The Partial Reinforcement Effect (PRE), also known as the Partial Reinforcement Extinction Effect (PREE), proposes that the resistance to extinction is more intermittent than a continuous reinforcement schedule. This calls for a look into the extinction rate concerning the different schedules of reinforcement, i.e., constant reinforcement, fixed interval reinforcement (FI-25), and variable ratio reinforcement (VR-25). The predictions say that the CRF schedule is bound to show that extinction will occur faster, while the partial schedule does show a rather remarkable degree of persistence. Understanding these dynamics informs behavior modification strategies across psychology, education, and animal training domains.
Method.
Sniffy was trained and extinguished on three different schedules of reinforcement, i.e., continuous (CRF), fixed interval 25 (FI-25), and variable ratio 25 (VR-25). Sniffy was conditioned to the ability to press a bar so that he receives food reinforcers on each of the schedules. In comparison, in a CRF condition, Sniffy could obtain food reinforcement following every response he makes to push the bar down. In the FI-25 condition, Sniffy received food after every bar press, more or less at an exact 25-second interval. In the VR-25 condition, therefore, Sniffy was reinforced after an average of 25 bar presses.
The extinction criterion was characterized by no pressing of bars within a given time frame, usually set for a period where bar presses were not occurring. The experiment had three independent variables, including the reinforcement schedule type (CRF, FI-25, and VR-25). For each of these independent variables, one level corresponded to the specific schedule of reinforcement. The mean number of bar presses to extinction and time to reach extinction were dependent variables for each schedule.
One statistical test compared the extinction rate for several schedules. In comparing the mean bar presses and the time to reach extinction between CRF and each partial reinforcement schedule (FI-25, VR-25) and the comparisons on extinction rates for FI-25 and VR-25 conditions, paired t-tests were used. Common levels of significance (α) of 0.05 in all the tests were used to allow valid results while at the same time ensuring there was reliability against errors from the side of type I.
Results:
The study permitted comparisons of extinction rates under various reinforcement schedules, which included continuous reinforcement (CRF), fixed interval 25 (FI-25), and variable ratio 25 (VR-25). It is the process of comparing the difference in the means of the number of bar pressings to reach the point of extinction and the time to get to that point between the different schedules using statistical procedures.
Comparison of the Mean Number of Bar Presses:
Extinction under CRF was compared to FI-25 regarding the mean number of bar presses until extinction. For CRF, the mean was 14.85 bar presses with a standard deviation of 2.65, whereas for FI-25, it was 22.3 with a standard deviation of 3.33. A paired t-test revealed a significant difference (t(60) = -15.179, p < 0.001), indicating fewer bar presses under CRF compared to FI-25.
Comparison of the Mean Number of Bar Presses:
Next, the mean number of bar presses for extinction under CRF was contrasted with VR-25. CRF showed a mean of 14.85 bar presses (SD = 2.65), whereas VR-25 had a substantially higher mean of 1459.05 bar presses (SD = 572.623). The paired t-test revealed a significant difference (t(60) = -19.705, p < 0.001), indicating fewer bar presses under CRF compared to VR-25.
Comparison of the Mean Number of Bar Presses:
Furthermore, a comparison was made between the FI-25 and VR-25 regarding the mean number of bar presses until extinction. FI-25 exhibited a mean of 22.3 bar presses (SD = 3.33), while VR-25 had a higher mean of 50.5746 bar presses (SD = 10.47251). The paired t-test yielded a significant difference (t(60) = -19.880, p < 0.001), indicating fewer bar presses under FI-25 compared to VR-25.
Comparison of Time to Achieve Extinction:
The time taken to achieve extinction under CRF was compared to FI-25. CRF had a mean time to extinction of 88.00 minutes (SD = 10.354), whereas FI-25 had a mean time of 325.25 minutes (SD = 38.960). A paired t-test demonstrated a significant difference (t(60) = -50.220, p < 0.001), indicating a shorter time to extinction under CRF compared to FI-25.
Comparison of Time to Achieve Extinction:
Additionally, the time to achieve extinction under CRF was contrasted with VR-25. CRF showed a mean time to extinction of 88.00 minutes (SD = 10.354), while VR-25 had a substantially longer mean of 1459.05 minutes (SD = 572.623). The paired t-test revealed a significant difference (t(60) = -18.771, p < 0.001), indicating a significantly shorter time to extinction under CRF compared to VR-25.
Comparison of Time to Achieve Extinction:
Lastly, a comparison was made between the FI-25 and VR-25 regarding the time to achieve extinction. FI-25 had a mean time to extinction of 325.25 minutes (SD = 38.960), whereas VR-25 exhibited a mean of 1459.05 minutes (SD = 572.623). The paired t-test demonstrated a significant difference (t(60) = -15.388, p < 0.001), indicating a shorter time to extinction under FI-25 compared to VR-25.
Discussion:
This study’s partial reinforcement effect (PRE) partially aligns with previous research findings; however, partial reinforcement schedules significantly slow down extinction compared to continuous reinforcement (CRF) (Killeen, 2023). This is evident in the fact that the FI-25 and VR-25 schedules yielded many more bar presses and sustained them for a prolonged time before extinction compared to the CRF schedule. Notably, VR-25 demonstrated the highest resistance to extinction, followed by FI-25 and CRF. This means that the variable-ration schedules will elicit the highest persistence strength, as predicted by the theory of intermittent reinforcement.
Different types of reinforcement schedules present different expectations for the presentation of reinforcement and differentially influence the extinction rate. When the CRF schedule reinforces every response, the subjects build a robust expectation of the reward, leading to rapid extinction upon its withdrawal. On the other hand, partial reinforcement schedules introduce variability in reinforcement delivery that provokes uncertainty and increases the persistence of responding. Knowing that reinforcement may occur intermittently, subjects react even without reinforcement.
Some theoretical explanations go into the mechanics of the PRE, such as the Frustration Hypothesis by Amsel and the Sequential Hypothesis by Capaldi. Amsel suggested that the frustration of inconsistency in reinforcement would produce heightened persistence in the behavior, whereby subjects continue to seek reinforcement despite the inconsistency associated with its delivery (Nicholson, 2021). Capaldi’s Sequential Hypothesis indicates that intermittent reinforcement disrupts the learning of an extinction response and hence promotes the prolonged persistence of behavior (Capaldi, 1967). Both theories give valuable insight; however, Amsel’s frustration hypothesis seems more comprehensive in terms of the cognitive aspects of the phenomenon (Nicholson, 2021). That is, inconsistent reinforcement leads to frustration and would cause more remarkable persistence in the behavior that brought forth emotional factors in shaping behavior.
Real-life examples of PRE include persistent begging behavior in children and pets. For instance, if a child persistently begs, even if she receives candy occasionally, she will likely persist in her behavior, even if she does not receive it in the future. If a pet gets treats occasionally for performing tricks, it might continue in the behavior without reinforcement, anticipating a reward. Appreciation of the PRE could provide strategies for behavior modification, focusing on the fact that steady reinforcement is essential for bringing about desired behavior.
Having observed the outcomes of this study, I may reconsider how I would behave with my pets or children. Instead of reinforcing the desired behaviors occasionally, I would want to be consistent in my reinforcements to achieve faster extinction and avoid perpetuating unwanted behaviors. Knowledge of these dynamics, the various reinforcement schedules, and their application to behavior will help me apply better behavior modification and training techniques.
Conclusion
In conclusion, this study shows how reinforcement schedules affect behavioral outcomes, mainly through the partial reinforcement effect. The PRE shows how reinforcement schedules affect behavioral persistence and extinction rates, emphasizing the complexity of learning and extinction. The study also indicates that cognitive elements like expectation and frustration influence behavior in different rewarding settings. Recognizing these intricacies helps people improve behavior change and training. Knowing how reinforcement schedules work helps people use better techniques for specific behavioral goals. Knowing reinforcement schedule dynamics allows for more exact and targeted interventions when shaping or eliminating behaviors. The study emphasizes the need for educated decision-making in reinforcement techniques to promote desired behavior and reduce undesirable behavior.
References
Capaldi, E. J. (1967). A sequential hypothesis of instrumental learning. In Psychology of learning and motivation (Vol. 1, pp. 67–156). Academic Press.
Killeen, P. R. (2023). Theory of reinforcement schedules. Journal of the Experimental Analysis of Behavior, 120(3), 289-319.
Nickerson, C. (2021). Frustration-aggression hypothesis. Frustration-Aggression Hypothesis-Simply Psychology. Retrieved January, p. 12, 2023.
write