Science desk
< October 31	<< Oct \| November \| Dec >>	November 2 >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

November 1

Is every living thing which has a nervous system can think?

Is every living thing which has a nervous system can think? for example ants, cockroaches can think or everything that they do is considered as instinct? 93.126.88.30 (talk) 00:25, 1 November 2016 (UTC)[reply]

It seems pretty clear that not everything with neurons can think. Jellyfish have a simple nerve net that I don't think anyone imagines is capable of thought. Yet mammals besides humans have such structurally similar brains it's hard to imagine that they don't think in a manner similar to how we do. Where is the line drawn? We have an article on Animal cognition that may be of interest to you. Someguy1221 (talk) 00:54, 1 November 2016 (UTC)[reply]

(edit conflict) Well, it depends on what you mean by "think". Scientists, philosophers, and all sorts of people who have dedicated their entire lives to the question have not come up with a widely agreed-upon definition. Concepts like consciousness, sentience, sapience, awareness, metacognition, will, etc. are all generally part of "thinking", which is usually thought of as distinct from reflex, instinct, or other sorts of automatic responses. Read any or all of those articles to get a more nuanced view of the issue. You could also review the Hard problem of consciousness, which is also a good overview of the issues. --Jayron 32 00:57, 1 November 2016 (UTC)[reply]

(edit conflict)Quite recently there was research in which bees could learn how to remove a cover from sugar water, and then teach other bees to do the same. So it look as as if insects are not just subject to instinct, but can learn / think, and certainly have memory. But cnidarians also have simpler nervous systems without a brain, but have a network of nerves. Still I would not be surprised if they too could learn. Graeme Bartlett (talk) 01:02, 1 November 2016 (UTC)[reply]

As previous responders have already explained, the notion of "thinking" is not well defined, especially for non-hominids. Learning, however, has been demonstrated in animals with rather simple nervous systems, such as C. elegans or Aplysia californica. If learning is taken as a hallmark of thinking, then even a single-neuron model system can be made capable of learning. You can easily demonstrate this for yourself, by introducing a mechanism that modulates the probability of response of the model neuron to stimulus S_A (e.g. concentration of some substance A) according to the history of presence of another stimulus S_B (e.g. concentration of another substance B). Imagine substance B is a nutrient, while substance A has no biological role but is correlated in its concentration with B. You can make your model neuron learn to respond to the change in concentration of A, by having it respond to B with which A is correlated. This is a single-neuron version of classical conditioning. A learning rule may be: increase the probability of X receptor triggering a response when X is present AND B is present. As a result the neuron will learn to respond to X = A, and will start to respond to A even when B is absent. This is a single-neuron version of Hebbian learning rule. If you add another rule: decrease the probability of X receptor triggering a response when X is present AND B is absent, you can make it unlearn (forget) erroneous associations, as well. Dr Dima (talk) 01:38, 1 November 2016 (UTC)[reply]

Very good, thanks. Taxis is another concept that applies to relatively simple animals, which may in some circumstances be related to "thinking". SemanticMantis (talk) 15:03, 1 November 2016 (UTC)[reply]

Even learning is not well defined. There are different phenomena which can be described as learning, and they are not interchangable. There's learning which merely produces a new reflex, and there's learning which produces the ability to understand an abstraction. Teaching a nematode to swim towards a certain color of light to find food is not the same thing as teaching a nematode wave-particle duality of light. Sensitization is a very different thing than conceptualization, and yet we call them both "learning". --Jayron 32 16:54, 1 November 2016 (UTC)[reply]

Putting aside the term "think" again and focussing on learning, Habituation is actually the process of learning not to respond to irrelevant stimuli. Try poking your deeply sleeping dog or cat. The first time they might lift their head. The second time they might flick their ears and open their eyes. The third time - no visible response. They have habituated. Many, many simple animals such as sea anemones and hydra will habituate. This includes jellyfish (Johnson, M. C., & Wuensch, K. L. (1994). An investigation of habituation in the jellyfish Aurelia aurita. Behavioral and neural biology, 61(1), 54-59.). DrChrissy ^(talk) 17:09, 1 November 2016 (UTC)[reply]

Yes, but that sort of training is still NOT the same think as learning to appreciate the nuance in a well-played piece of music or understanding the intricacies of integral calculus. Abstraction is a key element of thought, and represents a far different process than merely training or un-training a particular reflex. Plants display tropism but we don't call that learning or thinking... --Jayron 32 18:19, 1 November 2016 (UTC)[reply]

First of all, habituation is not training. It is a natural response to repeated, irrelevant stimuli. Secondly, I never claimed jellyfish could solve quadratic equations - I only pointed out that animals with relatively simple nervous systems are able to learn. DrChrissy ^(talk) 20:19, 1 November 2016 (UTC)[reply]

What is training if not "changes in behavior due to repeated stimuli..." --Jayron 32 12:20, 2 November 2016 (UTC)[reply]

"Training" implies to me an intention on the part of the trainer. Animals will habituate to totally natural stimuli such as shadows from overhead sources. I would not call this "training". Furthermore, "training" implies a long-term change in behaviour. One characteristic of habituation is that if the stimuli cease for a period of time, the original behaviour will return relatively quickly. DrChrissy ^(talk) 16:30, 2 November 2016 (UTC)[reply]

Fair enough. If you define the word your way, you are correct. I was thinking of training as "stimulus which elicts change in behavior without abstract thought or conscious effort". As in, when I train to shoot basketballs, I become better at shooting basketballs even if I don't know or understand why they go in the hoop, conversely I can learn why basketballs go into the hoop (by understanding physics and the laws of motion) but knowing that has no effect on my ability to score a basket. Changing behavior through repetitive action is what I think of as training, understanding an abstract concept is what I was thinking of as learning. They are not really that closely connected. --Jayron 32 17:36, 2 November 2016 (UTC)[reply]

I cannot recollect ever having seen habituation defined in terms of training. Do you have any supportive references for this? One other characteristic of habituation is that it usually applies to innate behaviours, not learned behaviours.

Your suggestion that "training" applies to "stimulus which elicts change in behavior without abstract thought or conscious effort" is potentially problematic. What about jellyfish which move up and down a water column due to temperature and/or light? These are probably taxis type behavioural responses and I suspect do not involve abstract thought or conscious effort. What about typical dog training? To train a dog to sit, we might push its bum to the ground while saying "sit". If the dog stays in this position for a period of time, he gets a food treat. We continue doing this until the dog sits immediately at the verbal command without requiring the push. We have trained the dog to sit, but this has used the dog's ability to make an association between the command, the behaviour and receiving the food. I suggest this is training using conscious thoughts by the dog. DrChrissy ^(talk) 18:06, 2 November 2016 (UTC)[reply]

Yes, you are entirely correct. --Jayron 32 18:22, 2 November 2016 (UTC)[reply]

I think it should be kept in mind that for the wild animals the major preoccupation is to get it through the day, not die from hunger or being killed and eaten. They don't have any incentive for abstract thinking but I do believe that even simple animals with a handful of neurons do think if this improves their chances of survival. I've read (don't ask for a reference, it is all in the memory) that some spiders can walk a complicated maze and count to 80 in the process and they seem to have just a few neurons. Not every person in the United States can count to 80 or do arithmetic in this range. --AboutFace 22 (talk) 19:49, 2 November 2016 (UTC)[reply]

Not sure about spiders counting to 80 but some jumping spiders take an indirect route to prey rather than the most direct route, thereby indicating flexibility in behaviour and route planning, and possibly insight learning. (Sherwin, C.M., (2001). Can invertebrates suffer? Or, how robust is argument-by-analogy? Animal Welfare, 10 (supplement): S103-S118) DrChrissy ^(talk) 20:10, 2 November 2016 (UTC)[reply]

Just as a follow up and perhaps more direct answer to the OP, I started thinking about teaching in animals. Teaching in behaviour usually implies that the teacher is sensitive to the student's behaviour, and adjusts their teaching behaviour accordingly. In my opinion, this shows a level of consciousness and thinking. So, do animals with simple nervous systems teach each other? Here is an article which presents 2 very possible examples of teaching in insects (Leadbeater, E., & Chittka, L. (2007). Social learning in insects—from miniature brains to consensus building. Current Biology, 17(16), R703-R713.). DrChrissy ^(talk) 21:13, 2 November 2016 (UTC)[reply]

Complete paper just quoted is available on the web.[1] --AboutFace 22 (talk) 14:06, 3 November 2016 (UTC)[reply]

Gardner Island

Is there any ocean current passing by Gardner Island, especially any current flowing from the vicinity of Howland Island toward Gardner? 2601:646:8E01:7E0B:CDAE:DD86:5B6B:23C9 (talk) 06:29, 1 November 2016 (UTC)[reply]

The South Equatorial Current flows through that area, but perpendicular to the path between those two islands. Someguy1221 (talk) 09:28, 1 November 2016 (UTC)[reply]

So if Amelia crashed into the ocean near Howland and drowned, her body probably wouldn't wash up on Gardner? 2601:646:8E01:7E0B:CDAE:DD86:5B6B:23C9 (talk) 11:55, 1 November 2016 (UTC)[reply]

There's a pretty good consensus by people who aren't all that interested in making it more controversial in order to get you to watch TV shows on the subject that she crashed on Gardner Island itself. Skeletal remains were found there consistent with her gender and age, and there were likely pieces of her aircraft on the island. The Wikipedia article Amelia Earhart goes into some of this detail. We've been pretty sure of this for years, and doing some research, it seems that just in the past few days, articles have been published that further confirm this: [2], [3]. It seems that it is likely she landed semi-safely, but starved to death or died of exposure later. --Jayron 32 12:02, 1 November 2016 (UTC)[reply]

Why don't they do a DNA analysis ? I would expect either we could find DNA known to be from her (did she ever give a fan a lock of her hair ?), or perhaps from a relative, to compare with the skeleton. StuRat (talk) 21:42, 1 November 2016 (UTC)[reply]

They did a DNA test, but according to the article linked above it was inconclusive. Llaanngg (talk) 23:15, 1 November 2016 (UTC)[reply]

Importantly, DNA degrades if not properly preserved. Left to the elements on a desert island for decades may mean there's little usable genetic code if any. Under IDEAL conditions, DNA does have a half-life of about 521 years, but this is preserved in an anaerobic, light-free environment. A body lying exposed to the elements on a tropical island is hardly ideal, this article notes that DNA left in the open may degrade to an unusable state in a matter of weeks. --Jayron 32 01:32, 3 November 2016 (UTC)[reply]

The "left out in the open" part isn't likely to happen to all the DNA, though, as DNA in bone marrow, and in particular the pulp from molars, is quite isolated from the environment. StuRat (talk) 03:55, 3 November 2016 (UTC)[reply]

That may or may not be true in every case of every corpse that is left to the elements for decades. What I will note is that we know it wasn't true in the Amelia Earhart case because we were actually told that they couldn't find any usable DNA from the female skeleton they found.

Inappropriate Ref Desk behavior.
The following discussion has been closed. Please do not modify it.
But you keep being smarter than the people who are trained and actual experts in this. I see it is working well for you.

--Jayron 32 11:48, 3 November 2016 (UTC)[reply]

A bit old but [4] may be of interest

Out of all processed samples, 15 remains were fully profiled at all 15 STR loci. The other 12 profiles were partial. The least successful profile included 13 loci. Also, 69 referent samples (buccal swabs) from potential living relatives were collected and profiled. Comparison of victims' profile against referent samples database resulted in 4 strong matches. In addition, 5 other profiles were matched to certain referent samples with lower probability.

They found remains of 27 individuals in WW2 mass graves. I expect, but aren't sure, that DNA recovery will probably be higher when someone was buried eve if in a mass grave. Either way though, it seems they only got full profiles of slightly more than half of the remains. This was in 2007, so techniques may be better now. Still I strongly suspect anyone who thinks the discovery of a 70+ year old skeleton is a slam dunk for DNA is no forensic scientist.

Nil Einne (talk) 13:45, 3 November 2016 (UTC)[reply]

Those results are actually pretty good. But far more relevant is what remains there are that are left to study. The 13 bones discovered of the skeleton were lost back in 1940. More recently in 2010, investigators found three very small bone fragments that were tested against a close relative. That the skeleton was lost and that additional bone fragments were found is mentioned in our article on Amelia Earhart and its references go into other details. There was also a program I watched years ago that presented the story behind these new fragments, but I don't think there is any real need here for me to go digging that up too. -Modocc (talk) 16:44, 3 November 2016 (UTC)[reply]

what kind of weather can I expect under the jet stream boundary?

I know we pay attention to fronts a lot but the jet stream is rarely indicated on "consumer" weather maps. Why? This is annoying because at times I really need to predict things for myself and I have started traveling and need to pack for multiple cities. What kind of weather can I expect UNDER the jet stream boundary other than warm/cold, in terms of say, air pressure? Yanping Nora Soong (talk) 19:29, 1 November 2016 (UTC)[reply]

Jet streams exist at high altitudes. They have no direct influence on the weather below. Ruslik_Zero 20:19, 1 November 2016 (UTC)[reply]

Yes, but I know about convection, and how low pressure at the surface leads to high pressure heights aloft etc. That is surely related to wind speed? Yanping Nora Soong (talk) 20:56, 1 November 2016 (UTC)[reply]

Weather fronts and even hurricanes can be steered by the jet stream, but they don't follow it precisely. Weather closer to the poles from the jet stream tends to be colder and closer to the equator tends to be hotter, so directly under it you would expect moderate temps, on average, with perhaps a slight increased chance of precipitation. But, you would do far better to look at a weather forecast (which combines jet stream locations with much more data) than to try to predict the weather based on the jet stream location alone. StuRat (talk) 21:34, 1 November 2016 (UTC)[reply]

I would agree with StuRat that you are better off with forecasts. The neat little band of the jet stream as shown on TV forecasts is also a little bit misleading. Actually there are many jet streams that come and go all the time. Here is an up-to-date North Atlantic jet steam chart. 250 mb jet stream Notice it looks bitty compared to what is aired on the simplified TV weather charts. Also, the jets streams are not high pressure, rather they are of high velocity and can thus drag the surface weather with them including low pressure systems. So they do have quite an effect for us here on the ground.--Aspro (talk) 21:59, 1 November 2016 (UTC)[reply]

You may be looking for the Area Forecast Discussion. In the United States, our National Weather Service publishes many weather products, but the AFD is the one in which one or more professional meteorologists describe how the weather forecast was influenced by meteorological knowledge, including high altitude winds, atmospheric models, and so on. If the Center Weather Unit experts believe that the jet stream affected the weather, the AFD is where they'll describe how and why they believe so.

On any given day, I might check the AFD five or six times - it updates frequently. For my region, (MTR), there is commonly discussion of the influence of upper level atmosphere on surface weather. Meteorologists sometimes call out atmospheric levels by pressure altitude. Sometimes there is a discussion about jet stream(s). Sometimes, the meteorologists describe a computer model prediction - like the HRRR - and describe how it aligns, guides, or deviates from human opinion. The computer models account for upper level atmospheric conditions, but the implications are not always obvious.

If you actually explicitly care about the jet stream itself, you're looking for the winds and temperature products, or the high level SIGWX products. If you want the weather under the jet stream, you're looking for the Mid Level SIGWX products... and if you want the weather under that, you want the Low Level SIGWX products ... always remember that the atmosphere is three-dimensional, and the weather "under" the flight levels is often different from the weather at the surface.

Nimur (talk) 14:27, 2 November 2016 (UTC)[reply]

Try to find a synoptic chart. That's the map which actually shows the weather systems. When I was a kid, these were shown on the nightly TV news forecasts. Today we just get idiot symbols of what's happening right now, rather than a good prediction of the future. Weather forecasting predicts systems very well, the precise position and actual results of those systems less well. Andy Dingley (talk) 15:21, 2 November 2016 (UTC)[reply]

Another resource you might like for the USA is the Forecast discussion provided by the National Weather Service for each region. What you're reading is the top-level comments from top-level meteorologists, describing what is happening and why. It will help you learn to understand the general concepts of what pushes weather, as well as what to expect in a general sense over the next hours to weeks in a given locale. SemanticMantis (talk) 23:04, 2 November 2016 (UTC)[reply]

"The jetstream plays an important role in the UK's weather" [5] Here we are sometimes shown a jetstream chart on the TV weather, but only if it's not doing what it usually does. "Scientists have discovered the cause of the recent run of miserable wet summers as they begin to unravel the mysteries of the Atlantic jet stream." [6]. Alansplodge (talk) 19:42, 3 November 2016 (UTC)[reply]

Oh, another delusion shattered ! I thought that that summer here in the UK was defined as that period of the year when is was miserable and wet. And how recent is recent? In the 1950's Flanders and Swann even noted it: A song of the weather. Jetstream or no jetstream, If it wasnae for our wellies (which are a far superior form of galoshes), (and popularized by an Irishman who's name escapes me) where would we be?--Aspro (talk) 21:21, 3 November 2016 (UTC) [reply]

Is any living being with a brain so simple that it can be predicted?

Can we analyze a living being with neurons so well that we will know what it will do next? Llaanngg (talk) 21:16, 1 November 2016 (UTC)[reply]

You might be interested in this discussion: [7], where they are talking about simulating the brain of a worm with 302 neurons. However, note that there could be "randomizers" that make even such a simple brain unpredictable. (Why would it have them ? Well, being less predictable might make it fare better against predators.) StuRat (talk) 21:26, 1 November 2016 (UTC)[reply]

The OP might like to re-state their question. Yes, we can predict what an animal will do next. Our physicians frequently use the Patellar reflex test. They hit our patellar ligament with a reflex hammer while knowing that the knee will jerk as a response (if the human has not experienced injuries). However, I get the feeling the OP is not asking this sort of question. DrChrissy ^(talk) 21:40, 1 November 2016 (UTC)[reply]

Nope, I am not after this sort of scenario. Does the brain get involved at all, BTW? Would the knee of a brain dead person still react? I was thinking more about predicting whether an ant will turn left or right in a maze (without chemical traces on it). Llaanngg (talk) 23:12, 1 November 2016 (UTC)[reply]

In the case of the petellar reflex, the brain is not involved. The nervous impulses from the stretch receptors are transmitted to the spinal chord which transmits impulses for the knee to flex. Not sure about recently dead people! The ant-in-a-maze example may be more complex than you imagine. Some animals learn mazes in a process called Chaining. Fist turn left, then turn right, then turn right, then turn left....and so on. Chemicals might not be used at all. Others might use visual cues internal or external to the maze. This includes ants (Sakiyama, T., & Gunji, Y. P. (2013). Garden ant homing behavior in a maze task based on local visual cues. Insectes sociaux, 60(2), 155-162.) DrChrissy ^(talk) 23:40, 1 November 2016 (UTC)[reply]

This is a case where random decisions would be helpful. The ant colony wouldn't want searchers to always turn the same direction, and thus potentially miss a food source. A scent trail could mark directions already tried, but after it rains, that trail may be gone. StuRat (talk) 23:45, 1 November 2016 (UTC)[reply]

The one example I presented was ants returning to their home. I'm not entirely sure what you mean by "random decisions would be helpful". Animals use different methods to spatially learn depending on the context. Even Spontaneous alternation (I turned left last time so this time I will turn right) is not random but has evolved for information gathering. DrChrissy ^(talk) 23:56, 1 November 2016 (UTC)[reply]

This is a classic case of Buridan's ass, where 2 exactly equal paths should pose a problem which can't be solved, but in reality, it doesn't, because all animals are capable of making random decisions. (Of course, there's also the philosophical question of whether anything is truly random, or if things which appear to be random are simply following patterns we can't decipher. Quantum mechanics predicts that truly random events do occur.) StuRat (talk) 00:02, 2 November 2016 (UTC)[reply]

E/C I'm not exactly sure where you are going with this because the OP's secondary question was can we predict whether an ant would turn left or right in a maze. If the ant has learnt the maze, I think we can predict left or right with a very high probability of being correct. If the ant is being placed into the maze for the first time, its first decision may be random but other factors could be considered. The same could be said for humans learning a maze for the first time. DrChrissy ^(talk) 00:19, 2 November 2016 (UTC)[reply]

In the case of a learned maze, the question was whether we would predict behavior based on analysis of the brain structure alone. In this case, probably not yet, but we should eventually be able to do that. As for the behavior going through a new maze, that I'm not sure of, as a random factor may make prediction completely impossible, even if we have perfect information. StuRat (talk) 03:30, 2 November 2016 (UTC)[reply]

Well, understanding a complex system depends on a lot of prior knowledge. For example, to calculate threshold potentials, you would need to know exactly how many receptors and ion channels are already present, exactly what concentrations of ions are already present, how much neurotransmitter is present in vesicules, etc. And then many processes (exactly when ion channels are open, exactly how much neurotransmitter diffuses across the synapse, etc) are random, although let's just accept that for many events in the brain the event would not be at the exact borderline of a threshold potential. (None of this could be known non-destructively, but we'll just leave that to one side for now.) So it's like predicting the weather: if you don't have absolutely accurate information about the starting state, inaccuracies could cause your simulation to spiral out of control very fast, and brute-force simulating would take a very very very powerful computer. Blythwood (talk) 00:21, 2 November 2016 (UTC)[reply]

I wouldn't count on it. Taking the nematode, for instance, that's 302 neurons with 9000 connections between them. The map has been known for decades, and brain-modelers still can't use it to predict anything from first principles. Someguy1221 (talk) 00:37, 2 November 2016 (UTC)[reply]

Well OpenWorm isn't doing too badly and have a look at [8] Dmcq (talk) 10:20, 2 November 2016 (UTC)[reply]

C. elegans is actually not the best example, because although it has the smallest nervous system, the neurons are so small that electrodes can't be inserted into them. As a result our understanding of how it works is very limited. A system that is understood much better is the stomatogastric nervous system of the lobster and other arthropods. It is probably the neural system that we come closest to understanding as an operational whole. Looie496 (talk) 15:04, 4 November 2016 (UTC)[reply]

How well did remarkable scientists from times past followed the publish or perish

Have scientists like Einstein, Feynman or more recently Hakings also followed the trend, and caved in to the pressure to publish as much as possible? Hofhof (talk) 23:35, 1 November 2016 (UTC)[reply]

The answer to that question requires knowledge we're unlikely ever to have of whether or not, say, Stephen Hawking or Richard Feynman are or were ever motivated by the "publish or perish" imperative. It seems unlikely of either man in their later careers. Stephen Hawking is a living testament to the British medical community's ability to keep someone suffering from amyotrophic lateral sclerosis alive well past the median survival period past onset for that condition, and in that time he has continued to publish professional and popular-audience work according to Google Scholar. Very little of that work likely is crucial to his tenure as Director of the Centre for Theoretical Cosmology, DAMTP, nor his tenure as Lucasian Professor of Mathematics at Oxford University. Likewise, by the time Richard Feynman attained fame for his work in quantum electrodynamics (for which he shared the Nobel Prize in Physics), his reputation was made, and his insights into such phenomena as the Challenger disaster were sought after irrespective of how many publications he'd made that year.

You'd really have to look on the Internet for any prominent researcher who'd actually said he felt the need to publish or perish. loupgarous (talk) 02:48, 2 November 2016 (UTC)[reply]

NB The Lucasian Chair of Mathematics is at Cambridge University. Wymspen (talk) 11:01, 3 November 2016 (UTC)[reply]

At times when scientific discovery contradicted deeply held belief, the risk was publish and perish. Galileo was lucky to get away with it. Wymspen (talk) 10:35, 2 November 2016 (UTC)[reply]

Le plus ça change Asmrulz (talk) 14:35, 2 November 2016 (UTC)[reply]

If you look at List of scientific publications by Albert Einstein, you can see that from his incredible "Annus Mirabilis" in 1905 to the late 1920s, he was constantly producing a lot of influential, genuinely innovative stuff. Then he got stuck on what we now see as the dead end of Classical unified field theories (theories that try to explain everything without using quantum physics). He didn't really make any progress, but he kept publishing. Smurrayinchester 09:59, 4 November 2016 (UTC)[reply]

Wikipedia:Reference desk/Archives/Science/2016 November 1

Contents