Science & Nature Archive

Monday, July 22, 2013

Sketching Art Masterpieces from Memory

Some of my fellow engineers and I watch the show, Brain Games, on the National Geographic Channel (I highly recommend watching the show if you haven't seen it, yet). On one episode, they performed an experiment based on the study, The Science of Cycology, by Rebecca Lawson. Subjects were given a basic, incomplete layout of a bicycle, showing just part of the frame, the wheels, the seat, and the handlebars, and then asked to fill in the rest of the drawing with the pedals, chain, and completing the frame. Below is a copy of the basic layout that they were given.

Cycology Skeletal Layout

This may at first appear to be a simple task, but it turns out to be surprisingly difficult for a large number of people. If you want to see what types of drawings Lawson got from the test subjects, follow that link above to read her paper (it's not terribly long, and you can jump ahead to the interesting parts, anyway). If you really want to play along, try it for yourself before following the link.

The study listed errors in frame, pedals, and chain independently, so I'm sure there was some overlap in the errors, and I'm unsure what that overlap was, but nearly half of the drawings had the chain drawn incorrectly to where the bike wouldn't work. In other words, at least around half of the drawings were wrong. If some people got the chain right but made mistakes in other parts, then the number is even higher.

The hypothesized reason is that our memories really aren't as good as we think they are. While we all know a bicycle when we see one, and think we have a good memory of just what a bicycle looks like, the truth is that many of us remember only just enough to recognize the bicycle, but not much detail.

Of course, us engineers in the office had no problem drawing a bicycle from memory. In fact, we didn't even need the skeletal layout to start off with. We could all just draw it from a blank sheet. But I didn't think this was exactly representative. For one, we're all mechanical engineers (if you count aero as a subset of mechanical), so this is exactly the type of thing we pay attention to. For another, with our bent towards mechanical design, we don't actually have to remember what the bike looks like. We can just remember a few details, and then fill in the rest as we go to make a functioning product. That's the type of thing we do on a daily basis.

So, we got to talking about another way to test Lawson's hypothesis. Was there something else that we should be familiar with, that we would recognize instantly, to try drawing from memory. And we decided that art masterpieces were the perfect objects. These are things you see repeatedly throughout your life. And we couldn't use our mechanical aptitude to fill in details. We had to rely on memory.

Before I go on, I'll list the pieces that we attempted to draw. If you want to have some fun, try drawing them for yourself before you scroll on further (or before you click on the links).

[Intentional blank lines to leave some space so that you don't accidentally see the real versions if you want to draw them yourself.]

.

.

.

.

How did we do? Pretty poorly, and not just on artistic merit. I don't have copies of every sketch that we did, since the other engineer who was trying it was doing his sketches on a white board and had to erase each one to have a clean canvas for the next masterpiece. But I was doing mine by pencil on scratch paper, so I at least have all of mine. Below are my sketches and one of his, followed by the real masterpiece.

The Mona Lisa

Jeff's Mona Lisa Sketch The Real Mona Lisa

The Scream

Jeff's The Scream Sketch The Real The Scream

The Persistence of Memory

Jeff's The Persistence of Memory Sketch
Martin's The Persistence of Memory Sketch
The Real The Persistence of Memory

We may have gotten a few of the big details right, but we really missed a lot when it came to filling it in. We both missed many details in the Mona Lisa, even something as prominent as her hands. Neither of us had the extra people or the sailboats in The Scream. And my Persistence of Memory was particularly dismal (the other engineer's was pretty good, but he admitted to having painted a copy of it in an art class).

I can think of another permutation on this, actually similar to what Lawson did in another section of her study - multiple choice. How many of us would be able to pick the correct version of a masterpiece given multiple similar choices (either through alterations, or probably even more difficult, through the time honored tradition of hand painted reproductions).

Anyway, while this post is nominally related to Rebecca Lawson's study, and our anecdotes fall in line with what she's proposing, the real reason I posted this was because it was fun. Get together with some friends and try doing this for yourself. Pick a masterpiece, try sketching it from memory, and then compare your sketches to each other and to the real thing. If you're like us, you'll be surprised at how little you can remember, but you'll have fun doing it.

Monday, July 15, 2013

Which Came First, the Chicken or the Egg?

Which Came First, the Chicken or the Egg?Jerry Coyne recently started a thread on his website in the entry, The Eternal Question, asking his readers to weigh in on the question, 'Which came first, the chicken or the egg?'. Some of his commenters left answers I largely agreed with, and I even added my own two cents in response to one comment, but I thought I'd leave my full thoughts here.

The short answer is that evolution is a gradual process that occurs among entire populations, and species are such a fuzzy concept, that it doesn't even make sense to think of a first chicken or a first chicken egg*.

One of Coyne's commenters suggested that even if the line is arbitrary, you should be able to draw it somewhere. What if you set up your dividing line such that there was a single last mutation responsible to shift a genome from an almost-but-not-quite-chicken to a true chicken? And that mutation finally appears in one organism, so presto, it's a chicken. But if you think about this threshold a bit more, it doesn't make sense. Imagine that your first chicken grows up, finds a mate, and reproduces. Since the chicks get half of their genetic material from each parent, and the division is random, somewhere around half of the 'first chicken's' offspring will get the mutation that defined it as a chicken, and around half will have the old version of that gene. So only half of its offspring would be 'true chickens' - there would be brothers and sisters that were different species! And then all those chicks would grow up and have their offspring, and on and on, and you'd end up with a breeding population composed of a mix of almost-but-not-quite-chickens and true chickens. That's just silly, and doesn't even meet the biological definition of a species.

In reality, species is a very fuzzy concept. The biological species concept is the one most used for sexually reproducing animals. Wikipedia gives a definition for it as follows.

A biological species is a group of individuals which can breed together (panmixia). However, they cannot breed with other groups. In other words, the group is reproductively isolated from other groups.

So, any organism will always necessarily be the same species as its parent. It's only after generations of reproductive isolation that two groups will gradually change to be sufficiently different that they won't be able to interbreed.

There are some interesting modern examples that show how even the biological definition of species can be difficult. One is polar bears and grizzly bears. They can, in fact, interbreed to create fertile offspring. But they don't usually do so naturally. So, it's not that the reproductive isolation has to be complete. It just has to happen little enough that the gene pools don't do too much mixing. How much is too much? Who knows. That's one of the grey areas.

A very interesting case is what's known as a ring species. These are animals that have a range that encircles some type of barrier. A classic example is the Larus gull, which lives in a band around the Arctic Ocean. If you start with the European Herring Gull which lives mainly in Great Britain, it can mate with the American Herring Gull to the west. And they do this often enough that their gene pools mix, which indicates that they're merely subspecies, not completely different species. And if you go west from there, the American Herring Gull can mate with the East Siberian Herring Gull. And you can keep going west, with the groups being able to interbreed, all the way until you get to Lesser Black-backed Gulls, which live mainly in Europe but which also stray into Great Britain. But guess what, the Lesser Black-backed Gulls in Great Britain don't mate with the European Herring Gulls on the same island. So how do these animals get classified as a species? There's one large interbreeding population right now, which would indicate one species. But what if all of the subspecies were to go extinct except for Lesser Black-backed Gulls and European Herring Gulls? Would they instantly become two new species?

Perhaps a more familiar example is dogs. Everybody knows that dogs can interbreed. That's where muts come from. But what if some super-villain were to come along and kill every breed of dog except for Chihuahuas and great Danes? Now, I know that technically you could probably give them a hand to do the deed and make puppies, and they might even be fertile. But, if left to their own devices, they'd be two pretty effectively isolated breeding populations. (I suspect this page is either a joke or an urban legend, but it still reveals the difficulties that would be involved). So, given the current dog population, chihuahuas and great Danes are part of the same species. But there's no way they would be considered the same species if they were the only dogs left in existence.

Given that species is such a fuzzy concept to begin with, it makes no sense to think of a 'first' of any species. There are gradually changing populations, and there's no point where you can pick one organism as being a different species from its parent.

So, the next time you hear someone ask, 'Which came first, the chicken or the egg?', you can tell them to go learn some biology before asking such a silly question that doesn't have an answer.

Image Source: Brain Pickings


*I'm assuming that the question implies it's a chicken egg. It doesn't make much sense to ask which came first, the chicken or the dinosaur egg, because then there's no conundrum at all.

Updated 2015-02-26: Slightly reworded 3rd paragraph for better flow.

Monday, June 17, 2013

Leaving Comments on Other Sites - Birds as Dinosaurs and Fossil Evidence for Evolution

Archaeopteryx - Berlin SpecimenOne of my habits when I'm getting ready to write a blog entry is to do a quick Google search to see if anyone's written anything along the same lines, before. If I find something that's very similar to what I was intending on doing, then there's no reason for me to repeat what's already been done. Sometimes I'll change direction on what I was going to write, and sometimes I'll just table the concept entirely.

Well, in the course of googling for the entry, Birds Are Dinosaurs, I came across a blog, Across the Fruited Plain, which had an entry, Are Dinosaurs Alive Today As Birds?: Refuting Archaeopteryx as "Evidence" for Evolution. Reading through the comments, I followed a link to another of his blog entries, Refuting Fossil "Evidence" for Evolution: The Data is NOT in the Strata. Despite it not being a particularly active blog, I caught a case of SIWOTI syndrome and couldn't resist commenting. Unfortunately, those comments are held up in moderation. My guess is because the owner of the blog just isn't very active in maintaining it (he's only posted three new entries so far this year). But, the only cure for SIWOTI syndrome is to see your comments get published somewhere, so I'm putting them here. So, if you just happen to be a regular reader of Across the Fruited Plain, here are some comments relevant to posts on that site.


First, here is my comment to his article, Are Dinosaurs Alive Today As Birds?: Refuting Archaeopteryx as "Evidence" for Evolution.

I tried leaving a comment to this article a couple days ago, but it didn't go through. If it's simply held up in moderation, then I apologize for being redundant.

I have a question for you, but first some background. Ignoring evolution, most people agree that organisms can be grouped into nested hierarchies. For example, there are prokaryotes and eukaryotes, with animals being one group of eukaryote, and then vertebrates as one type of animal, and mammals as one type of vertebrate, etc, etc. So, for example, in the group we call mammals, there are animals as diverse as whales, bats, platypuses, dogs, elephants, people, etc. These are all very different animals, but share common traits that are unique to mammals, so they all get grouped as mammals. Personally, I think that evolution is the best explanation for these nested hierarchies, but maybe that's just the way that a god/gods (depending on your religion) liked to create things.

So, if you look at say, a chicken, a deinonychus, and an ornithischian dinosaur like a stegosaurus, it seems that the chicken and deinonychus have much more in common than either does with the stegosaurus. They're bipedal, have feathers, hollow bones, an air sac respiratory system, etc. And if you pick a bird like archaeopteryx, then it has even more in common with the deinonychus, right down to the sickle claw.

So my question is, ignoring evolution, would you at least classify birds as a type of dinosaur?


Next, here is my comment to his article, Refuting Fossil "Evidence" for Evolution: The Data is NOT in the Strata.

I know this is an old article, but I couldn't help commenting on it. Here are some responses to statements you made, grouped by the headings you used.

Lack of Transitional Forms Disprove Fossil Evidence for Evolution

First of all, why would you expect there to be countless fossils of every evolutionary transition? For example, the modern phylum of platyhelminthes, or flatworms, consists of thousands of species, yet there's scant fossil evidence of these organisms. If living organisms are absent from the fossil record, why would you expect all extinct organisms to be present? Fossilization is a rare event, and it's even rarer still for fossils to be exposed in a location where humans can find them.

How can you claim there are not transitional forms? What about archaeopteryx, tiktaalik roseae, pakicetus, rhodhocetus, dorudon, australopithecus? What would you expect of a transitional form?

Your understanding of punctuated equilibrium is very muddled. You've described what's known as saltationism, which simply couldn't work in sexually reproducing organisms - where would the 'hopeful monster' find a mate? Rather, punctuated equilibrium describes periods of relative stasis punctuated by periods of change rapid on a geological timescale - thousands of years rather than tens or hundreds of thousands. In reality, both punctuated equilibrium and gradualism are detectable in the fossil record.

Dating Methods

Ideally, the way dating works is to find layers of igneous rock above and below what you want to date. The igneous rock can be dated very accurately with radioisotopes (I know many young earth creationists don't trust atomic theory when it comes to radiometric dating, but this really is accurate). If no igneous layers are bracketing the sample you want to date, then you can rely on index fossils. These are species that were very abundant but only alive for only a very short time, and so only appear in limited stretches of the geologic column. In fact, these index fossils were recognized before radiometric dating, and used to establish relative ages of different layers. In modern times, there have been enough of these index species dated relative to igneous layers that you can be reasonably certain of the age of a sedimentary layer even if all you can find are the index fossils. But it's only these special index fossils that can be used to date layers, not any of the other fossils you happen to find in them.

Distinct Strata Identification

I'm not really sure what you're getting at, here. I don't know of anybody who would propose a date for a fossil based solely on finding it in limestone. As discussed above, you'd have to have at least index fossils, or ideally, igneous rock above and below the limestone layer you're looking at.

No Fossil is Conclusive Evidence for Evolution

Very true. A single fossil is not evidence. It's the pattern that emerges when you compare multiple fossils. For example, I cited a few examples above of whale evolution. Finding any one of them in isolation wouldn't be terribly strong evidence for evolution. But when you find multiple fossils like indohyus, pakicetus, ambulocetus, kutchicetus, rodhocetus, dorudon, and basilosaurus, it presents a much more cohesive picture.

The Fossil Evidence Supports the Biblical Worldwide Flood

First of all, most animal fossils are not of whole, complete animals. Most are fragmentary, the result of predation and scavenging. And the fossil record doesn't at all match what would be expected from a world wide flood. Organisms are found only in specific strata. Now, I know that some creationists like to explain this with 'hydraulic sorting', or positing that organisms got grouped by their ability to escape rising flood waters, but that doesn't match the reality of the fossil record. And that would still only be an average. Surely, if a worldwide flood had occured, some 'fast' animals would have died for various reasons before reaching higher ground. Yet there are no fossil rabbits in the cambrian, nor are there any ammonites that happened to make it to a higher strata (to pick just two examples). There are too many other problems with a global flood to list here, so I recommend googling "problems with a global flood talk origins" and reading that article.


Update 2015-02-23:My comment was finally approved on that site, and it spawned an entire debate. I also had a few follow-up posts on this site. For a summary of all the posts on this site dealing with this, take a look at Creationist Dishonesty and a Follow Up to Previous Entries.

Thursday, June 13, 2013

Birds Are Dinosaurs

Birds are dinosaurs. At this point, that's not a very ground breaking statement, but I'm surprised by the number of people I talk to who don't know that already, or who balk at the idea.

So, let's get the basics out of the way first. Birds are definitely descended from dinosaurs. There may have been some question of this a few decades ago, but this is pretty much a settled question now. Just take a look at this figure from Donald Prothero's book, Evolution: What the Fossils Say and Why It Matters.

Non-avian Dinosaur & Bird Homology

Here's another similar diagram from Peter Wellnhofer's Archaeopteryx: The Icon of Evolution.

Comparison of Bambiraptor, Archaeopteryx, and a Modern Chicken

Both diagrams very clearly show just how similar early birds like archaeopteryx were to their theropod cousins. Here's another diagram I like from Wikipedia. It compares the hands of a deinonychus and an archaeopteryx. If you were just shown flash cards of one these at a time without the other to compare to, would you be able to tell which was which?

Deinonychus vs. Archaeopteryx Hand Comparison

And here's one more picture, also from Wikipedia, showing where birds fit into the dinosaur family tree. They're nestled right in among the theropods, in the Saurischian branch of dinosaurs.

Dinosaur Family Tree


So, that much, at least, seems to be settled. But many people still seem to have a problem with calling birds dinosaurs. For example, go read this posting on Yahoo! Answers, Are birds considered dinosaurs? (one of the answers there even linked to this website). Most of the answers there fell into the form of, birds are descended from dinosaurs, but not dinosaurs themselves.

Now, on one hand, I can appreciate this line of reasoning. After all, we are evolved from lobe-finned fish, and while I suppose that you could think of us as just very specialized fish, I don't think anybody but a cladist would actually call us fish.

So, the question becomes, are birds dinosaurs, or have they changed so much that they should no longer be considered dinosaurs? Or in technical terms, should dinosaurs be a monophyletic or paraphyletic term?

To answer this, let's leave dinosaurs behind for just a bit. Take a look at the pictures below, none of which are dinosaurs. (Click on any picture to go to the Wikimedia Commons source. Note that all have been touched up in some way, some more than others.)

Kangaroo Skeleton Horse Skeleton Baleen Whale Skeleton Dolphin Skeleton Fruit Bat Skeleton

That is a whole lot of variation. The first animal hops. The next walks on only one toe per foot, on a highly modified toenail only. After that is an animal that swims in the ocean and eats by using its mouth as a strainer. Next is another ocean dweller, but one that 'sees' through sound waves. And last, and perhaps most relevant to this discussion, is an animal that flies. Despite all this variation, these animals all share some common traits. For example, they all breathe air (even the ones that live in the ocean), and they all feed their offspring milk from a special organ in the females. So, these various animals all get grouped together as mammals.

Now, let's get back to dinosaurs. Here are a few more skeletons to compare. (Clicking on any of them will take you to the original source - some of these are copyrighted.)

Emu Skeleton Gallimumus Skeleton Eagle Skeleton Stegosaurus Skeleton (Outdated Reconstruction)

There's still a fair amount of variation here, but not as much as in the group of mammals above. There are three bipedal animals, one of which can fly. The real outlier is the stegosaurus in the bottom right (note that the image is an outdated reconstruction, but close enough for this discussion). Just to be clear, the two skeletons on the left are modern birds. The two skeletons on the right are non-avian dinosaurs.

So, considering how much variation there is among animals that are still all classified as mammals, and considering how much more similar the gallimimus above (top right) is to the emu and eagle than it is to the stegosaurus, I don't see how you can group the gallimimus and stegosaurus in one group, yet leave the birds out. I just don't see any reason to classify birds as anything but flying dinosaurs.

Friday, June 7, 2013

Where Would Newton Weigh a Newton?

Newton on a ScaleI got into a conversation about units the other day (yeah - some of my friends are as nerdy as me), and it got me to thinking about Newtons. They're named after the guy, but not based on him in any physical sense. But what is the physical relationship? Where would you have to go for Newton to weigh a Newton?

This is actually a pretty simple calculation. If you think back to your high school or college physics days, the force due to gravity is:

F = G*m1*m2/r^2

where G is the universal gravitational constant, m1 and m2 are the masses of the two objects attracting each other, and r is the distance between their centers of mass. G is known thanks to science, as are the masses and radii of various bodies in our solar system. That leaves just one more unknown - the mass of Sir Isaac Newton himself.

So just how big was Newton? To tell the truth, I doubt that anyone knows for sure. Doing a google search on "How much did Newton weigh" didn't yield anything concrete. But I did come across an interesting article on the blog, And Now You Know:

How tall was Isaac Newton? 5 feet 6 inches, perhaps shorter

So, the title of that article gives the answer in itself. Newton wasn't very tall by today's standards. In fact, he wasn't even very tall by the standards of his day. John Conduitt, who knew Newton personally and saw him on a regular basis (he was married to Newton's niece), described him as "he was short of a \middle/ stature & in \plump/ \in/ his later years inclining to be fat."

So assuming Newton was 5'-6" and on the 'plump' side, how much would he have weighed? Here's an interesting chart from the UK's NHS, Height/weight chart. For someone 5'-6", the middle range for 'overweight' is just over 12 stones (who actually weighs themselves in stones?). So, let's round that up to 12.5 stones, or 175 lbs.

Okay, so now we've got Newton's weight on Earth as 175 lbs, which is equivalent to 79.5 kg. With that in hand, let's go through one example calculation for Earth, just to double check that we're doing everything correctly.

F = G*m1*m2/r^2
F = (6.67e-11 m³/kg-s²) * (5.97e24 kg) * (79.5 kg) / (6,371,000 km)^2
F = 781.3 N

In normal units, that's 175.65 lbs - close enough to my original estimate once you account for rounding errors, so it looks like everything's being done correctly. But that means, on Earth, Newton would have weighed far more than a Newton. The moon's smaller. What about it? Well, once you go through the calculation for the Moon, it turns out to 29.06 N - still too much. Even the dwarf planet of Pluto has too strong of gravity. Below is a table showing various bodies in our solar system, and how much Newton would weigh on each one (I also included pounds for the people like me who don't have a good feel for Newtons). Just so you know, those last three bodies are moons of Saturn, and they're all more or less round.

Body Mass, kg Radius, m Newton's
Weight, N
Newton's
Weight, lbs
Earth 5.97E+24 6,371,000 781.34 175.65
Moon 7.35E+22 1,737,100 129.28 29.06
Pluto 1.31E+22 1,153,000 52.12 11.72
Enceladus 1.08E+20 252,100 9.02 2.03
Mimas 3.75E+19 198,200 5.07 1.14
Janus 1.9E+18 89,500 1.26 0.28

So, you have to get down to something as small as Janus, which only has a diameter of about 180 km, before Newton would weigh roughly a Newton. That's not a lot of force.

Image Source: Photoshopped from IGS.net and Wikimedia Commons

Archives

Selling Out