Ratna vs. Sniffy

Ratna in the Operant Box( 3/12/16)

Sniffy the Virtual Rat

Even though my live rat, Ratna, and Sniffy the Virtual Rat were not exactly the same, they both were very beneficial to me. I worked on Sniffy the Virtual Rat before I began training and this allowed me to see what to expect for magazine training, shaping, and other reinforcement schedules as well as the extinction process. Furthermore, it helped me understand how the Operant Box worked. Personally, I felt like completing Sniffy before moving on to training Ratna helped me a lot because I was more comfortable about any uncertainties I had about training. One thing I liked about Sniffy was that he could be trained in one session. In training Ratna, I was forced to cut off the training session after 30 minutes. However since I did not time out the sessions with Sniffy, I was not able to fully see how the behaviors vary on different days like it did for Ratna. Furthermore, Sniffy did not satiate or get worn-out. I tried to limit the amount of reinforcements I gave Ratna, because I wanted her to perform the behavior throughout each session. Ratna was also more hyper or aggressive some days and not as active others days. Sniffy, however, had the same energy and relatively the same behavior. It was also very helpful to visualize Sniffy's responses because it helped me understand how cumulative records work. The x-axis is time and the y-axis is the number of responses.  Response rate—a steep slope of the line indicates a high response rate while a flat line indicates a low response rate. Understanding the response rates on the cumulative record for Sniffy helped me understand the cumulative records for Ratna.
              Overall, magazine training for Ratna and Sniffy were relatively similar besides the amount of time it took for them to grasp the technique. Magazine training took Sniffy about 15 minutes, while Ratna took 21 minutes to be fully magazine trained. I was able to keep both Sniffy and Ratna interested in the food magazine area by rewarding them immediately. Sniffy and Ratna quickly learned the association between the magazine and the delivery reinforcement. One difference I noticed was that if Ratna was on the opposite side of the Operant Box and I wanted her to come back to the magazine, she would respond to the delivery of the reinforcement and come back to the correct side. Sniffy on the other hand did not. Sniffy continued his behavior for a few minutes before slowly going back to the magazine. Furthermore, Sniffy had to rear over the lever in order to be considered magazine trained, while this was not the case for Ratna.

            Shaping Sniffy was definitely way easier than shaping Ratna. The program for Sniffy had a low variability of behaviors compared to the Ratna’s assortment of behaviors. For Sniffy it was easy to just watch the bar-sound graph's progress and see the trend behaviors to reinforce.  For example, Sniffy had to rear up to receive reinforcements. Ratna on the other hand was reinforced after she reared and then came down on the bar. Furthermore, I had to change my shaping technique several times because Ratna was not pressing the bar hard enough. One advantage I had over Ratna’s various behaviors is it allowed me to determine which behaviors to reinforce quickly for her to gain an association. Shaping took Sniffy 1 hour to complete, and for Ratna it took over 7+ hours over the course of two-three weeks to shape her partially( **** See Shaping Blog Day 8-14 for Cumulative Recorders).

            All in all, I found it very useful to train the virtual rat before training my live rat. Sniffy the Virtual Rat helped me pay close attention to the behavior performed because if I missed a behavior that should have been reinforced, then the bar-sound association graph decreased. This essentially helped me pay very close attention to Ratna throughout my shaping training sessions. Due to Ratna's high variability of behavior, time constraints, and Ratna’s compliance, I was only able to partially complete the shaping technique. I highly recommend using Sniffy for future classes. I also believe that requiring the students to do all the different types of reinforcement schedules in Sniffy will be beneficial because it will help them understand how the behaviors vary across the different schedules.