Classical and Operant Conditioning in Reptiles

[TylerDurden]

So in an earlier thread there was a debate going on (that seems to have gotten pretty heated from what I read) on whether geckos like to be handled.

A discussion came up about operant conditioning. A few of us (including myself) suggested it wasn't possible in reptiles because they lack a lot of the reward centers of mammalian brains

However, there is research demonstrating operant and classical conditioning in reptiles

Here's just one source but upon further reading I'm finding more as I go along

http://www.ncbi.nlm.nih.gov/pubmed/12795857

This means reptiles can be trained. The capacity of each species for operant and classical conditioning, and the complexity of stimuli they can "learn" I'm still not clear on, but I bring this up because I'm definitely already seeing classical conditioning in my own gecko.

I posted this video in another thread but I figured I'd post it here too since it's relevant.

http://www.youtube.com/watch?v=UhXYGZoi1MA

Whenever I feed her I do it using forceps. I've noticed whenever I lift up the forceps she now runs out to me and waits for food. This doesn't occur with any other object.

To make sure it wasn't just the forceps being shiny and moving that was attracting her, I used a laser and a blinking light that she will also follow. However, when the stimuli of lights are removed, she walks away. When the forceps are removed, she stands at attention waiting for food.


I think this could definitely be relevant to handling, and theoretically it may be possible to get your gecko to do some very very simple things like stand in a certain area of the tank when you put your hand in.

If anyone has any other sources on operant or classical conditioning in reptiles, especially any peer reviewed journals or secondary sources that cite one, it would be much appreciated! I may experiment a little with operant conditioning, to see if I could get her to stand on a certain spot of the tank. Before I even bother putting in any effort in that though I'd want a few more sources demonstrating it is actually possible to do (and even if it is I'd imagine it is difficult since they lack the "working memory" of mammals, and can't really plan out tasks, they can only react. This means operant conditioning would have to be extremely simple and clear-cut. Any ambiguity between stimulus, action and reward would likely make this impossible).

 

[reps4life]

Actually I tried different things as well with Savannah Monitors. They would always run up to the glass when I came near with a yellow container. So I wanted to know if they could see yellow so I found another container with the same shape, but was red. I came close to the glass and they did not react.

I would also feed in a certain section which I would block off due to the substrate and everyday at the same time they would walk over and stand in that section waiting.

If I was home they would do their normal activities and go to sleep by 8pm. I would always go downstairs and check to make sure lights when out and everything was fine. On the few occasions I was away from home and arrived after the lights went out, I would find them awake in the dark laying their bodies on the glass looking. How strange is that.

 

[TylerDurden]

wow that's exactly what I was hoping for lol I'm definitely going to try this out then. Maybe get her to stand in a certain spot for food, that would make picking her up and feeding a lot easier

 

[M_surinamensis]

I'd imagine it is difficult since they lack the "working memory" of mammals, and can't really plan out tasks, they can only react. This means operant conditioning would have to be extremely simple and clear-cut. Any ambiguity between stimulus, action and reward would likely make this impossible.

That's really the portion of it that many people fail to grasp. You can manipulate them over time to form some extremely basic associations between pre-existing instinctive behaviors and new stimuli. You will find that anything involving the development of new behaviors, or where the association is not direct and concrete will fail to materialize in any kind of reliable fashion.

So pattern recognition associated with a feeding trigger (as an example) is a possibility.

Responding to verbal commands and being rewarded with praise or petting is not (despite some anecdotes shared by people who are actively imbuing their perceptions with their own desires for interaction).

The manifestation of genuinely new behaviors is also off the table as a possibility.

 

[gecko4245]

Is it because the size of there brain? will a larger reptile be smarter than a smaller one?

 

[Desdemona]

I am not an expert.. but from my limited knowledge and based off of some experiments with other species (which may not be relevant).. from a evolutionary standpoint it makes sense that they would have some level of basic behavioral conditioning.. it's what keeps you alive in nature. If you eat a bug that makes you sick, you avoid that bug in the future or you die.

I think I was the first to bring up the topic, but I never said petting or praise, that was assumed or added. I was talking about simple cause and effect.. give the creature a reason that it understands to continue the behavior (it understands food) and over time it will. Of course if the animal has a short memory it will quickly forget if not constantly reinforced and repeated. I have no clue what the memory span of a reptile is. It probably varies by species, some are more simplistic than others.

If you read about experiments regarding animal intelligence, learning ability, etc.. you will find that we are constantly discovering that animals have abilities we never before considered or assumed they did not have. We just have to find the right way to trigger or make it apparent to us (since obviously there are communication barriers). Of course some animals with simplier brains will never (unless they evolve) develop certain abilities, such as complex communication. Animals also evolve so who knows, 20 -40 years from now some domesticated reptiles could start showing more social skills. Other animals who have been long breed in captivity have developed animal to human social skills that their wild counterparts do not have. However, these so far (as far as I know) have all been mammals who have a leg up on reptiles in that catagory.. so it may never happen. Only time will tell.

 

[TylerDurden]

Is it because the size of there brain? will a larger reptile be smarter than a smaller one?

no it has to do with the anatomical structures of the brain. Mammals have a cortex with structures that enable higher processing such as planning, "cognitive map" making, facial recognition, etc.

This is why you can train a mouse to go through a maze for instance, or train a dog to do tricks (some that have multiple steps). It's why chimps can do complex tasks

Humans also have the frontal lobes which are really unique and are pretty much responsible for our ability to perform complex tasks, make complex decisions, etc

Reptiles lack these structures. Their brains are very simple. They have what is sometimes referred to as the "old" brain, or "lesser" brain which is referring to the fact that it evolved long before the mammalian brain and doesn't have all the new bells and whistles that come with it.

So reptiles are limited to very basic functions. They cannot plan, they can only react. They can't make complex associations, they can only be conditioned to know "this stimulus leads to this response" or "this stimulus is safe, this one is not". They're pretty much completely instinctual, meaning they don't really learn beyond what is "hard-wired" in their brain, such as food-seeking (they seem to easily pick up what is associated with food for instance), and their "emotions" are likely limited to simple fear vs safety and hunger vs satiety

The size of the reptile is not going to change that because it's an anatomical issue, their brains simply lack the structures needed for all those other functions. Brain size is not likely to have much of an effect either although I'm not sure if there are spatial/temporal relationships in their brains for working (aka task) memory. What I mean by that is, it may be possible that a larger brain could make them easier to condition and more likely to pick up simple associations, but beyond that there won't be much difference.

 

[TylerDurden]

a simpler analogy would be:

Say reptile brains are regular phones and mammal brains are smart phones.

You can make the regular phone smaller or bigger, or even make it faster, but it still will never have the ability to go on the internet and check your e-mail, watch videos and take photos.

 

[TylerDurden]

I am not an expert.. but from my limited knowledge and based off of some experiments with other species (which may not be relevant).. from a evolutionary standpoint it makes sense that they would have some level of basic behavioral conditioning.. it's what keeps you alive in nature. If you eat a bug that makes you sick, you avoid that bug in the future or you die.

I think I was the first to bring up the topic, but I never said petting or praise, that was assumed or added. I was talking about simple cause and effect.. give the creature a reason that it understands to continue the behavior (it understands food) and over time it will. Of course if the animal has a short memory it will quickly forget if not constantly reinforced and repeated. I have no clue what the memory span of a reptile is. It probably varies by species, some are more simplistic than others.

If you read about experiments regarding animal intelligence, learning ability, etc.. you will find that we are constantly discovering that animals have abilities we never before considered or assumed they did not have. We just have to find the right way to trigger or make it apparent to us (since obviously there are communication barriers). Of course some animals with simplier brains will never (unless they evolve) develop certain abilities, such as complex communication. Animals also evolve so who knows, 20 -40 years from now some domesticated reptiles could start showing more social skills. Other animals who have been long breed in captivity have developed animal to human social skills that their wild counterparts do not have. However, these so far (as far as I know) have all been mammals who have a leg up on reptiles in that catagory.. so it may never happen. Only time will tell.

you're correct and I agree with most of what you're saying but I wouldn't hold out on domesticated reptiles developing some sort of ability to interact with humans beyond their current levels.

What you're referring to is true - many domesticated animals over the centuries have developed the ability to interact with humans in ways no other animals could, and develop complex social skills they didn't previously have

but those animals were mammals.

Again it becomes a matter of anatomical structure. Reptiles are not going to evolve a new cortex in our lifetime, that's something that would take a few thousand to a million years (look at how long reptiles roamed the earth before the first mammals arrived).

Using the analogy of electronics again:

The reptile brain is like a watch, and the mammalian brain is like a computer. They both have the basic functions to tell time, but the watch is limited in its capacity, there are many things it can't ever do unless many many changes are made to make it more like a computer. The computer already has the ability to do things like run software, so changes can occur much faster, because it already has that ability and can build on it.

Mammals already HAVE brains with the anatomical structures for more complex integration of information, and they have structures for reward centers, planning, etc. So selective breeding can alter all of these things at once, and even give rise to some new functions (dogs are the only known animal to read human facial expressions, the ability does not exist in wolves, it arose from domestication breeding over centuries)

 

[M_surinamensis]

I am not an expert.. but from my limited knowledge and based off of some experiments with other species (which may not be relevant).. from a evolutionary standpoint it makes sense that they would have some level of basic behavioral conditioning.. it's what keeps you alive in nature. If you eat a bug that makes you sick, you avoid that bug in the future or you die.

That's an incorrect assumption. You are assuming the animal is capable of drawing a connection between two events; eating a specific bug and then feeling sick. Those events are displaced in time, in space and in association. Making the connection requires a species to be able to understand the abstract, to associate the action with the reaction despite no concrete and immediate identifiers tying them together.

Their behaviors evolve between generations, because the ones who eat poisonous bugs do die, removing their genetic material and all the associated instincts, from the gene pool. Adaptations to or reinforcement of behaviors doesn't happen beyond the most basic level inside individuals.

If you read about experiments regarding animal intelligence...

... if? Do you really want to whip 'em out and start measuring... references... in a conversational forum thread?

Animals also evolve so who knows, 20 -40 years from now some domesticated reptiles could start showing more social skills.

You may be slightly unclear on the number of generations it usually takes for natural selection to produce measurable results. Even artificial selection for characteristics such as the behavioral functions of reptiles; they have very simple behaviors as far as animals go, but that doesn't mean that it's easy to analyze and manipulate as a breeding goal in a short time frame. It is not the same as just sticking together two animals with the best color and compatible genetic anomalies.

 

[OnlineGeckos]

I think this could definitely be relevant to handling, and theoretically it may be possible to get your gecko to do some very very simple things like stand in a certain area of the tank when you put your hand in.

I have 2 geckos that know where I'm going to reach my hand in to pickup the water bowl to clean. Whenever I open their tanks, they would head over to that side of the tank and stand up on their hind legs looking to climb up onto my hands. I didn't do anything specific to train them that way, but those 2 just started doing it on their own.

Interesting creatures, they are definitely not smart at all. But some of their behaviors do surprise me and make me smile

 

[Desdemona]

There are random mutations that occur in not so small incriments.. generally they are useless and die, but they do happen. The chances of a reptile being born with a cortex-pre cursor.. the trait even being recognized.. and of course passed on.. are highly improbable. It would probably have to be a gecko in a lab to start out with. But, its not completely impossible. I would not bet money on it, but I would not be that astonished either if it did happen. Some species have shown a lot of mutation in short periods of time. Chiclids are, from what I know, the best known species for this, they will modify behavior and physical apparence in only a few short generations, especially in a lab. They live in a tough envrionment though (in nature), so its a adatablity trait.

I find both psychology and evolution fascinating... but my reading has been all over the map on the evolution subject and of course my psychology training has been in humans. I have read a few, "wow this animal showed abilities we thought they would never have" type of studies (and not by fruity PETA types..actual studies) and "we started X evolution in sepcies and everything else went with it and in only a few generations we had almost a completely different animal (not genetically just behavioral and visual). From what little I've read on reptiles and line-breeding reptiles are probably more immune to quicker evolution than mammals and fish. I am not saying it will happen by any means, but I don't think its 100% impossible. Gecko acting like a dog in 20 years.. yes that is impossible.. but a single behavior that makes them stress less with human contact maybe. So far my enigmas have shown a better ability to adapt to human changes than my "regular" Leos.. but I am only compairing two animals to three and its by no means a rule for the species.

 

[TylerDurden]

There are random mutations that occur in not so small incriments.. generally they are useless and die, but they do happen. The chances of a reptile being born with a cortex-pre cursor.. the trait even being recognized.. and of course passed on.. are highly improbable. It would probably have to be a gecko in a lab to start out with. But, its not completely impossible. I would not bet money on it, but I would not be that astonished either if it did happen. Some species have shown a lot of mutation in short periods of time. Chiclids are, from what I know, the best known species for this, they will modify behavior and physical apparence in only a few short generations, especially in a lab. They live in a tough envrionment though (in nature), so its a adatablity trait.

I find both psychology and evolution fascinating... but my reading has been all over the map on the evolution subject and of course my psychology training has been in humans. I have read a few, "wow this animal showed abilities we thought they would never have" type of studies (and not by fruity PETA types..actual studies) and "we started X evolution in sepcies and everything else went with it and in only a few generations we had almost a completely different animal (not genetically just behavioral and visual). From what little I've read on reptiles and line-breeding reptiles are probably more immune to quicker evolution than mammals and fish. I am not saying it will happen by any means, but I don't think its 100% impossible. Gecko acting like a dog in 20 years.. yes that is impossible.. but a single behavior that makes them stress less with human contact maybe. So far my enigmas have shown a better ability to adapt to human changes than my "regular" Leos.. but I am only compairing two animals to three and its by no means a rule for the species.

well geckos acting less stressed around humans wouldn't have to require new behavioral evolution. If geckos that are generally less stressed are continuously selected for and bred, yes in a few decades we could have geckos that just in general have a suppressed stressed response.

also you're on the right track with reptiles being less likely to evolve rapidly

Bacteria can evolve in a few days time. We see it every day with acquired drug resistance. They have mechanisms that really speed up the process, such as horizontal gene transfer, and bacteria don't have advanced genetic repair mechanisms

As you go higher up the evolutionary heirarchy, you see more and more repair mechanisms preventing autosomal mutations as well as mechanisms preventing germline mutations. It's advantageous because traits that are "good" are conserved throughout evolution

So reptiles are not going to evolve massively new characteristics in short timespans like bacteria can (we literally see new proteins every decade or so, some being virulent factors and some being drug resistance factors).

A realistic prediction would be reptiles could be selectively bred to be less stressed and better adapted for captivity (reptiles that are over-stressed or not well adapted for captivity are less likely to make it to breeding. Also if breeders wait until reptiles are slightly older before breeding it increases the environmental pressures for selection, as survival curves obviously decrease over time)

Evolving a cortex is really an unrealistic prediction as it took tens of millions of years for that to happen (dinosaurs were around for millions of years before even birds began appearing, likely branching from archaeopteryx or a similar transitional species of feathered reptiles). I don't know if you realize how many actual genes are involved in creating the cortex. It isn't like one generation had a cortex out of nowhere, we're talking many many small changes over millions of years with very harsh selective pressures. In captivity animals have veterinary medicine, no competition of resources, etc. So evolutionary pressures are really limited to a very small subset of factors.

So yes I could definitely see selective breeding for lower stress responses happening, as you're modifying an existing response. I could not see any circumstances which would possibly warrant expecting a major anatomical change to their brain in any smaller period than a few thousand years at least, and it would be more likely to be seen in the wild in an isolated population than in captivity.

 

[chrisherp]

So in an earlier thread there was a debate going on (that seems to have gotten pretty heated from what I read) on whether geckos like to be handled.

A discussion came up about operant conditioning. A few of us (including myself) suggested it wasn't possible in reptiles because they lack a lot of the reward centers of mammalian brains

However, there is research demonstrating operant and classical conditioning in reptiles

Here's just one source but upon further reading I'm finding more as I go along

http://www.ncbi.nlm.nih.gov/pubmed/12795857

This means reptiles can be trained. The capacity of each species for operant and classical conditioning, and the complexity of stimuli they can "learn" I'm still not clear on, but I bring this up because I'm definitely already seeing classical conditioning in my own gecko.

I posted this video in another thread but I figured I'd post it here too since it's relevant.

http://www.youtube.com/watch?v=UhXYGZoi1MA

Whenever I feed her I do it using forceps. I've noticed whenever I lift up the forceps she now runs out to me and waits for food. This doesn't occur with any other object.

To make sure it wasn't just the forceps being shiny and moving that was attracting her, I used a laser and a blinking light that she will also follow. However, when the stimuli of lights are removed, she walks away. When the forceps are removed, she stands at attention waiting for food.


I think this could definitely be relevant to handling, and theoretically it may be possible to get your gecko to do some very very simple things like stand in a certain area of the tank when you put your hand in.

If anyone has any other sources on operant or classical conditioning in reptiles, especially any peer reviewed journals or secondary sources that cite one, it would be much appreciated! I may experiment a little with operant conditioning, to see if I could get her to stand on a certain spot of the tank. Before I even bother putting in any effort in that though I'd want a few more sources demonstrating it is actually possible to do (and even if it is I'd imagine it is difficult since they lack the "working memory" of mammals, and can't really plan out tasks, they can only react. This means operant conditioning would have to be extremely simple and clear-cut. Any ambiguity between stimulus, action and reward would likely make this impossible).

Would you say they have the ability to recognize their owner?

 

[endrien]

Would you say they have the ability to recognize their owner?

They have no perception of "Owner", they just see where they get food from.

 

[iLEO]

I've noticed this as well with my geckos. When I feed them I throw their food into a container with Repashy Calcium and shake it. They seem to have associated the shaking sound as "It's time to be fed" because now when I do this they all come running to the front of their homes. Even if they don't see me.

 

[chrisherp]

They have no perception of "Owner", they just see where they get food from.

well that's a bummer

 

[Desdemona]

I wasn't expecting an entire cortex in one generation.. just a small start of what could turn into one over time. Some animals in captivity have shown major evolutionary change.. though the cases I have read about were fish and mammals.. and not major brain changes like that.

I was thinking about some humans who are born with their brains wired wrong, giving them unusual abilities.. and normally social difficulties. Due to their social difficulties, in a social environment, the few I have read about never married or had kids. I think the general consensus is that it was a gentically normal brain that suffered some trauma in birth, however. If there was a person born with that brain anomalie and it was genetic they could potentinally pass it to their children and develop a population of people with special mental abilities (IQ in general as seen as being genetic to some extent). Of course this is really just a enhancement on a ability that is already there, ie most humans can do math to some extent and some can just do complicated math problems in their head instantly.

My expertise is not biology so I am just theorizing on limited knowledge I've gained in my free time. I think it would be a little assumptous to assume that brain development could "ONLY" happen in a few million years just because that is how it happened previously. Look at the changse in the human race compaired to others... dolphins might be half as smart as us, but we have out evolved the rest by far. Young kids of this generation seem to be better adapted at learning than kids of just a few generations ago (though maybe they were just limited by the assumption that they could only learn a limited amount). Even the current "your kid should meet these milestones" seems to be behind what I've seen in actual kids (my son has gone to a few daycares since I've moved since he was born). My son's Pedi thought he was accelorated, but quite a few kids at his daycare was at the same level as him. It was a pretty average center too, no special training to make them smarter.. or whatever.

But, of course it still comes down to compairing reptiles, who seem to enjoy holding onto their ancient genes... and mammals who would rather not (canine seem to evolve pretty quickly as well). I am not expecting any great reptile feats to actually happen. I don't expect a fully developed cortex to develop in one while I'm still alive. I just have a tendency to try and play devil's advocat against any "NEVER" statement... ie reptiles will never want to be handled by humans.

 

[gecko4245]

well that's a bummer

Yep, that's the reality so if you can't live with it then it means you want conditional love so buy a dog instead.

They don't need a friend to play with. they don't love you or miss you emotionally.

No offense to you chrisherp but some people live in a delusional world when it comes to reptiles to please themselves. :main_thumbsdown:

 

[Tammy24]

I bet you can train them using operant conditioning, especially with the use of a conditioned reinforcer (clicker training). I think you can train them to go on your hand "on command", not a verbal command necessarily but just your hand laying flat being a cue to go on it.

Karen Pryor has trained Oscar fish to go through hoops when they are placed in the water, their "clicker" was a pen light being flashed. She trained a marine hermit crab to yank on a string, the "clicker" was her slapping the top of the water. The reward in both cases was food. I bet you could train a Leo that approaching the hand leads to food, and then fine tune it to BEING on the hand leads to food. If fish and hermit crabs can do it why not reptiles.


Here's a good book to check out (where I got those examples from):
http://reachingtheanimalmind.com/