Simulations Relationship To Learning In Operant Conditioning

Background on Operant Conditioning

Discuss About The Games Simulations Relationships To Learning.

Operant conditioning is a type of learning where an organism learns through the consequences of its behaviour (Skinner, 1938). For instance, when studying the Sniffy the rat’s behaviour under a particular condition, one variant behaviour would be constantly moving towards the bar to get the reinforcement (food pellets). A second variant behaviour could be Sniffy standing on his hind legs because it is one step closer to the target behaviour (Elcoro, 2013). This second variant is then reinforced until Sniffy constantly does this behaviour. The reinforcement of variant behaviours continues until the subject finally completes the target behaviour. In this way, a regular pattern of behaviour is conditioned to the rat. In the operant conditioning the subject has a choice to respond different from classical conditioning where the subject does not have a choice to respond. In the experiment, Sniffy the rat is put in a chamber and its behaviour is observed and specific data relating to the dependent variable of bar pressing behaviour are recorded for analysis. It imitates many of the behaviours one would observe in a real rat learning to operate in the controlled environment of an operant chamber (Graham, Alloway, & Krames, 1994). It was designed to provide a realistic conditioning experience in a cost effective and ethical way. The Operant chamber or “Skinner box” was constructed by B. F. Skinner while developing the laws of operant conditioning in 1930, to be used in experiments aimed at identifying an animal’s response under controlled conditions (Skinner, 1935). These modified behaviours through operant conditioning are called operant behaviours which occur spontaneously (naturally). In the context of the Sniffy program, Sniffy has an entire repertoire of possible operant behaviours such as standing up against the wall, walking around, cleaning its genitals, scratching its ears or even occasionally pressing the bar mounted on the wall.

 This paper contains the laboratory report that summarizes details about the Sniffy program as well as the many steps for conducting operant conditioning with Sniffy the virtual rat. Using the Sniffy program, we are able to train Sniffy’s operant behaviour to demonstrate learning through basic operant conditioning.

Apparatus.

·         The lab manual

·         Sniffy Program and MAGTRAIN program (optional)

·         Stop watch, or stop watch app on your phone

·         Paper & pen

The experiment was conducted with a class of undergraduate psychology students chosen through convenience sampling. The students were required to work in teams of two experimenters to each rat. The two were to agree on the role of each person and stick to the job each one agrees to perform. For instance, when one is operating the apparatus and the other one recording data.

The Sniffy Program as a Simulation

In preparation for each Step, they were instructed to ensure that their equipment is operating properly and have pieces of paper prepared to record data. For Step 1, they were not to operate the food reinforcement delivery mechanism or press the bar. The essential first step would be to measure the baseline level of the operant prior to any learning.

  The experiment’s independent variable would be the different operant behaviours to be reinforced and the dependent variable would be the total cumulative bar presses. There were 5 steps in total, Observation, Magazine training, Shaping, Conditioning and Extinction. Students were asked to follow specific steps given to each pair to stimulate a live rat experiment. The experiment was completed under the supervision of the tutor assigned to the class.

Step 1. Observation. The students were required to place the rat gently in the box and quietly close the lid and then observe its behaviour for 3 minutes, making notes on everything it does minute by minute. The recordings were supposed to be specific for example "washes genitals," "stands on hind legs," "sticks nose into corner," “scratches left side” and so forth. At the bottom of the sheet, they were to summarize their observations by indicating the number of different behaviours noted, the four most frequent, and the number of times each occurred. In any case they would experience difficulty increasing the probability of a response they were to tally (///// //, etc.) every time the rat touches the bar with his paw, and in a separate column, every time it presses it. A press is arbitrarily defined as one that produces a click in the apparatus. After observation, they were to record the Cumulative response data in a table (see Table 8).

Step 2. Magazine training. Here, the rat was to go to a particular place (the magazine) for his reinforcements. The requirement was to deliver one reinforcer at a time when the rat is relatively inactive. The click and the sound of the reinforcement delivery might disrupt the rat and it may take it a while to find the food. This was to continue as the rat consumes each reinforcer and moves away before delivering the next one until a clear pattern of behaviour is identified.

Step 3. Shaping. In this step the target was to condition the rat to press the bar in order to get the reinforcement. Starting with behaviour only remotely resembling bar pressing, gradually shifting to behaviour that come closer and closer to a true bar press. "Coaxing" was used to get the rat closer than before in every subsequent reinforcement until the rat is at a point very near the bar. Reinforcement would start when Sniffy faces the bar, then it was required to stand up near the bar. For the next one, it was required to touch the bar; then to push it a bit; and finally to press it to produce a click. This would take between five to ten minutes depending on the skills of the students and the cooperation of the rat. Some behaviours would be clear bar presses while others would be “almost-bar-presses” and both would be treated them same way and the reinforcement given. The students were to record the bar presses in the table (see Table 10).

Steps for Conducting Operant Conditioning Experiment with Sniffy

Step 4. Conditioning. In this step they were to observe the increase in rate of the bar pressing response. The apparatus were to be set so that reinforcement is delivered each time the bar is pressed. This is called continuous reinforcement or sometimes CRF. Records of the bar presses on a minute-by-minute and cumulative basis were to be made. They were to allow the rat 20 minutes of reinforcement during which one of the experimenters would record every incidence of any of the four most frequent behaviours noted and tabulated in Step 1. After the rat has consumed the last reinforcement, discontinue conditioning was done.

Step 5. Extinction. At this point, the rat was fully conditioned. For the extinction step, the students were instructed to select the "Experiment" command, and then select "Design Operant Experiment" then click the "Extinction" checkbox. Under this procedure, Sniffy would no longer get any reinforcement for pressing the bar. Sniffy should gradually stop pressing the bar.

Step 6. Spontaneous recovery. Once the bar pressing response is extinguished, Sniffy’s conditioned response is observed at a different time, may be after 24 hours. To finish up the experiment, students were instructed to select the "Experiment" command, and then select "Remove Sniffy for time out" and then click OK. This operation simulates having removed Sniffy from the chamber for a 24 hour rest, and then returning him.

Behavioural observations made were: cleaning face, sticking nose, grooming, standing up, bar touching and bar pressing. Out of those the most frequent behaviours exhibited by Sniffy was “standing on hind legs” and “cleaning genitals” with a total of 36 “standing up” and 22 “cleaning genitals” actions. Cleaning nose was 20 times and sticking nose into the corner was 12 times (see Table 7).  Three bar touches were recorded in the first minute, then two then three in the subsequent minutes with a cumulative total of 7 bar touches. Similarly, three bar presses were recorded for the first minute, two bar presses for the second minute and 3 bar presses was recorded for the third minute, with a cumulative total of 8 (refer to Table 8). Of all the behavioural observations, standing on hint legs was the most frequent.

Conclusion

According to the analysis of the data collected and the graph above, the cumulative curve of responses of the rat for the entire sequence shows a gradual increase in the conditioned behaviour. The baseline (natural) behaviour that was exhibited initially gradually decreases under the conditioned stimuli. The stages of extinction and spontaneous recovery can be used to check the memory of the operand that is being conditioned.  For instance, when Sniffy is removed from the chamber for 24 hours and the procedure of conditioning is repeated afterwards, this will show whether Sniffy can actually remember that procedure for bar pressing for it to be reinforced (given food pellets).

In conclusion, Sniffy program allow the user to control, manipulate and modify several aspects of the program to simulate operant conditioning according to the needs of the user.

References

Graham, J., Alloway, T., & Krames, L. (1994). Sniffy, the virtual rat: Simulated operant conditioning. Behavior Research Methods, Instruments, & Computers, 26(2), 134-141.

 Pavlov, I. (1977). Classical conditioning. Learning and Instruction, 4, 26.

Skinner, B. F. (1935). Two types of conditioned reflex and a pseudo type. The Journal of General Psychology, 12(1), 66-77.

Skinner, B. F. (1938). The behavior of organisms: an experimental analysis. Appleton-Century. New York.

Aloway, T., Wilson, G., & Graham, J. (2011). Sniffy the virtual rat pro, version 3.0 (with CD-ROM). Nelson Education.

Gaham, J., & Alloway, T. (1995). Sniffy the virtual rat. Wadsworth Publishing Company.

Graf, S. A. (2008). Three nice labs, no real rats: A review of three operant laboratory simulations. The Behavior Analyst, 18(2), 301-306.

Venneman, S. S., & Knowles, L. R. (2005). Sniffing out efficacy: Sniffy Lite, a virtual animal lab. Teaching of Psychology, 32(1), 66-68.

Gredler, M. E. (2007). Games and simulations and their relationships to learning. Handbook of research on educational communications and technology, 2, 571-581.

Lieberman, D. A. (2009). Learning: Behavior and cognition. Wadsworth/Thomson Learning.

Elcoro, M., & Trundle, M. B. (2013). Student Preferences for Live versus Virtual Rats in a Learning Course. International Journal for the Scholarship of teaching and learning, 7(1), n1.

Lemmon, C., Lui, S. M., Cottrell, D., & Hamilton, J. (2012, October). Challenges to develop an interactive 3D virtual world for psychological experiments. In Proceedings of the European Conference on Games Based Learning (pp. 278-284).