There are two independent variables in the experiment. The first independent variable is the type of reinforcement schedule: fixed-ratio versus variable-ratio schedule of reinforcement. The second independent variable is the overtraining time ”“ low, medium and high. The dependent variable is the resistance to second extinction of the magazine trained rat in trials with above-mentioned parameters.
The purpose of the experiment is to study the effect of these variables on the second resistance to extinction, and to determine how type of reinforcement schedule and overtraining time interact with each other in these trials. Null hypothesis in this research can be stated as follows: there are no statistically significant differences between the second extinction for rats with variable or fixed reinforcement schedule, and different overtraining time. Alternative hypothesis is the following: there exist statistically significant differences between the second extinction for rats with variable or fixed reinforcement schedule, and different overtraining time; in other words, alternative hypothesis states that the rats can be extinguished twice.
This experiment was held using software “Sniffy Pro 3.0”ť in a laboratory setting. This software models the behavior of a magazine trained laboratory rat in a Skinner Box (laboratory chamber). SPSS software was used to perform statistical data processing.
The subject involved in the experiment was computer-simulated rat.
The virtual rat was trained to press the bar at different level increment (5, 15, 25, 35, 50 times). At the second step of the experiment, the mouse was over-trained at different levels (low, medium and high), and at each of these levels, fixed-ratio and variable-ratio schedules were used for extinguishing bar-pressing behavior. Resistance to extinction was measured for each combination of values of independent variables, and after than, spontaneous recovery along with resistance to second extinction was measured.
Operational definition of fixed-ratio schedule: the virtual rat is rewarded a pellet when pressing the bar for a given number of times during 5 minutes; in this case, there will be experiments with 5, 15, 25, 35 and 50 times. Operational definition of variable-ratio schedule: the virtual rat is rewarded a pellet when pressing the bar for on average a given number of times during 5 minutes; the exact occasion of rewarding a pellet is non-predictable (computer-generated random reinforcement). Operational definition of over-training level: the virtual rat is receiving an electric shock for 1 sec; the intensity of shock can be low, medium or high; shock intensity is recorded as the overtraining level for the trial.
Operational definition of resistance to extinction: following previous response conditioning with the help of the reinforcer, the outcome is further removed, and the time needed to revert to the behavior which was prior to initial learning is measured as the resistance to extinction (in this research, extinction criterion is the situation when the virtual rat presses the bar no more than two times during a 5-minute period). The second response to reinforcement has the same operational definition, with the only difference than the extinction takes places after the first extinction, and not after the conditioning with the help of reinforcement.
One-way ANOVA and univariate analysis (used as part of SPSS package) were applied for the analysis of data and for testing the null/alternative hypothesis; confidence level was selected as 0.05.