How do spiders know how to build webs?

You’re right, I don’t mean to trivalize the incredible capability of a machine as complex and intricate as even something as biologically simple as a spider. But once you have said organism, the instructions to build a weapon are, relatively speaking, pretty simple, requiring nothing that we would call awareness or cognition beyond a simple instruction set. The spider building a web is a particularly good example of this; while you note that it must “accomodate a huge range of possible problems, hindrances, prevailing conditions, situations, locations and events,” but truthfully it doesn’t accomdate these things except by incidental design.

A spider doesn’t understand that building a web in a doorway is a waste of valuable time and energy, nor will it learn when the web is destroyed not to build there again if it’s an otherwise ideal spot. The instructions for building a web are really quite simple, and the spider can be tricked into doing it wrong by altering conditions and cues in ways that the programming doesn’t allow for. In other words, spiders don’t exhibit awareness or judgement based upon experience; they build where, when, and how because that’s what the hardwired instructions tell them to do. Most spiders fail to reproduce and die (hence, why they are hatched by the hundreds); those who survive to reproduct statistically increase the fitness of the species by passing on the traits that permitted them an advantage.

Stranger

Spiders in general are interesting lessons in evolutionary specialization—particularly the web-builders. Those that I find most incredible are the ones that use their silk actively, as a weapon, to hunt for prey (i.e. Bola Spider, Net Throwing spider). I imagine their natural selection journey was even more complicated than the static web builders. (Also, of interest: The record holder for the highest-flying animal is the ballooning spider).

http://szgdocent.org/resource/ff/f-ssilk.htm

http://homepage2.nifty.com/singingsand/miwa/e_gossamer.html

I am not sure what makes you claim spiders are biologically simple.

I also find the claim that they don’t learn by experience remarkable. All you need to do is a web search on the jumping spider, Portia, such as here .
Incredible - and smart and anything but simple.

Lynne

I should declare a vested interest here and say that I am not only obsessed by spiders, but also writing a book on them, so all your answers are sending me in all sorts of directions. But I think it is ethical I tell you that.

That site was great, Schuyler, and I would be really pleased if you could remember the reference to funnel webs. What we call funnel-web spiders in Australia are ‘primitive’ spiders who live in burrows. They have silk trap lines coming out from their silk lined burrows, which led to the term ‘funnel’. This includes the Sydney funnel-web which is deadly.

Other countries call ‘modern’ spiders which build webs with funnels in them ‘funnel-webs’, for pretty logical reasons. We have them, too. I suspect it is the webs of the latter you mean and I am very keen to know if you can come up with a reference.

Lynne

Not completely related to the subject, but just an interesting website.Kythereia

http://spiderwebfarm.com/

This “Spider Web Farm” is just a little farm in a no-name town in Vermont. Basically during the Spring and Summer they lure the spiders into empty frames and get them to spin webs. Then they’ll go through several light spray painting and aerosolyzed gluing until they can pull a piece of wood through the frame and capture the web. I met this wonderful couple and they have a map of the world in their workshop that shows visitors from all over the world. And I have one of my own hanging up in my room.

This is why Vermont is my favorite state, it is covered, like little nooks and crannies, with all sort of little hidden gems like this.

I’d just like to add that I profoundly disagree (in a friendly way - this isn’t GD or the Pit), and I think your qualifiers (‘relatively speaking’ and ‘pretty’) are only there to hedge your bets.

I have no idea what the neurological ‘program’ (using the term loosely) for web-building behaviour looks like. But I would suggest it’s more complex, by many orders of magnitude, than any instruction set or computer program that we humans know how to write. I know plenty of men and women who can write the software for, say, a busy trading floor or a jumbo jet’s navigational system. I don’t know anyone who can write a program that you can run – inside any machine you care to imagine or construct – such that if you drop the machine at random on a bush, it will be able to work out how to build a web somewhere in that vicinity. And remember, this program has to work without any visual input beyond the extreme near field.

I know that the ‘program’ can be derailed (ie you can trick spiders into building webs that don’t make sense). But this only indicates that the program has limitations, not that it isn’t complex. Any piece of software can be derailed, no matter how complex, if you provide input that it wasn’t designed to handle. And whereas every piece of human software ever written is prone to ‘crashing’, such that the system locks and needs to be rebooted, I’ve never seen a spider simply ‘crash’ or lock into a static position such that it seems to need ‘re-booting’. Somehow, the program never actually seizes up. Even if you do your best to derail it.

Simple? I don’t think so. Not even ‘relatively’ so. Just my opinion.

First of all, I was speaking of the anatomical and neurological simplicity of spiders relative to other animals of orders Sauropsida (reptiles), Aves (birds), or Mammalia (mammals). Certainly spiders are complex than a single-celled organism, and vastly more complex and compact than any mechanism we could build to perform the same tasks, but they are a stone hammer to a rocket booster in comparison to a dog or parrot.

Second, the context was speaking in terms of web-building spiders; generally speaking, it is accepted that a hunting predator will be more sophisticated and aware than one which traps prey. It is inarguable that jumping spiders (family Salticidae) are the most complex and intelligent of all spiders; in comparison to even the most primitive rodent, however, it’s unclear (despite the linked pop science article which is short on detail and long on misapprehensions, such as spiders being like “other insects”) that it can be considered on par in terms of cognitive problem solving capability.

Third, while there’s no question that the biological machinery and neural sophistication of even a simple organism like a spider is vastly more reliable than a computer (which is understandable; Intel and AMD are going on a third decade of building microprocessors, whereas Nature has had a couple of billion years to refine brains), the complexity of a set of algorithms to replicate the behavior of web-building spiders is not that great, certainly not compared to a mouse or a moose.

Even assuming the kind of complex visual processing capability that the linked article posits for the jumping spider doesn’t translate directly into cognitive capability; many creatures have much more capable and and accute eyesight than primates, or even other complex and intelligent animals like the bear or octopus. “Knowing” (as contrasted with conditioned response to external stimuli) requires a degree of awareness both of the environment and one’s place in it, and a complex response to new stimuli (curiosity, caution, probing, trial and error, progressive learning). A jumping spider might exhibit these behaviors to a limited extent, but doubtless not to the same degree as a neurologically more complex creature which undergoes a long period of rearing and exploratory experience. And certainy web-building spiders display almost no degree of cognition or broad-scale comprehension beyond very simple operant conditioning.

Spiders know how to build webs not by trial and error (the way we learn how to do things) but because the conditions and cues that lead to the activities of web-building are coded genetically into what passes for their brains.

Stranger

I’ve seen a spider lock up into a static position from a re-booting. Actually it may have been the first booting that did it. Poor little spider :frowning:

Actually, the twin studies don’t mean that those are inherited, just that they are shared. To show they were inherited, you’d have to show that parents/children married people with the same names, etc.

w.

I still can’t agree. I have just spent two years writing a book on crocodilians (order Crocodylia) that is crocodiles, alligators, caimans and the gharial. I am now engrossed in my favourite animals - spiders - for my next book. I am not finding the physiology any simpler. The gospel on spider biology is Rainer F. Foelix’s Biology of Spiders, and for complexity in anatomy, these guys lack nothing! The web builders depend on very sensitive response to the vibrations of anything in the web to assess not only the location in the web, but the size and nature of the animal trapped, and then behave accordingly. If you start saying all this can be programmed, it can be argued that everything we do can be as well.

I have the same word length on the spider book as I did on crocodilians, and will struggle terribly with it - there is just so much more variety in spiders. That’s not to knock down the crocs - their heart and immune systems, among many other aspects, are extraordinary and more advanced than ours in some ways. But like us, they have nothing remotely like the spider families ability to create a huge range of silk types - a single species can create a variety of silks for different purposes. So comparing animal species for any single ability is worrying - you need to consider them in terms of the ecological niches for which they have evolved.

I struggle with this claim as well. Let me give a single example of an orb weaver. I have the article in front of me if you want the reference. Fifty Australian leaf curling spiders (Phonognatha graeffei) were taken into a lab - so out of their environment and where the ‘program’ was supposed to work. 33 managed to build a web, 17 didn’t. Programming? All were in an identical situation. They were placed in a frame and given a choice of leaves. 24 of the 33, chose green leaves which are much more suitable for creating their leaf retreats in the middle of their orbs. The other 9 chose dry leaves. The article then says: “Although the spiders usually selected a leaf for their retreat before completing construction of the web, the decision process appears to be ongoing: four spiders that initially selected dry, brown leaves subsequently removed them from the web in favour of moist, green leaves.”

This is in the lab with leaves placed on the base of the frame - nothing like their usual environment.

So how did these incredibly simple animals with no “knowing”, in your opinion, manage to build its web in totally unnatural environment, and then display a decision making process adapting to the situation in hand - including the trial and error you dismiss as being possible for spiders?

I am not arguing that these are the most intellectually capable creatures, but simple they are not. They are far from studied to the degree other animals are. In fact, it is estimated thet the 40,000 or so known species may be less than half of those on earth. We haven’t even classified them all yet, let alone studied their cognitive abilities. In fact, the experts I talked to when investigating crocodilians all acknowledged they were far from intelligent - in the way we humans define intelligence. Armchair experts on crocs tend to claim they are intelligent, those who work with them don’t. They don’t really need great cognitive abilities in the ecological niche they have so successfully evolved to fill.

As do many animal species, including the reptiles you quote as so vastly superior. Crocodiles don’t need great intelligence in their lofty position at the top of the food chain - when they get there. 99% of saltwater crocodiles don’t make it to their first birthday - and just like the spiders, they fail to reproduce and die. (Reference: my own book! Oh, and the the references quoted in it.)

I better get back to studying the physiology of spiders - it’s complex stuff!

Lynne

Spiders lack a complex respiratory system, digestive tract, or muscular arrangement. (Spiders have an “open” circulartory system, in which haemolymph is pumped by a primitive, single chamber “heart” into sinus cavities around their organs.) This is not to say that their anatomical features and arrangement are not in their own way ingenious, but they lack the complexity of other animals such as mammals. They are certainly more complex than flatworms, or indeed, many insects, but not as complex (certainly not in the neurological sense) as animals with which we observe indications of higher cognition and problem solving.

In general, you seem to be repeatedly misstating the context, as if I’m making spiders to be no more complicated than a child’s wind-up toy. There’s no question that even a primitive spider species is more capable than anything we can build. But their behavior is quite simple, and response to stimuli very predictable. There is, of course, a wide variety and range of complexity in the order of Araneae, but even the most complex of them fail to display the kind of learned behavior one would expect from a dog. Evolutionary fitness doesn’t enter into it, either; while the adaptability of higher cognition has its advantages, it’s clearly not the single path toward reproductive success; bacteria with only a rudimentary ability to respond to stimuli and nothing anyone would consider a brain are highly successful. Whether jumping spiders are smart or dumb is irrelevent with regard to evolutionary success; what matters is how effectively they are able to thrive and reproduce in the environment they occupy.

Coming back to the OP’s question, again, it’s clear that the knowledge that a “spider” as in building a web is innate; it’s not passed on by parents, nor (given that they can and have to immediately start making webs) a product of trial and error. This doesn’t mean that they can’t be intelligent–the hunting capability of an octopus or squid is also, at least initially, an built-in quality (despite the demonstrated intelligence of these animals, their parents are dead or dying by the time they are born) but the ability to construct even complex webs is not a reflection on their intelligence any more than the olfactory capability of canines makes them smarter than humans. Utility is not intelligence.

Stranger

These are two spiders who I would long to see and have never managed to observe. Yet! In terms of Stranger’s argument that spiders are such simple creatures, these raise a few issues.

The Australian version is called a bolas spider, and like the Bola Spider you linked to, gets a sticky piece of silk, loaded with sticky beads and waits for the moths it feeds on to approach. It then swings the ‘bolas’ and clobbers the moth, which is much bigger than itself. It battles it, much like a fisherman with a fish on a line, until the moth is exhausted, when the spider hauls it in and kills it. The net-casting spider creates an elastic web which it holds on its first pair of legs, like a net, to wait for an insect to pass below which suits their size and abilities. It then throws its net over the prey and the battle begins.

Do these examples sound like using a tool to you?

Isn’t using a tool considered one of the signs of advanced cognitive behaviour?

Stranger wrote:

Are these the more sophisticated hunters or less sophisticated trappers? Who ‘accepts’ that hunting predators (generally speaking) are more sophisticated and aware than trapping ones?

I become more amazed by spiders with every word I read!

Lynne

Yes and no. The spider isn’t creating tools out of objects found in the environment, and “tool” use (particularly tools formed by one’s own body) isn’t the sine qua non of intelligence. Nor is this an example of learning and propogating tool use memetically, i.e. by example, passed from generation to generation. We can (and I have) programmed CNC milling machines and welding robots to perform highly complex operations with extensive feedback control. This doesn’t mean that the machine is intelligent.

There’s no question that spiders are well-adapted to what they do, and that many of those adaptations and capabilties are astonishing and amazing in their effectiveness and apparent complexity (i.e. elaborate web-building, or hunting behavior). But being able to do something well does not translate into displaying higher congnitive function. Many viruses attack cells, hijack their normal functions to reproduce, trick the immune system into short circuiting or attacking the body, and generally do all sorts of clever things, and not only are they clearly not intelligent, they aren’t even alive by any rational standard. Spiders are well-adapted; but that doesn’t translate into sophistication in any sense that we would consider general intelligence.

Stranger

The debate over spider intelligence reminds me of a passage in The Hitchhiker’s Guide to the Galaxy. (Talking the original BBC radio program here; whether the same exchange occurs in the book I can’t recall; also, I’m paraphrasing.) As the gang is fleeing the Restaurant at the End of the Universe, Marvin remarks that he can see the ultimate question to the ultimate answer (42!) engrained on Arthur Dent’s brain. Arthur practically squeaks, “You can see into my brain?” Marvin, “yes.” Arthur, “What do you see?” Marvin, “It astonishes me that you can live in anything so small.” In other words, I think Stranger’s fundamental point is sound. (Don’t agree about the programming point, for the reasons argued by Ianzin.) Intelligence is a continuum ranging from bacteria through hyper-intelligent pan-gallactic beings. The fact that each creature on that continuum is, in it’s own way, a marvel of nature, does not make them equivalent. Does anyone really dispute this?

As for the OP, I believe the complete, true and correct answer is “We don’t know.” But it sure is pretty amazing.

Oops, that should be “hyper-intelligent pan-dimensional beings.” Sorry, Frankie and Benjy mouse.

I was being facetious with the tool claim, and I agree substantially with the concusions of Pbear42. I suspect Pbear42’s pan-gallactic slip was due to one too many gargle blasters.

I have found this discussion invaluable at this early stage of researching a book, and will ask more spider questions in my own thread rather than continuing to derail Ianzin’s. As a specialist in gifted education, having sat through debates on the nature of intelligence for decades now, I know we can go around in circles. I am much more interested in the original OP. I also supsect we just don’t know, but I find it astounding and would love to know how they do it. I’ve watched it often and never been less than overawed.

Lynne

I just wanted to thank Stranger On A Train for all his (or her) contributions to this thread. While I may disagree here and there, I’ve found all SOAT’s points well-argued and fact-based, and delightfully free of any personal ‘touches’ that sometimes tend to clog up GQ answers (and yes, I’m as guilty as anyone). When I grow up, I want to respond to GQs like SOAT does.

I found a couple of things, although 2 caveats: 1) I am not a biologist, so these searches aren’t in my area of expertise, and 2) this is by no means meant to be exhaustive.

Stuff I did find:

Concerning the relative energetics between orb webs and sheet-funnel webs - the sheet-funnel webs are more costly to contstruct, which means the spider is less-likely to relocate their web site. However, the sheet-funnel webs, being in most cases non-adhesive, tend to last longer with lower recurring maintenance costs:

Energetic cost of web construction and its effect on web relocation in the web-building spider Agelena limbata

It appears, if I interpret correctly what is said, that 2-D orb webs evolved first, with the 3-D sheet-funnel webs beings a later development. Also, the sheet-funnel webs appear to serve a defensive purpose with respect to predatation by mud-dauber wasps, while the orb webs leave the spiders relatively exposed.

Are three-dimensional spider webs defensive adaptations?

Also, apparently the zebra tarantula secretes silk (or a silk-like material) from its feet, leaving open the question about whether silk evolved initially for prey-capture or whether its first function was to protect against falls.

Biomaterials: Silk-like secretion from tarantula feet

Finally, I think that perhaps this discussion is going at cross-purposes. When it is said that web-building is relatively simple and/or built on relatively simple rules, this is not meant to imply that the mechanics of implementing these instructions is simple, just that a description of the high-level steps is fairly straightforward. Lemur866, Colibri, and Stranger all offered some examples of steps in a web-building procedure - they are not arguing that the spider’s execution of these steps is simple. For example: “going to the grocery” is a simple instruction, but the execution is complex if we break it down to the individual muscle motions required to balance the body, move the legs, swivel the eyes to track traffic, step on/off curbs, etc.

Schuyler, you’ve nicely summed that up. The mechanical complexity of a spider, and the controls that allow it to perform such nuanced actions are nothing short of miraculous; and when you look at it in a vacuum, it seems almost incomprehensible that such complexity could arise naturally. Advocates of Intelligent Design seem to have a very strong case, until you look at the evolutionary development of such mechanisms. But this doesn’t correspond to what we’d consider higher cognition; indeed, much of the control over spider movement is actually local to the limb, which have nerve ganglia that essentialy act as semi-autonomous regulators for movement and response. (This can be seen in other animals as well; cephalopods, for instance, have a highly distributed nervous network, despite their largish central brains.)

The jumping spiders may be more intelligent as a class than other spiders, and may even demonstrate something resembling curiosity and awareness of their environment beyond instinctual impulse and response; given the diversity of the order, I wouldn’t be surprised if there’s a broader range of intelligence in spiders than in primates. But the web-building behavior of spiders, while sometimes resulting in elaborate, beautiful webs, is not really all that complex in high level decision making.

Intelligence is not the primary or even preferred route toward evolutionary success, though; the vast majority of animals survive not by being smart, but by being otherwise well adapted and adaptable. Spiders can be amazing, even if they can’t quote Shakespeare.

Stranger

Schuyler and **Stranger **- thank you so much. I love being pushed to think differently and greatly appreciate what you and others have said. I guess I really did overdo my cure for arachnophobia and have become so spider obsessed as to be almost as irraitonal as I was fearing them! And thank you Ianzin.

Thank you!