An Experiment of Rat Proportions!

Experiment Diagram

Hypothesis: Sprague-Dawley rats will demonstrate learned behavior after being exposed socially to the behavior of another pre-conditioned Sprague-Dawley rat.

Method:

Apparatus: A wooden Box (76.2 cm x 30.5 cm x 45.7 cm) will be used. Inside a small box approximately the height of two baking soda boxes stacked on top of each other (5.1 cm x 7.6 cm) for the rats to stand on. A standard round bell on a red ribbon will be suspended from an eye hook screwed into the side of the wooden box at approximately (2.5 cm) from the top of the box. Standardized 45 mg rat pellets will be used as a reinforcer.

Procedure:
Part 1-Rat A (Tinkerbell) will be classically conditioned to climb the boxes and move the bell hard enough to make a ringing sound and will then be reinforced with rat pellets.
Part 2- A separate baseline over the course of three days will be established based on how many times Tinkerbell climbs on the box and rings the bell within an hour. A baseline will also be found for Rat B (Rooth) and Rat C (Ambrosia), based on the number of times they successfully climb on the box and ring the bell over the course of an hour. This baseline will be taken with each rat in the box separately.
Part 3- Tinkerbell and Rooth will be placed in the box at the same time. Tinkerbell will be free to climb on the box and ring the bell at which point both rats will be reinforced. Rooth will not be reinforced if she rings the bell on her own at this point. The procedure given will be repeated after Rooth is removed and Ambrosia is put in with Tinkerbell.
Part 4- Rooth will be placed by herself in the holding box where the number of times she climbs on the box and rings the bell (and is therefore reinforced) is recorded for an hour per day over the course of three days. This same procedure will be repeated for Ambrosia after she is placed within the box by herself. Afterwards we will take a ratio of the number of successful bell rings at baseline and after they have been subjected to the experiment.

It’s Generalization Mr. Watson And Learning in Practice

So today I thought I’d try something different. I kept Rooth in the lab and put her in the big wooden box to see if she would generalize her trick to the box. At first she was confused and wouldn’t do any behavior that would have earned her a click in the aquarium so I gave her time to settle down then clicked when she stood up anywhere in the box. After a few repetitions with the clicker (which noticeably different in sound quality from in the aquarium even to me, especially with the loudness of the lab) she started offering more standing behavior. A click and treat or two after that and she was starting to stand up in the same general area as she would have in the aquarium if I had said ‘stand up’ when I did say ‘stand up’. I think today’s trial was a success.

Also, while horseback riding today it suddenly clicked to me how what we’re doing in class applies to other animals and other parts of life. I was riding a horse named Dice in the same field as his fieldmates and he started to pull on the reins and angle his body so that he was moving back towards another one of the horses. I pulled on the reins steady and my riding instructor reminded me to give him a quick pop instead of gradually increasing the pressure on the bit as he was continuing to move towards the other horse. One quick pop and another as we came around to the same spot the time after that and he no longer moved his body towards that other horse when we passed that direction. Of course this is a technique that I’ve used for a long time in riding, but it never made as much sense as today after putting it together with the ideas we’ve been learning in class.

One More Video

Here are is a video of Rooth doing her trick. After two days off to socialize her with Tinkerbell and Ambrosia I put her in and went to give her a few tries to see how she’d do before we started taping. All of a sudden she was going to the correct spot and rearing up with one ‘Stand up’. It was interesting to see, especially since I had figured that her schedule would be all messed up after two days off.

Tricks of the Trade

Well we’re definitely making progress to the point where I started to add a voice command to Rooth standing up. Dr. MacEwan suggested that I say something repeatedly and only reward her when she stands up when I’m talking. When I’m not talking she doesn’t get a click and a treat. This seems to be working.

In other news, my group and I are trying to do a social experiment with our three rats so Joni got a washable marker and marked Tinkerbell and we put her and Rooth together in the big box to see how they would do. I had the idea of putting in their cages and putting things on top to give them a chance to smell each other without touching each other. That worked for about two minutes until they both figured out how to jump out. For a while they did stop rolling around and interacting with each other so roughly at least while they moved around the cage and sniffed at the water bottles and the cages. Rooth was sniffing and getting into things more than Tinkerbell was.

We’re making a little progress I think, but this isn’t going to be a quick process I don’t think. I’m going to continue to think on how to get them to interact and get used to each other without any of them getting hurt or avoiding the others.

Correct Video

Here is the correct video showing the very beginning of me training Rooth by clicking when she stands.

Adding My Voice

So I more or less had Rooth doing what she was supposed to. The trouble was…she does it so often I wanted to make it a little more difficult for her and put it on a voice command. I know that Joni started using her voice from the very beginning. I didn’t want to do that because I wasn’t sure that it wouldn’t just confuse Rooth. Of course the joke is on me. She was doing a lot of the jumping around recently and rubbing against the corners when she doesn’t get the food and she expects it.

I think she’s finally getting the idea, but it’s hard to tell because she offers the behavior I want so often anyway. The method I’m using is to try to give the command before she rears up or at least before she gets there all the way. At least I know that she hears the command as at first, after a while, she would turn around and look for food right after I said the command. After that I made her stand longer before I clicked.

I’m not even sure that I’m going about it right to add a voice command to ask for what I want I wanted her to do. Has anyone else decided to add a voice command at the end? Any tips?

Training Video

Here is a post of  me training Rooth to do her trick.

Rooth Training Procedure And Diagram

Here is my diagram of how I am training Rooth. I am starting with only clicking when she rears up on her hind legs at all without touching the sides of the aquarium between markers one and two. Then as she gets better she has to rear up between markers one and two and marker three. Afterwards the area will be narrowed to markers two three and four until eventually she is only rearing up on marker 7.

If I have enough time I will increase the length of time until she can rear up for longer amount of times, up to five seconds.

And On To Trick Training

I discovered the other day that Rooth was finally clicker trained, so I moved on to teaching her her trick. I’ve decided to train her to rear up on her hind legs in the middle of the tank at a relatively specific spot in the tank and facing the right side of the tank as it relates to me. I want her to hold this position for five seconds.

So far she’s doing well. She’s rearing up a lot and trying different places and seems to get agitated if I don’t click when she does so. I only click when she rises at the right spot, though maybe I should be clicking every time she rises in the middle of the tank for now, I’m not sure. I’ve been doing this for a few days and it seems to work, so I hesitate to change the way I’m doing things.

The only thing that seems to get in the way is that she itches a lot and stops to bathe herself a lot. That’s probably because I’m not feeding her as often so she gets distracted more often.

Feeding

On a different note, I noticed earlier this week that Rooth was losing weight. I thought that it was just a bloop but I gave her a little more food anyway. This happened for a few more days. Today I’m happy to report that after boosting the amount of food I gave her she is up over 200 grams! I was starting to get worried.

Another thing that I found interesting was that Rooth no longer leaves presents when she’s on me. This I very much appreciate. It seems like she has learned not to pee on me, or at least that she’s no longer as afraid and feeling as much of a need to spot pee. I’ll have to see if this remains true in the future.