by studying color-coded shapes. CNET reports: To test the buzzers' ability to perform arithmetic, the team used a three-chambered maze shaped like a Y, training bees to enter through a hole into a small chamber where they would see their first stimulus: blue or yellow shapes on a plain, grey background. The number of shapes varied between 1 and 5 and the color of the shapes told the bee whether it needed to add one (blue) or subtract one (yellow) from the initial number. The bee then flew into a subsequent chamber which presented both a correct option and an incorrect option. To train the bees, the correct option rewarded the critters with a drop of tasty sugar solution -- a delightful dessert for the bee. On the other hand, selecting the incorrect solution resulted in a nasty drop of quinine -- like a slab of Brussels sprouts slathered in chocolate.

The testing procedure itself focused on 14 bees undergoing four tests of 10 choices. The tests themselves were "non-reinforced," so they didn't receive reward or punishment when selecting their "answers" during testing. Because the bees were subjected to two answers each time, the expectation is that -- purely by chance -- they would select the correct answer 50 percent of the time. But the bees performed significantly better than chance would predict, selecting the correct answer around 65 percent of the time.

Comments

New way for students to cheat on math exams . . .

[PolygamousRanchKid]

. . . Bee Bearding:

https://en.wikipedia.org/wiki/...

And here all along I thought they were just trying to be cute or relive some of the stress of the exams with their "comfort" pets.

Did they try this with people?

[Anonymous Coward]

Without any instruction, would a 4-year old be able to figure out the color code or even the fact that a mathematical operation was required? We know 4-year olds can do simple arithmetic, so administering the test to them would help validate the method.

Re:Did they try this with people?

[omnichad]

Exactly. How do we know there's no pheromones at play after multiple runs through the puzzle? Or some ability to smell the sugar. Or any of a number of confounding factors.

Re:Did they try this with people?

[jellomizer]

Addition and subtraction is often rather instinctual. What makes it difficult, is when we symbolize the numbers in an abstract base, and use symbolic representations of addition and subtraction.
so
1 = *
2 = **
So if we say put * and ** together we get *** no problem. However saying 1 + 2 = 3 requires higher brain power. Because the symbolic numbers is more abstract and doesn't always mean the same thing.
so
1 = A^
2 = #$
So if we put A and #$ together we get possible combinations of A#$, #A$, #$A, A$#, $A#, $#A but we still have 3 characters. However a character is more abstract then just a * it is just the same thing, and we know instinctively what would happen if we add or subtract the same thing.

Re:Did they try this with people?

[BringsApples]

If I throw you an apple, and you reach out and catch it, we'll all be extremely unimpressed. And yet, if someone sitting there watching, pulled out paper and pensil and started writing out the formulas that we each just unconsciously used to do all of the physical activity that we did while tossing and catching an apple, it'd look quite complex.

For whatever reason, mathematics is universal, built-in to everything within the universe. Each species has it's own method of going about life. It's literally impossible for any species to go about life without acting in some mathematical way.

That we are able to detect that the bees are doing math, means that we've only caught onto a way that they do it, that we can recognize. Who's to say that they're not preforming mathematics at levels that we're unable to detect?

Re:Did they try this with people?

[Potor]

I am so stealing (and modifying) your example for my philosophy of science classes.

It is one thing to say that bees interact with the world mathematically, and another to say that they understand math.

I don't see how science can make the jump from the former to the latter in a way warranted by measurement.

Re:Did they try this with people?

[BringsApples]

I think that everyone has a grand understanding of the universe in some meaningful and useful way. I mean, even something as simple of "ownership" applies here. For instance, the bees that are all flying around your yard, on your flowers... they feel that they're in their yard, on their flowers. And according to Nature, they're correct, and so are you.

What one may understand as (what we call) mathematics, another may understand as color. Take the example of butterfly wings. The color from their wings comes from their structure, and not a pigment.

Re: Did they try this with people?

[Anonymous Coward]

Soon...

Biological bee computers

Re:Did they try this with people?

[Dragonslicer]

Except that as the summary says, it wasn't merely counting items. It also involved translating color into an arithmetic operation, so there was a symbolic representation of addition and subtraction. I can't speak to the rest of the experiment's design, but if the results are valid, it would seem that the bees do have some amount of understanding of the abstract concepts.

What a load of

[Anonymous Coward]

That's the exact definition of Pavlovian conditioning. You don't need the reward / punishment in the actual test, that's the exact point of it.

Also 14 is not nearly a large enough sample size to determine whether that slight deviation from a perfect 50:50 result is not by chance.

So this is not even a good enough experiment to determine whether pavlovian conditioning works on bees.

Re:What a load of

[Oswald McWeany]

That's the exact definition of Pavlovian conditioning. You don't need the reward / punishment in the actual test, that's the exact point of it.

Also 14 is not nearly a large enough sample size to determine whether that slight deviation from a perfect 50:50 result is not by chance.

So this is not even a good enough experiment to determine whether pavlovian conditioning works on bees.

Laughably small test sample numbers! Insects and spiders have shown that they DO possess more intelligence than we give them credit for. (I've seen studies suggesting cockroaches could have similar intelligence to rodents); but this test is just statistical noise. You don't get any confirmable answer with these numbers.

However... put that aside... imagine they did 14,000 tests and got the same result %; does that show that insects can do maths? Probably not, there could have been some other form of pattern matching going on that the researchers weren't aware of.

Re: What a load of

[Anonymous Coward]

These were my thoughts, too. There may be neurons firing, but this is a far, far cry from 'doing arithmetic'. And the sample size - give me a break. Was this 'study' done at a high school science fair?

Re:What a load of

[jellomizer]

I also expect the inverse too. We humans are not as smart as we think we are also.

A lot of our actions and responses is more instinctive then intellectual. Political/Religious/Sports Teams/Text Editors/Cell Phones leaning is less about an intellectual response to the policies,rules and teachings. But the instinctive need to be part of group, and be protective of such group.

An Evangelical and an Atheist (Both English Speaking Americans) will normally be diametrically opposed to each other. However we take these same people and put them somewhere where they are surrounded by Non English speaking foreigners who are neither Evangelical or Athiest, they will probably be rather close friends for a while, as they will classify themselves as English Speaking Americans first and their Religion secondly, and stick together for a while. This is instinctive not intellectual.

Or back in my elementary school days. K-4 were separated by 6 schools, 5th and 6th grade was combined into one school. For the first couple of weeks I befriended some kids, who I never got along with and didn't really like in K-4 for the sole reason that they were in my K-4 and I knew who they were. Only to make new friends/allegiances later.

There are a wealth of other behaviors that are instinctual which seem like intellectual activities, from diet, sexual activity, grooming needs....

No way this is significant

[Anonymous Coward]

14 bees undergoing four tests of 10 choices getting 65% right? Are you kidding me?

Re: No way this is significant

[Anonymous Coward]

Seriously. Numbers like don't even mean shit to kindergarteners.

math not needed

[phayes]

Memorization of the correct and incorrect answers is all that is needed for the described (too small of a sample size to be considered an) "experiment".

How far /. has fallen...

Re:math not needed

[umafuckit]

Memorization of the correct and incorrect answers is all that is needed for the described (too small of a sample size to be considered an) "experiment".

How far /. has fallen...

It is true that memorisation could explain this. You point about the sample size is trickier, though. Firstly there is no magic number that constitutes a large (vs small) sample size. What is suitable depends on the size of the effect, the variance, and the degree to which you want to generalise the results to a wider population. This is often balanced against what is possible. In biology a lot of experiments have a small sample size because of the cost or difficulty in gathering the data. For instance, I just reviewed a paper where the authors have gathered data from just a single subject. However, they gather a vast amount and do a very thorough job. Their work still stands as it is (it's in a sense a methods paper) and given that they aren't targetting a big name journal or over-selling their results I'm going to let the n=1 slide.

In the particular case of this paper, what I find most annoying isn't the n=14 but that their graphs hide the underlying data by displaying them as just bars with a 95% confidence interval for the mean. I would also agree, however, that I don't see why in this case they couldn't have produced a larger sample size. That's not the main issue, IMHO, however.

Re:math not needed

[Anonymous Coward]

Not even that (as I've said in another post). Bees have been around for literally millions of years; it would be amazing if they still weren't any good at picking up clues on which plants deliver the goods. Want to bet that a bee's visual cortex can't classify the difference between a "busy" pattern of lots of patches and a "simpler" pattern? All it has to learn is to associate busier patterns with the reward for one colour of "flower", and simpler ones for the other. Job done, no counting required.

Re:math not needed

[Anonymous Coward]

Look at the editor and tell me how surprised you are.

Re:math not needed

[SunTzuWarmaster]

This isn't like they are studying the remaining living WWII veterans or Japanese anorexics. They should be able to find some extra BEES to run the tests on. Presumably they have access to a hive, at a minimum.

Re:math not needed

[umafuckit]

This isn't like they are studying the remaining living WWII veterans or Japanese anorexics. They should be able to find some extra BEES to run the tests on. Presumably they have access to a hive, at a minimum.

I agree a larger n would be nice (say, n=30 at least) and I *think* it's likely not too hard to obtain in this case. I would caution, however, that sometimes it's a lot harder than it looks to obtain these data. It could be that n=14 is hard to do.

I used to work in insect neuroscience and I collaborated with people who did experiments of the general sort described in the paper. The issue was of course not finding insects -- we had lots of insects -- the problem we had was that running the experiments was very time consuming and could often fail for unclear reasons. You may get drift in behavioral scores over time, batches of insects that produce suspect results, etc. All sorts of really weird stuff happens with animal behavior and so to get solid results you believe in might require throwing out most of your data (e.g. because variance was weirdly high on some days). After all is said and done your sample size isn't always what you hope for. I've seen really good people work for years and still end up with sample size of less than 10 animals.

Re:math not needed

[Anonymous Coward]

You may get drift in behavioral scores over time, batches of insects that produce suspect results, etc.

Yes, that's exactly the problem here. These guys got a batch of bees that happened to score 65% and cherry picked that as proof that the bees can do arithmetic.

Re:math not needed

[Oswald McWeany]

Not even that (as I've said in another post). Bees have been around for literally millions of years; it would be amazing if they still weren't any good at picking up clues on which plants deliver the goods. Want to bet that a bee's visual cortex can't classify the difference between a "busy" pattern of lots of patches and a "simpler" pattern? All it has to learn is to associate busier patterns with the reward for one colour of "flower", and simpler ones for the other. Job done, no counting required.

Indeed, this is my first thought too. Not that I think it is out of the realm of possibility that bees can count; but it is probably much more likely there is some sort of pattern matching is going on, as you suggest. Bees need to be able to recognise visual patterns to find flowers; you'd expect them to be good at that. It's a much more feasible explanation than that they can count and do simple maths.

Re:math not needed

[umafuckit]

You may get drift in behavioral scores over time, batches of insects that produce suspect results, etc.

Yes, that's exactly the problem here. These guys got a batch of bees that happened to score 65% and cherry picked that as proof that the bees can do arithmetic.

If they did that, it's gross data manipulation. You don't know that they did that and there is no evidence that they did.

Re:math not needed

[Anonymous Coward]

As I read it, they trained the bees by displaying 1, 2, or 4 blue shapes and rewarded the selection of a feeder with 2, 3, or 5 respectively (the "addition" test). They also trained the bees by displaying 2, 4, or 5 yellow shapes and rewarded selection of 1, 3, or 4 respectively ("subtraction").

It seems to me the training was "see blue, select the feeder with more blue; see yellow, select feeder with less yellow".

 

Re:math not needed

[phayes]

I agree that the small sample size is no proof of any malign intent, however, the tiny sample size with no explanation of why the sample size is so small, the use of graphics instead of presenting the raw data and the fact that they never appear to have considered memory being sufficient to explain the results without any claims that "bees can add/subtract" are all worrying signs that the study is worthless.

Re:math not needed

[Anonymous Coward]

p>It seems to me the training was "see blue, select the feeder with more blue; see yellow, select feeder with less yellow".

Then you need to read that article again: "Two of the tests involved the incorrect answer being the same direction as the correct answer (addition: sample = 3, correct = 4, incorrect = 5; subtraction: sample = 3, correct = 2, incorrect = 1)."

Re:math not needed

[umafuckit]

I agree that the small sample size is no proof of any malign intent, however, the tiny sample size with no explanation of why the sample size is so small, the use of graphics instead of presenting the raw data and the fact that they never appear to have considered memory being sufficient to explain the results without any claims that "bees can add/subtract" are all worrying signs that the study is worthless.

Based on what I see in the field in general, I wouldn't say n=14 is "tiny". I would say it's "OK could be better" but of course that depends on SD and effect size too. As for the lack of raw data points, that's inexcusable (i.e. there is no reason not to do it) but unfortunately very common indeed in this field. I've found from experience that you can't write off a paper just because they miss out the raw data points: you'd throw out 90% of the field and it wouldn't make sense. FWIW, authors are obliged to send a reader raw data upon request. I've not really read the paper, but based on the figures it seems they lack controls of the sort you mention. That's the problem. The rest I can live with.

Re:math not needed

[umafuckit]

I agree that the small sample size is no proof of any malign intent, however, the tiny sample size with no explanation of why the sample size is so small, the use of graphics instead of presenting the raw data

On reflection, I take back the raw data statement (which was made WRT to lack of data points overlaid onto error bars -- a common thing in the field). There ought to be more detailed analyses of what individual bees do.

Re:math not needed

[Anonymous Coward]

I didn't see any claim that the bees were "doing math" -- only that they were solving simple math problems. Humans solve math problems. So do calculators, in a different way. Who knows what bees do? Whatever it is, they get the job done which is interesting and worth investigating further.

Re:math not needed

[epine]

Your point about the sample size is trickier, though.

No, the problem is not sample size.

The problem with this study design is that they never exposed the bees to a number/combination they had never seen before, to see whether the inductive gaps were filled in, or 404 in the memory table.

And there are other possible controls for memory effects, such as whether this numerically consistent pattern is easy to learn than purely random patterns.

The wall has been hit

[AndyKron]

With 65% they get a failing grade in math. Science has been reduced to tinkering

Alternately...

[Anonymous Coward]

...bees learned that bigger bunches of blue flowers are better, and smaller bunches of yellow ones. That doesn't need maths per se; just a rough visual classification.

Impressive

[Quakeulf]

That is more than what I am capable of after only 4 hours of sleep per night.

pattern matching (not addition/subtraction)

[Anonymous Coward]

Sorry dudes, but this is just pattern matching, not addition and subtraction.
Hint: Bee just thinks "if the color color is X and shape is Y, go to location Z for the yummy stuff."
But we already knew bees can recognize things and memorize complex routes to teach their friends. (That's basically all they do all day every day.)

IMO this study proves absolutely nothing.

In other news... this just in: water is wet. Film at 11.

So can a calculator

[eggstasy]

A Half adder consists of TWO logic gates, a full adder can be 5.
Bees have approximately one million neurons. Fuck's sake, we should be able to run far more complex computations on it.

Re:So can a calculator

[omnichad]

Imagine a Beewulf.... I'll just see myself out now.

Yum!

[Anonymous Coward]

"like a slab of Brussels sprouts slathered in chocolate"
I've never had that, but it actually sounds appetizing! (I do like Brussels sprouts in the first place, though)

Re:Yum!

[fonske]

Guinea fowl bathing in a raspberry vinegar, pink pepper berry cream sauce with melon balls, grapes and litchi. Brussels sprouts with chocolate is a perfectly good idea to go along with it. And croquettes. Damn, I'm dreaming up the ideal filling for the poultry already...

If you think this is wild...

[Anonymous Coward]

If you think this is wild wait until you find out about the bee's Waggle Dance. That's a real mind blower as far as intellect, hive mentality and communications.

Re:If you think this is wild...

[Anonymous Coward]

Not forgetting the spelling bee.

Re:If you think this is wild...

[GoTeam]

Just don't dance in a way that insults their fat queen. They hate that

Trump owns you

[Anonymous Coward]

He's in your head all day and all night. And there isn't a thing you can do about it.

Low scoring

[billybob2001]

I think I'm being generous if I score this study B+

Bees can either subtract or smell

[abies]

So, bees can smell minimal amount of quinine or sugar from dispenser, even when it is still closed. This would require very sensitive sense of smell, only possible in animal which has to go large distances to find food based on smell alone. Yep, basic hypothesis is that bees can add and subtract. Only after we prove they cannot, we can look for other explanations.

Re:Bees can either subtract or smell

[omnichad]

Yeah. Should have seen the results of putting the sugar and quinine in opposite spots on the last test just to control for this. If they follow the cues, thru did math. If they follow the smells, then this study is worthless.

Re:Bees can either subtract or smell

[Anonymous Coward]

The other issue is if they let the bees interact between the experiments. If that was the case, they basically enabled the bees to inform each other of the correct solution and hence uncovered the abilities of group intelligence instead those of an individual bee, by the ants in a colony!

Memory not needed

[Anonymous Coward]

Memorization of the correct and incorrect answers is all that is needed for the described (too small of a sample size to be considered an) "experiment".

From the study: "As the sample stimulus of three elements had never been presented during training, the bees had not previous received reinforcement on the number four for addition or two for subtraction trials. "

How far /. has fallen...

As far as its comments by people who don't RTFA

Imagine a Beeowulf Cluster of these!

[jddj]

Damn, I gotta get some bees!

Re:Imagine a Beeowulf Cluster of these!

[Anonymous Coward]

I'll be impressed if they can find spelling bees.

Re:Imagine a Beeowulf Cluster of these!

[Anonymous Coward]

No, let them live in peace.

Myself, I'm going all fucking Pavlov for another Chocolate Brussels, ding!

check your premise

[Anonymous Coward]

So, "purely by chance" is more like 65% than it is 50%.

Re: No math needed play politics

[Anonymous Coward]

"WTC7 collapsing at free-fall speed which would require the use of all gravitational potential (meaning there was no resistance below"

More idiocy. I suppose that you have not seen the photos of the substantial damage to the corner of WTC7 after the towers collapse? Have not bothered to look either? I suppose they are fake photos, right? No, you just want to spread BS. How do I know this? I've been a forensic structural engineer for 35 years. You haven't a clue.

THAT is why

[mark_reh]

we had to kill all of them. We can't have insects that are smarter than people buzzing around.

Use this in some way to fight Monsanto?

[Anonymous Coward]

Yes those Monsanto bastards use those Bees to cheat at agriculture and dominate the neighboring farmers so maybe this could be used to strategically tip the scales maybe? Come to think about it, Monsanto could do the same. Uh oh..... now bees could get caught in the middle of an industrial war. Crap....

You mean to tell me

[Anonymous Coward]

That if I throw shit at the wall at a known pattern, I can count it in groups later?

Then why do math?!?

Then get your bees to video chat you and me across the ocean, or perhaps have your bees fly back to the coast guard and act as your epirb.

It's not colony collapse disorder

[pierceelevated]

It's "So long, and thanks for all the pollen"

ridiculus

[Anonymous Coward]

more stupid "science" from the same dumb monkeys that brought you "global warming is undeniable" and "evolution is a fact".

the only REAL truth to be found is in the BIBLE

Don't care

[kaizendojo]

Still not using a bee calculator. Where would you carry it?

Re:Don't care

[Bite The Pillow]

In my Osh Kosh BeeGosh pocket of course, my dungarbees. Right next to my pistil.

Beezes Buzzes Lesses In Eclipses

[Anonymous Coward]

Buzzin' Bees

For this reason, God sends them a powerful delusion(operation of wandering)(planet) so that they will believe the lie.

Mystery Red of the Great American Eclipse
It has blood on it!
ABCNews: Eclipse makes pendulum wander

This isn't Pavlovian conditioning.

[denzacar]

Unless you're talking about the "scientists" who've breezed through all those years of school and training without ever picking up on the basics of statistics and probability.
On account of just repeating the correct answers they memorized earlier.

Basically, you could "train" a group of 14 coins to do the same task.
https://en.wikipedia.org/wiki/...

Although, to be fair, they did fuck up on the Pavlovian conditioning as well.
In half of the experiments, "correct answer" was always going straight from the point of entry, not changing the direction.
To make sure bees actually "made choices" - both addition and subtraction were tested for.
Addition choices were blue, while subtraction choices were yellow.

"Amazingly", bees were not only "getting it right" more often if the answer is just "fly straight ahead".
When subtracting, they were "getting it right" more often when the answer demanded flying away from the wrong answer.
Clearly, bees are doing math, right?

Except, experimenters rigged the game.
For addition, "correct answers could be 2, 3, and 5 and the incorrect answers could be 1, 2, 3, 4, and 5".
While for subtraction, "correct answers could be 1, 3, and 4 and the incorrect answers could be 1, 2, 3, 4, and 5".

I.e. When "adding" bees were trained to just go straight if the color is blue, and for 2 out of 5 "answers" that would be the correct answer.
When "subtracting", flying AWAY from the presented option would be the correct answer in 2 out 5 "answers".

Leaving 3 "answers" where the bee would have a 50:50 chance of getting it right.
I.e. 2 answers the bees were conditioned for + 1.5 answers where they'd get it right 50% of the time = getting 3.5 out of 5 answers correctly, or 70%.

I.e. By color conditioning and random choice alone, bees should be getting it correctly more often than most of their measurements show.
Though right smack in the middle of their error bars when accounted for the color conditioning training.
72.1 +/- 3.20% for addition and 67.9 +/- 3.66% for subtraction.

In each of the four tests, the bees performed at a level that was significantly different from chance.
In the addition (same direction) test, the bees chose the correct option of 4 in 72.1 +/- 3.20% (mean +/- SEM) of choices (z = 5.05, P < 0.001; Fig. 2B).
In the other addition (opposite direction) test, the bees chose the correct option of 4 in 66.4 +/- 2.69% of choices (z = 3.81, P < 0.001; Fig. 2B).
In the subtraction (same direction) test, the bees chose the correct option of 2 in 63.6 +/- 2.89% of choices (z = 3.17, P = 0.002; Fig. 2B).
In the other subtraction (opposite direction) test, the bees chose the correct option of 2 in 67.9 +/- 3.66% of choices (z = 4.13, P < 0.001; Fig. 2B).
There was no significant difference between the performance of the bees in any of the four tests (z = -0.887, P = 0.375), demonstrating that the bees performed equally well on all tests.

They've "discovered" that bees can tell colors and tastes.

Bee Gees, before coffee

[Tablizer]

I misread it as "Bee Gees", and a tune started playing in my head: "Ah, Ah, Ah, counting to five, counting to five..."

If they learn NAND

[Anonymous Coward]

they can build their own computers.

65% is hardly definitive

[Rick Schumann]

That's all I have to say on the subject -- other than how much 'studies' like this perhaps inflate or spin their findings in such a way that maybe they're just trying to attract more funding sources; basically, clickbait for investors.

and they can do basic algebra!

[sootman]

2b + 3b = 5b

Sample size too small

[juancn]

The sample size is way too small to draw the article's conclusion. It could still be by chance.

a slab of Brussels sprouts slathered in chocolate

[Anonymous Coward]

As long as it is Belgian chocolate, all is fine.

At first glance, I thought the title said,

[Anonymous Coward]

"Beer can solve math problems..."
Hmm...

Math bees?

[The Snazster]

I knew about spelling bees, but who would have guessed there are also math bees?