Sniffy the Virtual Rat

As untechnologically savvy as I am, I was finally able to figure out my problems with the Sniffy program. Sniffy is a computer program that simulates the type of environment that is involved with training in an operant box. Sniffy is not designed to actually produce result, but instead to help Psychology of Learning students practice training using operate conditioning works before trying to train a real rat.

Sniffy and I were then able to begin magazine training. It only took us about 30mins, but I got a really good idea of how long a session with Ruby will last. The purpose of magazine training is to shape Sniffy to associate the sound of the bar with a pellet dropping in the food magazine.  Magazine training prepares Sniffy for shaping as it begins to recognize that when food is dispensed it is doing s behavior that I want it to increase, positive reinforcement. Below is the cumulative record of his magazine training.

After Sniffy was fully magazine trained, I began shaping Sniffy to press the bar. Having such a smooth time with magazine training, I definitely thought shaping would be easier than it was. The instructions in the book say that it should take about 30 to 45 minutes to shape Sniffy, it took me approximately an hour and a half. I would find myself talking to sniffy trying to instruct him to press the bar which of course did no good because the virtual rats if he can't hear my instruction it's also probably not advised to talk to the real rat as verbal communication can be seen as reinforcement.

Following Sniffy's initial shaping, I began transitioning from a Fixed Ratio 1 (every time it pressed the bar it received a reinforcement) to a Fixed Ratio 2, where it would be awarded a pellet every second time it pressed the bar. This took much longer than I thought, approximately 1 hour, simply because I didn't realize that the Sniffy program didn't allow me to go to other websites (wanted to work on my blog and leave Sniffy to it's own devices). That was my fault for thinking I could multi-task. Honestly, I think it's beneficial that the program doesn't allow the trainer to venture elsewhere since he or she also shouldn't be multi-tasking when shaping his or her live rat. I noticed that Sniffy appeared to be confused and even a little frustrated when I put it on an FR 2 schedule. It would press the bar and then press it again and no food would come out so it would run to a different area of the box seemingly in frustration. I then progressed to an FR 5 schedule and then began extinction. Sniffy was extinct faster than it was able to be shaped which I don't think the same will be the case for my real rat, Ruby. 

FR 2 Cumulative Record

 Sorry this one is a little blurry, but I thought it was still important to post so you viewers can see the decrease in Sniffy's behavior after extinction.