Wikipedia:Reference desk/Archives/Science/2013 December 13

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.

December 13

No seagulls in Malta

native range of the rock pigeon in deep red

When I visited Malta I realized that the "flying rat" niche in Valletta and Sliema was occupied not by seagulls as I expected in a European harbour but by pigeons. Talking with a resident of several months he also remarked this. Reading around Times of Malta, The Independent I see opinions that it is because hunting in Malta is so excessive. So my question is: Is really the population of Maltese seagulls very low or nonexistent or was I unlucky? If so, why? Why are pigeons not affected? --Error (talk) 01:20, 13 December 2013 (UTC)

Why would you hunt seagulls? I'm given to understand they taste terrible --Trovatore (talk) 01:30, 13 December 2013 (UTC)

So do foxes. What does that matter? Rojomoke (talk) 05:20, 13 December 2013 (UTC)

• Pigeons originally nested in stone cliffs rock pigeon, of which Malta has plenty, while probably not having so many nesting places for gulls. μηδείς (talk) 02:11, 13 December 2013 (UTC)

Yellow-legged Seagull eating a pigeon!
(but this one isn't in Malta)]

• You might contact BirdLife Malta Org., but our article List of birds of Malta lists 12 gull species which occur in Malta. ~E:71.20.250.51 (talk) 03:31, 13 December 2013 (UTC) ~P.s.: the Yellow-legged Gull is the only species that breeds in Malta; Here's one eating a pigeon! →

"Hunting" is probably not the right word, but when I visited rural Tuscany I saw very few birds of any sort because the locals take pot-shots at them (regardless of species). On the plus side, this did mean that the butterfly population did very well.--Shantavira|^{feed me} 10:32, 13 December 2013 (UTC)

Fetal consciouness

Is there a scientific consensus that fetuses don't have consciousness until the third trimester? It's a claim I heard someone make, and google gives conflicting results. 74.15.137.253 (talk) 03:28, 13 December 2013 (UTC)

They certainly won't have consciousness before quickening, which is when their musculoskeletal nervous system starts firing spontaneously. An awareness of the body requires feedback to the brain from the body. Before bodily motion occurs (after about four months) there would be no feedback to the brain, and no consciousness of the body. There certainly might be a further reason why awareness of the body might not occur until the sixth month, say, integration of the brain inputs. But I have no personal knowledge of this. See quickening, sensory cortex, motor cortex, Antonio Damasio, and somatic marker hypothesis. μηδείς (talk) 03:48, 13 December 2013 (UTC)

We don't have a widely accepted definition of what consciousness is, we have no way to measure it, we have no good idea of what mechanisms might be behind it - and there is even some doubt that it exists! Given all of that, saying whether a fetus - a baby - or even a toddler - has consciousness is essentially a meaningless question. Anything you've read about this is likely to be shots back and forth between Pro-life and Pro-choice lobbies - and neither set of opinions is to be trusted because there is simply insufficient science to know. SteveBaker (talk) 04:13, 13 December 2013 (UTC)

That's a rather silly, but common trope. If one doesn't know that toddlers or babies, or late-term fetuses are conscious, as opposed to sleeping or comatose, he should be confined to his residence and be denied visitors as a public menace. It is very easy to say that certain objects lack consciousness, based on theory. It is not so easy to define the exact point when it arises. But a useful answer to this question would be one that explains why someone would draw the line at six months, rather than saying one can't draw a line at all. μηδείς (talk) 04:24, 13 December 2013 (UTC)

(ec) ... No. In fact, there is no scientific consensus on what consciousness is, or how we define it. I'm baffled how the previous comment can posit such certainty; why should "spontaneous neuron firing" be a prerequisite for consciousness? There is not any scientific consensus on whether adult humans are conscious, let alone whether it is a property shared by infants, fetuses, rodents, flowering plants, natural geological processes, or anything else along the spectrum of complex, life-like activity. There are numerous political factions who seek to draw sharp legal demarcations about the calendar for human fetus development - mostly because these factions hold strong viewpoints related to the topic, rather than reasoning about any specific observable details. But the incessant bickering of pundits does not alter scientific consensus: human prenatal development is a gradual and continuously-changing process. We understand many details about the schedule of normal physiological development. But I think you will find very few scientists who subscribe to a theory that specifies an exact time-period at which conscious thought - or any other property - begins. Such things develop gradually.

Debates about consciousness typically do not fall in to the regime of scientific investigation. "Consciousness" is an unscientific term for an ambiguous grouping of philosophical concepts. It is used by many people in many disciplines to mean many different, mutually-exclusive things. So, the presence or absence of consciousness is unfalsifiable; at best, we can develop an operational definition that some scientists may agree upon; and test specific cases. I find that approach unsatisfying, and altogether tangential to the issue of how human thought actually works. I prefer to consider the sophisticated interactions amongst intelligent life forms in the context of emergent behavior in complex and self organizing systems. As more sophisticated biological machinery evolves - within the life-cycle of an individual organism, or during the evolution of an entire species - the machinery becomes capable of yielding ever more sophisticated interactions.

If you are interested in the subject, I recommend some of the writings on this topic by Stephen Jay Gould, one of the more influential scientists from the 20th century. His writings are very influential, but I would stop short of saying that any one individual's work reflects the "consensus." Nimur (talk) 04:29, 13 December 2013 (UTC)

Yes, more eloquently stated - and MUCH better researched...but, yes - exactly what I'm trying to say here. SteveBaker (talk) 15:30, 13 December 2013 (UTC)

What Steve is arguing, properly, is not that the question is meaningless, but that there is no way to find out the answer. A group of philosophers called the logical positivists, in the early part of the 20th century, promoted the idea that, if there's no way to find out the answer, then the question is meaningless. But they were utterly wrong. --Trovatore (talk) 04:29, 13 December 2013 (UTC)

I know where you're coming from - but is the question: "Does a zworch gooble a snarj?" meaningless - or do we just have no way to answer it? Clearly, the former. We have no definitions for the words "zworch", "gooble" or "snarj" - so the question itself has no meaning. From a scientific perspective, the same thing is true of this question because it uses the term "consciousness" - which has no solid scientific meaning (in the sense of "feeling of self"). There have been a few half-assed experiments - such as putting a smudge of paint on a sleeping chimpanzee's forehead then waking it and presenting it with a mirror and seeing whether it reaches up to remove the smudge. I'm not much convinced by that test. I could easily build a robot that could pass it - a mere machine that nobody would agree had "consciousness". Without even a definition of the word - we really can't accept this as a meaningful question. The "logical positivists" are not the only ones to take that view - essentially all of science is founded on the concept that "unfalsifiable" questions are essentially meaningless.

Well, either way - there isn't an answer for our OP here...and it's fairly clear that this is at least unanswerable, whether or not it's meaningless. SteveBaker (talk) 15:30, 13 December 2013 (UTC)

What is it like to be a bat? We have no way of knowing, or at least I know of no way of knowing. Nevertheless I insist that it is a meaningful question, and it has an answer, although it is just possible (I think it's unlikely) that the answer is that it is not like anything at all; that is, that bats do not have qualia. But either it is like something, or it is not like anything, and if it is like something, then it is something definite. --Trovatore (talk) 18:00, 13 December 2013 (UTC)

The state of the art doesn't seem all that elevated beyond Aristotle's concept of ensoulment, though maybe due to confusion of interpretation of the genital tubercle he came to a weird conclusion the different sexes took different amounts of time. Generally speaking I subscribe to the "brain life" theory which suggests consciousness may begin as early as eight weeks,^[1] though this does not translate directly into calling abortion murder at that point due to the general legal requirement of proof beyond reasonable doubt. (I've seen this figure for fetal electrical activity rendered as ten or even twelve weeks, though I'd have to reexamine the definition used - LMP vs. conception) Still, by and large, an ancient holding an aborted fetus in his hands probably had about the same degree of subjective understanding as we do, and we don't have much more than that. Wnt (talk) 07:46, 13 December 2013 (UTC)

I remain unconvinced that even the lowly amoeba doesn't have some level of "consciousness" (or "self-awareness"), albeit a level that we can't understand. ←Baseball Bugs ^{What's up, Doc?} carrots→ 10:39, 13 December 2013 (UTC)

In the absence of a definition of what "consciousness" is - it's impossible to argue anything about the subject...it's irrelevant what User:Wnt subscribes to - and it's unimportant that Bugs wishes to attach a meaningless term to an amoeba, a virus, a DNA strand or even a rock. Just about the only thing we know about consciousness is "we know it when we see it"...but even that's not reliably true because no two people would place their finger at the same point on the line between rock, amoeba, snail, lizard, mouse, dog, dolphin and human and say "To the right of here, everything is conscious and to the left, nothing is. Convincing people that Artificial Intelligences are or are not "conscious" is going to be a really interesting problem in the future...and issues of whether turning off an intelligent computer is murder will be right up there with when it's morally acceptable to terminate a pregnancy or whether it's OK to do medical testing on chimps. Until we understand this feeling we each have that we're "alive", we can't pin down a definition and we certainly can't devise tests that we can run on unborn children to see when they "get it". My opinion is akin to Bug's - that consciousness is not a binary phenomenon - it varies from what we feel - all the way down to what a single atom "feels" in a continuous spectrum that defies anyone to put a sharp line between things that have it and things that don't. SteveBaker (talk) 15:30, 13 December 2013 (UTC)

The appeal of "brain life" is that it is a symmetric definition. I can say that my willingness to allow someone to let me die if brain activity has ceased gives me some Golden Rule based moral right to abort fetuses that have not yet developed brain activity. However, like brain death at the end of life, it is an outer limit, not a guarantee anything much is really going on, and one should naturally be reluctant to turn to harsh penalties where it isn't clear a crime exists. So for the political definition of the limit for abortion laws one needs (at least) some backup indicators. Wnt (talk) 04:05, 15 December 2013 (UTC)

Is the question "when is a fetus sentient?" a more meaningful question, or is sentience as scientifically vague a concept as consciousness? 74.15.137.253 (talk) 18:17, 13 December 2013 (UTC)

There's a lot of screaming in the literature and insistence based on false philosophical principles that we cannot define or know what consciousness is. There's also the problem that the word is used to mean different things. Julian Jaynes, for example uses consciousness to mean reflective conceptual self-awareness. But it's ultimately silly to say we don't have a notion of consciousness, or we couldn't understand what it meant to deny something has consciousness.

If you define consciousness as sensory/perceptual awareness of one's body and its surroundings it is undeniable that (normal) newborns have consciousness and that before quickening fetuses do not have consciousness, because their nervous systems haven't made the necessary connections yet. Talk of amoebas and 8 week fetuses having consciousness is speculation based not on evidence or our understanding of the mechanisms by which brain activity leads to awareness.

So we have a window from four to nine months. Presumably this can be narrowed down somewhat. We do know that late-pregnancy babies recognize voices and respond to them and that they struggle in resistance to abortion, so we know they have some awareness. If, say, it were found that 5 month fetuses didn't struggle during abortions we might assume they are not conscious yet.

Going back to the original question where the OP said someone claimed there is no consciousness until the third trimester, does that correspond to any argument anyone here knows of, or can find a relevant source for? I have never heard such an argument, but I would like to hear it if it exists. μηδείς (talk) 19:01, 13 December 2013 (UTC)

Medeis, I have a few shelves full of computers that have sensory and perceptual awareness of their enclosures and their surroundings. These computers know what time of day it is; and they know how bright the room is. They know when they are hot; they know when they are cold. They know when they need to rest; they sometimes wake up in the night because they think they've heard a noise; but mostly they sleep very well. If you visit them, they can see you, and they know if you are a human. They can even identify which human you are by sight; and if you've met them before, they can remember your name, and who you are, and what you said last time, and where you came from, and what you talked about. They will even say hello in your language. Actually, some of these computers can say hello, and they can speak and answer questions in more languages than you. Are you going to tell me that these devices have less sentience than a newborn human, just because their brains are implemented in silicon, instead of meat? Is sentience measurable? Do you have a list of criteria that establish it, as either a "yes-no" checkbox, or even as a "gradient" or spectrum?

This is called the Turing test. It has been known to scientists and philosophers for decades; in earlier forms, it was known to philosophers many centuries ago. It is probably the single strongest argument against any effort you make to define "sentience." In this amazing era of human development, we are now able to formulate concrete example representations of the test that make sense to ordinary human brains who are not very good at abstract thought. We can even build a machine that looks like a human, if that makes the example easier for you to understand.

Here is what I posit. For any definition of sentience that you can provide, I will provide a machine counter-example which will either (a) refute your definition, or (b) leave you to ponder a severe moralistic quagmire. Nimur (talk) 20:16, 13 December 2013 (UTC)

There's a basic problem with the Turing test, in that it doesn't address the key question, "Does that machine know that it exists?" ←Baseball Bugs ^{What's up, Doc?} carrots→ 23:56, 13 December 2013 (UTC)

Maybe you don't understand. These machines speak English (and other human languages). You can ask them - point blank - if they know they exist. They answer in the affirmative. Perhaps that's just a clever in-joke that a natural language programmer placed in the source-code... but, so what? When a human answers in the affirmative to such a question, aren't they just processing a set of input sounds, according to rules that another human taught them? Nimur (talk) 02:26, 14 December 2013 (UTC)

I do understand. It's a flawed premise. However, you may find this of interest. ←Baseball Bugs ^{What's up, Doc?} carrots→ 02:32, 14 December 2013 (UTC)

You are using those words metaphorically or in different senses, Nimur. I am as ready to debate you as I am to debate Aristotle when he said rocks "desired" to role down hill because it brought them closer to the center of the earth--AKA the center of the universe. You are simply doing tendon wrenching mental gymnastics if you want to deny that at some point between conception and birth babies attain consciousness, or perceptual awareness, or the like, depending on what terms you prefer. Indeed, we can narrow that range down to somewhere between quickening and the beginning of the third trimester. But going on about "atoms" and "computers" is just crazy talk. μηδείς (talk) 20:25, 13 December 2013 (UTC)

It's by no means obvious to me that a newborn baby is conscious, or sentient, or whatever words we prefer. I'd like to second Medeis' request above for a citation that consciousness arises between birth and six months, with my gut feeling being that a _longer_ timescale is needed. Of course, from a _legal_ point of view, the moment of birth is critical, but that isn't the issue we're discussing. Tevildo (talk) 20:30, 13 December 2013 (UTC)

Again, this all depends on how you want to define the term conscious. I would point out that a newborn will be asleep at certain times, and not respond to stimuli, while it will also be awake at certain times, and aware of touch, sounds, and faces, and will react with pleasure and displeasure based on what it experiences. Of course the mind of a newborn is not as developed as the mind of a 20-year-old. But there's an obvious continuum of awareness. μηδείς (talk) 20:50, 13 December 2013 (UTC)

Indeed. The question "where on that continuum is the typical foetus/baby/child at a particular stage of development?" is one that is (potentially) answerable, and I would be interested in reading about any studies of it. However, "where on that continuum do we set the boundary of consciousness?" is more of a philosophical question that depends on our definition of consciousness, and "where on that continuum does abortion become unethical?" is a political question that may be informed by the answer to the first, but for which neither the first or the second questions can provide a complete answer. Tevildo (talk) 22:26, 13 December 2013 (UTC)

Did the OP ask about abortion? μηδείς (talk) 23:22, 13 December 2013 (UTC)

No - but this question only ever comes up in that context - and we well understand the point of the question. SteveBaker (talk) 02:31, 14 December 2013 (UTC)

Actually, I think this is a fascinating question whether or not I am about to have an abortion. Why wouldn't it be? (That's rhetorical.)μηδείς (talk) 02:37, 14 December 2013 (UTC)

The question of how we define these terms - and then measure them in various systems - is definitely fascinating. The question of if/when abortions are ethical is a separate matter. The sentience/consciousness of the fetus at various stages of development is only one of a very large number of other considerations - and (IMHO) possibly one of the least of them. SteveBaker (talk) 03:06, 14 December 2013 (UTC)

"Sentience" has exactly the same problem as "consciousness" - no solid definition, no way to measure it. The Turing test isn't a test of sentience - it's a test of whether an artificial machine is intelligent. There are very few 2 year old children and no higher non-human mammals who could pass it - yet we don't (generally) deny that they are sentient. SteveBaker (talk) 02:31, 14 December 2013 (UTC)

So, when doctors prick people to see if they respond, flash a light in their eye to see if it contracts or follows, when they ask the patient to give their name, say where they are, give the day of the week, name the president, or to count backwards from 100 by sevens, in none of these cases are they measuring anything that has to do with consciousness or sentience? μηδείς (talk) 02:37, 14 December 2013 (UTC)

Certainly not! Nothing whatever. It would be trivial to devise a computer that could do that. The questions you pose are easily answered by a computer like IBM's Watson (which annihilated the best human "Jeopardy" players in the world with ease). The physical tests are also very easy to pass - they are just a matter of having the right hardware. The my phone responds to a light shone into it's ambient light sensor by brightening or dimming the screen. If you drop my laptop, it anticipates the "pain" of hitting the ground by retracting the heads of it's hard drive before it hits the ground. Many much lower animals such as the humble amoeba would respond to a pain prick or light flash. Nobody thinks Watson is conscious or sentient - although "intelligence" is perhaps something some people might concede (I wouldn't). SteveBaker (talk) 02:52, 14 December 2013 (UTC)

Actually no, that's rather puerile. Again, you are using words like pain in a metaphorical sense, and you are also assuming that we know what pain is in regard to computers while denying that we actually know what pain is in regard to humans. The fact that a computer has been programmed to mimic humans proves two things. One that humans (not the computer) are clever enough to write programs that mimic humans and (2) That some humans are foolish enough to think such mimicry is comparable to the real thing. μηδείς (talk) 03:33, 14 December 2013 (UTC)

No, it's not puerile. You offered this "doctor's test" as a way to measure "consciousness" - and I've clearly shown that it does no such thing in a general case. You can't use it to test whether an arbitrary object is, or is not conscious/sentient/intelligent. All it actually tests is whether a system (specifically, a human being) that was once known to be conscious is still conscious...and if we can make a computer pass the test without actually being conscious (whether by mimicry or not), that is definitive proof that the test isn't solid enough to provide a definition. The uselessness of it as a test is self-evident in applying it to resolving the question at hand here. We can't even use this test to figure out whether (say) a 2 year old child is "conscious" or not...heck, even an adult human might not be able to count by 7's if they happen to be innumerate. We're talking about a solid scientific definition for this word - and a means for measuring arbitrary systems against that definition...and this is nowhere near to being that. SteveBaker (talk) 14:32, 14 December 2013 (UTC)

Jehovah's Witnesses have published information about fetal learning and memory (http://wol.jw.org/en/wol/d/r1/lp-e/102001411#h=9:0-10:899), quoting a Reuters report about research at the University Hospital in Maastricht in the Netherlands.

—Wavelength (talk) 21:24, 13 December 2013 (UTC)

From a Google search for reuters maastricht fetus memory, I found http://www.ncbi.nlm.nih.gov/pubmed/19630906 and http://www.eurekalert.org/pub_releases/2009-07/sfri-fsm070109.php.

—Wavelength (talk) 00:48, 14 December 2013 (UTC)

Yes - but learning and memory isn't generally taken to be a measure of sentience or consciousness - a very clearly non-sentient/conscious computer can easily learn and memorize just like a fetus - and amoebae have been shown to be capable of simple learning and memory. The point that the Jehovah's are trying to make is neither proved nor disproved by these results - regardless of whether they are well-performed experiments or not. SteveBaker (talk) 02:31, 14 December 2013 (UTC)

In other words you do have a definition--a set idea of consciousness--you just don't want to state it? You can't deny that you know what conscious is, then presume to judge whether certain states are typical of consciousness. μηδείς (talk) 02:41, 14 December 2013 (UTC)

I have (like everyone else here) a sense of "I know it when I see it"...but I don't have a definition or a test or measurement of it. I know that I am conscious and sentient - I'm fairly sure that other people are also conscious and sentient...although I'll concede that I don't know that for sure. I know that I would not label an earthworm as conscious and I wouldn't say that a virus was sentient...but Baseball bugs probably would. But if I ask 100 people: "What is the lowest animal who you'd say was 'conscious' or 'sentient'?" - then I guarantee that I'd get 100 different answers. Some people would deny consciousness to anything other than a human - others would concede that dolphins and chimpanzees are sentient - others would go as far as including dogs and cats - yet others would probably allow any mammal, some larger reptiles and birds - and some people (Baseball Bugs being a case in point) would include the amoeba. So what is the definition of these words if no two humans have the exact same understanding of the term? If we lack such definition - then how can science measure this thing? If you can't measure it - then how can you study it in any meaningful way whatever? Just about the only definition that everyone would agree on (well, very nearly everyone) is that humans have these attributes and rocks don't...but you'll get all shades of opinion for dogs. SteveBaker (talk) 03:01, 14 December 2013 (UTC)

This debate and posturing is rather pointless because a computer analogy doesn't hold much water as an argument. It's a rather straightforward strawman, because most people understand that we are conscious humanbeings that can feel pain and have lives that we are self-aware of and can measure, but rocks and our current computers are not anywhere close to what we constitute. We are not here to draw lines in the sand or argue over them and regarding the OP's question, some useful references have been given above, I suggest we shut down the debate though. --Modocc (talk) 03:18, 14 December 2013 (UTC)

Steve deserves credit for what he has admitted, he does have an intuition that certain entities have consciousness while others don't. As for dogs, of course they are conscious. They just don't have a conceptual/linguistic consciousness the way we do. They are more like babes and apes. μηδείς (talk) 04:38, 14 December 2013 (UTC)

But that's the entire problem here. A LOT of people would disagree with you about whether dogs are conscious - and because we have no test or definition for it we can't say whether they are right or wrong. For example: those people who would use this test of "consciousness" in a fetus to determine when it's ethical to terminate a pregnancy would have to also deny the right of an overloaded animal shelter to euthanize a dog based on the same reasoning.

The question you have to ask yourself is this: "What is the most primitive organism that possesses consciousness?" - you might decide that it's a hummingbird (say)...you might argue that Alex (parrot) was clearly conscious - so maybe the smallest birds are also conscious - but maybe a gecko isn't...or maybe an earthworm isn't. At some point on this scale, you're going to have to draw the line at what you, personally, believe possesses this attribute. Maybe an amoeba passes your test (Baseball Bugs says it passes his test)...but eventually, there must be a sufficiently primitive thing that you'd agree is not conscious...even if it's a DNA strand or a rock.

The problem is that no two people agree on where this cutoff lies. Which means that we can't use people's gut feelings to determine at which month a fetus becomes conscious - and if we can't all agree on that, then it's not a basis for setting a moral principle because moral principles need to be things we can all more or less agree upon. I suspect that you (like me and Baseball Bugs) would agree that consciousness isn't a "Yes/No" thing...that a dog is less conscious than an adult human but more conscious than a mouse - which in turn is more conscious than an earthworm and so forth. Once you decide that - then consciousness isn't a "Yes/No" kind of thing and you can't ask: "Is a second trimester baby conscious? Does it become conscious in the third trimester?" because you've already admitted that this is a sliding scale. You could try to say "Is a second trimester baby's consciousness level about that of a mouse or about that of a dog?"...but even with your personal scale of consciousness fully determined - you'd have to admit that we have no means to do that measurement without an analytical definition of consciousness.

This would be different if we could say (for example) that: "Consciousness in mammals is a result of brain activity in such-and-such group of cells in the frontal cortex - and a measure of consciousness is the number of firings of neurons in that part of the brain measured over a 24 hour period."...with that definition in place, we could design a "consciousness-o-meter" and point it at things to get a reading - and pass laws to say that it's unlawful to kill a creature if the meter reads more than 7.54 mConscs. But without such a definition - this question simply cannot be answered. SteveBaker (talk) 14:32, 14 December 2013 (UTC)

Does the Turing test include questions like "What is your favorite color?" or "What kind of music do you like?" Or how about, "What do you like to think about when no one is asking you anything?" ←Baseball Bugs ^{What's up, Doc?} carrots→ 03:58, 14 December 2013 (UTC)

Really Bugs? You couldn't even be bothered to read the article about it? How can you possibly claim to be helping someone with a question when you can't be bothered to do even the most basic research required to make a useful contribution? SteveBaker (talk) 14:32, 14 December 2013 (UTC)

The other day you were lecturing me about how something works on Wikipedia, and you were wrong on all three counts. So, you worry about your own most basic research, and I'll worry about mine. ←Baseball Bugs ^{What's up, Doc?} carrots→ 14:48, 14 December 2013 (UTC)

The Turing test is not a set series of questions. In simplest terms, the Turing test merely consists of a normal human interviewing the subject in natural language with whatever the interviewer wants to ask. Given enough time and creativity, a human will always be able to deduce the difference between a machine and another human, THAT is the Turing test. It should be noted that it is far easier for a human to mimic the responses of the machine; that is it is quite easy for a human to fake being a machine to fool the Turing test. It isn't possible for a machine to fake being a person; unless of course it really WAS sentient/conscious/whatever because the ability to "fake" it requires that. --Jayron32 04:45, 14 December 2013 (UTC)

"It should be noted that it is far easier for a human to mimic the responses of the machine" -- no, that is exactly wrong. A human does not have a snowball's chance in hell of mimicking even a 1950's computer. If you don't believe me, answer this question within 2 seconds: what is 3178350710375067*10238590170186037143? A modern computer can do this calculation before your nerves can even transmit a signal from your eyes to your brain. --Bowlhover (talk) 07:18, 16 December 2013 (UTC)

Specifically, a human subject and whatever system you're trying to test are each given a computer connection to a researcher. The researcher can type questions and receive answers from the subjects but is not told which is which. Turing said that if the researcher cannot tell which of them is which, then the system you're testing is "intelligent". It's a tough test to pass - but some of the Loebner prize winners come surprisingly close sometimes. However, most serious Artificial Intelligence experts do not believe that Turing's test is a useful one. So they can certainly ask "What is your favorite color" or any other of Bug's questions. Those are fairly easy though - the tough ones for a computer to fake are things like: "Imagine that your garage has a shelf full of paint cans and suddenly the shelf support at one end breaks. Please give me a rough description of what you think you might see afterwards." I suspect that (for example) a two year old child and a chimpanzee could answer that question (if it could type and speak English - which are not really things we associate with intelligence) - but no computer we've built could come anywhere remotely close to coming up with a convincing answer to questions of that nature. SteveBaker (talk) 14:32, 14 December 2013 (UTC)

Why are you juxtaposing a non-scientific point in time with a baiting question? It's like 'Do 13 year-old girls menstruate?' There is no clear answer and it does even come close to answering whether the are ready to bear children. For most abortion laws the the third trimester was chosen due to the presumption that prior to that point, the fetus would not survive outside the womb. It's an arbitrary point in time development unrelated to anything other than viability. "Age of consent" is similarly arbitrary. [UNSIGNED]

All the Turing Test does is prove some people can be fooled by computer simulations. μηδείς (talk) 04:34, 14 December 2013 (UTC)

Did you happen to see the Dilbert strip that I linked to earlier?[2] ←Baseball Bugs ^{What's up, Doc?} carrots→ 05:03, 14 December 2013 (UTC)

Hmm, there's no guarantee that many humans aren't stupider than computers. How does that make you feel? μηδείς (talk) 05:48, 14 December 2013 (UTC)

Above average. :) ←Baseball Bugs ^{What's up, Doc?} carrots→ 06:20, 14 December 2013 (UTC)

The Turing test was Turing's proposal for an operational definition of machine intelligence. You don't have to accept his definition, though it would be nice if people who didn't would propose an operational definition of their own. It would also be nice if they learned what the test actually is, perhaps by reading his paper. In any case, as a definition, it can't prove anything. The Loebner Prize does demonstrate that people are gullible, both in being fooled by conversation bots that are not noticeably more advanced than those developed in the 1960s, and in believing that the Loebner Prize has something to do with the Turing test. I'm not sure what any of this has to do with fetal intelligence, not that I want to get involved in that debate anyway. -- BenRG (talk) 08:10, 14 December 2013 (UTC)

I think that it demonstrates that we don't have a good definition (or test) for common concepts like "intelligence", "sentience", "consciousness", "self-awareness" or even "thinking". Turing's test was a great way to start a conversation...but it's widely understood that a machine that was successfully programmed to be "intelligent" might easily fail the test because of a lack of "knowledge" - and that a sufficiently clever (but clearly unintelligent) "chatbot" might fool enough people to pass it. So it's not the test that most of us would want to use to ascribe this property of "intelligence" to arbitrary systems. I think that most dogs possess "intelligence" - yet they certainly can't pass the test because even if they could speak fluent English and type, they can't answer questions like "What color is Santa Claus' beard?". The Turing test really only asks you the question "Is this system a good emulation of the human brain?" - which isn't the same thing as asking whether it's intelligent or not. SteveBaker (talk) 14:32, 14 December 2013 (UTC)

Computing Machinery and Intelligence (1950) (PDF) is cited as Turing's most compact description of what we call his "test" today. Actually, he proceeds along the line of reasoning that I used: because we can't agree on our definitions, let's instead reformulate the problem and consider how well a machine can imitate human behavior - up to the point of indistinguishability. Alan Turing describes some of the limitations of technology in 1950 terms, just as René Descartes similarly described such limitation for his hypothetical automata in the seventeenth century. Nimur (talk) 15:40, 14 December 2013 (UTC)

Again, it's utterly pointless to claim that people cannot/should not define what we measure in terms of consciousness. Doctors measure peoples' perception of pain on a daily basis, but it is somehow unscientific to do so because people cannot agree on it? That is far too cynical to be of any utility, for the claim that there is an inherent lack of precision does not imply inaccuracy. Let's move on. --Modocc (talk) 16:13, 14 December 2013 (UTC)

What you call "cynical," I call "logical," and Alan Turing calls "impolite." "It is in fact the solipsist point of view. It may be the most logical point of view to hold but it makes communication of ideas difficult. A is liable to believe 'A thinks but B does not' whilst B believes 'B thinks but A does not.' Instead of arguing continually over this point it is usual to have the polite convention that everyone thinks." (Emphasis added), I think Mr. Turing is telling you all to please stop this impropriety, and asking you to politely concede that my machines are sentient; and in exchange, I will politely concede that you and your fetuses are sentient, (even if we each secretly know the other side is completely wrong), and we can all politely move forward with our lives. Nimur (talk) 16:24, 14 December 2013 (UTC)

You can learn about fetal origins at Ted.com talk - What babies learn before they're born - YouTube (19:25). (At about 12:32, the speaker uses a common mispronunciation of the word processes, ending with "eez", which is not even correct Latin, in which the plural form is prōcessūs.)

—Wavelength (talk) 17:10, 14 December 2013 (UTC)

That's a red herring. 'Processes' is a purely English word, the plural of 'process'. It's very normal in certain (British) idiolects to pronounce plural nouns like 'processes', 'focuses', 'circuses' etc with an -eez ending. I'm not sure whether it applies to verbs like 'processes', 'fixes', 'messes' etc. -- Jack of Oz ^{[pleasantries]} 19:24, 14 December 2013 (UTC)

(Admittedly this is not proper science; but it is the explanation I have been driven to...) Consciousness is, fundamentally, a paranormal phenomenon. There is no combination of equations e + H₂O + 17 eV + Na⁺ -> "I really feel it, I'm not just making noises that sound like I am". We should therefore consider that consciousness, being a paranormal phenomenon not recognized by science, has a paranormal cause. Traditionally indeed this has been quite popular. I would argue that consciousness exists only where free will exists (also a paranormal phenomenon!); that the ability to feel is based on an ability to act. And I would further argue (very idiosyncratically) that the functioning of truly free will - not random numbers, not causal consequence, but real free will - depends on acausal events, which depend, ironically, on the precognition of the immutable future to create multistable solutions for loops of causality over short time frames. This phenomenon, however, is a sort of "working fluid" kept in a suppressed form, dedicated solely to the action of free will, in order for the free will itself not to be badly degraded; also, because uninhibited it is extremely dangerous, and prone to be taken advantage of by future events. The creation of consciousness, therefore, depends on some method, perhaps quantum mechanical in nature, that makes this possible. It is difficult to suggest a way forward with this, however, because of the moral problem with the investigation. Wnt (talk) 04:25, 15 December 2013 (UTC)

This is traditional substance dualism, as expressed by Descartes and Roger Penrose. A perfectly respectable metaphysical position, but not really amenable to scientific investigation. Tevildo (talk) 10:32, 15 December 2013 (UTC)

Is Penrose a dualist? (Dualism is definitely not a currently respected view in any modern-non religious context I am aware of.) I thought Penrose linked consciousness to cell microtubules. μηδείς (talk) 01:25, 18 December 2013 (UTC)

mice in winter

I'm in Connecticut. How do field mice survive the cold of winter? Do they hibernate? (There's nothing about that at mouse but it also doesn't say they don't.) RJFJR (talk) 04:09, 13 December 2013 (UTC)

The article you need is actually called Wood mouse - (which is another name for the field mouse) and it says: "During the colder months wood mice do not hibernate; however, during severe winter seasons they fall into a sort of torpor, a decrease in physiological activity." SteveBaker (talk) 04:15, 13 December 2013 (UTC)

Actually, the wood mouse (Apodemus sylvaticus) is not native to the Americas; White-footed Mouse is the most common mouse in Connecticut^[3]. See also: torpor

For details or sources, search in Google Scholar: white-footed mouse torpor winter ~ E:71.20.250.51 (talk) 05:12, 13 December 2013 (UTC)

Incidentally, dormice (Europe, Africa and Asia) do hibernate.--Shantavira|^{feed me} 10:37, 13 December 2013 (UTC)

My sister, who lives in Massachusetts, just called me upset on Monday that she had killed her second white footed mouse of the season with a standard spring trap. She's tried have-a-heart traps, but the mice chew their way out. She says the mice come indoors every winter and build nests from things like pot holders they chew up. A few years ago, when my nephew was about 3, they had a mouse and she took him shopping, and to get a trap. To his mother's mortification, and a bit confused taxonomically, he joyfully told the woman at the checkout register, "We have a rabbit. We are going to kill it." μηδείς (talk) 17:38, 13 December 2013 (UTC)

"they had a mouse and she took him shopping" ? I suppose to the local cheese shop ? StuRat (talk) 09:43, 15 December 2013 (UTC)

Mice don't like cheese. They took the mouse shopping to the rabbit-killing store. It was the mouse's speaking that shocked the cashier, not that he said they were going to kill a rabbit. μηδείς (talk) 00:45, 17 December 2013 (UTC)

Outer planets question

Hi again, On Dec 10th I asked a question about the outer planets (see above) but unfortunately didn't get any answer. I don't know if the question was not comprehensible or if it was posted in the wrong place - if so, where should it be?, could anyone help me please (as I'd really like to know the answers), thank you very much. SGAst (talk) 13:52, 13 December 2013 (UTC)SGAst

Sorry you didn't get a response. This does seem like the right place, to me (an argument could be made for the Math Desk, but I think this is the better choice). I can't do the math myself, but would like to note that they will never be in a perfect line, as they aren't exactly in the same plane. And, even if they were, also note that such conjunctions don't cause anything to happen. One final comment: they may also appear to be in a line, from Earth, when they are not collinear, in 3D.StuRat (talk) 14:10, 13 December 2013 (UTC)

The Jupiter-Saturn conjunction is [conjunction] every 19-20 years, and according to what I understand the Saturn-Uranus conjunctions are 1896, 1988 and 2079; therefore in 2261 (the date I need confirmed) there would be a complete Jupiter-Saturn-Uranus conjunction (or not?). Also, how do I calculate the distance in that moment. In a straight line I have in such a conjunction approx. Jupiter -> Saturn 4.5 au, Saturn -> Uranus 9.5 au, Earth -> Uranus: total 18 au, but as you say it isn't a straight line, how is the interplanetary distance calculated then or is there any webpage to show this?? If anyone could help, much appreciated, thanks — Preceding unsigned comment added by SGAst (talk • contribs) 14:28, 13 December 2013 (UTC)

I seem to see a logic error there. For two planets, "conjunction" apparently just means they are the closest they get to each other in their orbits, as two planets always form a line. For 3 planets, however, the time when they are closest together and when they most closely fall into a line is not likely to be the same.

Also, I think there's sufficient error in those calcs, "give or take a year or two", that the two conjunctions aren't likely to line up precisely. So, the 3 planets will probably be somewhat near each other then, but not particularly close to being in a line. StuRat (talk) 14:58, 13 December 2013 (UTC)

Online solar system model ?

There should be computer models out there which allow you to run the solar system forward and backward to any particular date. If anyone knows of such a site, that would answer the above Q nicely. StuRat (talk) 14:58, 13 December 2013 (UTC)

Sky and Telescope magazine has a suite of interactive online charts and almanacs. They even have a very nice high-precision Jupiter Moons code implemented in Javascript. But why would you want such a tool on the web? Stellarium is free software for most common platforms; KStars is excellent free software for Linux; and CyberSky [4] is great free-to-use shareware for Windows. These tools don't require a network connection, which is handy for the observer who's off the grid. Nimur (talk) 15:10, 13 December 2013 (UTC)

In my case, I want to be able to answer this Q quickly without having to download and install anything, so an online model is best. StuRat (talk) 15:37, 13 December 2013 (UTC)

You'd better hurry; there are only a few hundred years until the conjunction, and there is no time to ensure that your methodology is sound. You had better take the very first answer you find on an internet webpage. Nimur (talk) 15:44, 13 December 2013 (UTC)

My time constraints are not due to motion of the planets, except maybe the Earth's rotation, as I get sleepy when it get dark here, so leave the Ref Desk alone and go to bed. Also, are you saying the site I found is inaccurate ? StuRat (talk) 16:30, 13 December 2013 (UTC)

I can't assess the accuracy of your link, as it requested me to download and install software that I do not normally install on any of my computers. Typically, when I see web content that is implemented in Adobe Flash, I proceed no further, and I am rarely disappointed. This is, perhaps, an unfair personal bias that I have developed, but it's a strategy that's been working for a while across many types of computers, and it does amazing things to improve my user-experience browsing the web.

As a FORTRAN programmer yourself, StuRat, don't you find it personally insulting to see astronomical data calculation programs implemented in Adobe Flash? For me, this sort of thing is an affront against my better nature. Nimur (talk) 16:42, 13 December 2013 (UTC)

No, I use Flash all the time, and this seems like a good way to implement a light version of the software in question. StuRat (talk) 15:38, 14 December 2013 (UTC)

I found an online one: [5]. This seems to show they are close to each other in 2080 and again somewhat close in 2259 (not 2261). StuRat (talk) 15:11, 13 December 2013 (UTC)

That link of yours http://www.solarsystemscope.com is absolutely perfect and just what I wanted. Thank you very much. SGAst (talk) 16:00, 13 December 2013 (UTC)SGAst

Your welcome, did you find all the controls ? You can drag the solar system so you are looking at it from above, for the best view, and zoom out, so you can see it all. Also be sure to change how fast it moves through the days so you can quickly get near the date you want, then slow it down to watch it closely. They also have controls to let you view it at actual scale, but I wouldn't recommend that, as everything but the Sun would just be a single point of light. StuRat (talk) 16:26, 13 December 2013 (UTC)

Yes, I've been playing around with it and it's a wonderful tool. The only thing I'm can't work out is how to measure the distances. For example, as you say, in 2259 there is a very good allignment of Jupiter-Saturn-Uranus, so I would guess that in a straight line in that year the distance is approx.:-- Jupiter-> 4.5 au -> Saturn -> 9.5 au -> Uranus. In 2256 though there are big differences, so it would take a theoretical spacecraft say three times as long, for want of a correct number, to get from Jupiter -> Saturn in 2256 than in 2259 (my maths is terrible so I don't even know where to begin). That would have been of interest, to see actual distances between planets, but the site as such is still brilliant. SGAst (talk) 17:18, 13 December 2013 (UTC)SGAst

Yes, it doesn't appear to have a tool for measuring distances. The best you could do is set the distances to the realistic setting, and measure them on the screen, using the Earth-Sun distance as 1 AU. This will give you very approximate distances. However, as noted below, a spacecraft moves slowly enough that planets move significantly in their orbits during the trip. StuRat (talk) 15:42, 14 December 2013 (UTC)

No spacecraft moves that fast. You would have to consider the planets' locations when the spacecraft reached them. New Horizons (our fastest yet) passed the "orbit of Mars on 7 April 2006, Jupiter on 28 February 2007, the orbit of Saturn on 8 June 2008; and the orbit of Uranus on 18 March 2011." Over 4 years to get from Jupiter to Uranus. Rmhermen (talk) 19:08, 13 December 2013 (UTC)

A key term to use in searching might be orrery. DMacks (talk) 15:18, 13 December 2013 (UTC)

Jupiter, Io, and Europa in Celestia -- OBSIDIAN†SOUL 12:50, 14 December 2013 (UTC)

Yes, and more. Celestia. I actually remember downloading it and playing with it during an eclipse once. Note that unlike Stellarium (which is a planetarium - i.e. a sky simulator from a fixed point), Celestia is an actual "space exploration" simulator. As in you get to see the planets, the comets, the stars, etc. and their movements in 3d. -- OBSIDIAN†SOUL 12:50, 14 December 2013 (UTC)

EDIT: Oh sorry, you meant "online" as in browser-based. My bad then. :P It requires a pretty hefty download. But yeah, I would recommend the OP download this for any questions on conjunctions, eclipses, etc. really. It answers all of them. -- OBSIDIAN†SOUL 12:55, 14 December 2013 (UTC)