Evaluating the Effects of Two Key Temporal Intervals During Pavlovian Training on Conditioned Reinforcement

Reinforcers are anything that increases or maintains the frequency of a behavior that precedes their occurrence. Reinforcers can be separated into two types, primary and secondary. Primary reinforcers are biologically relevant stimuli like food, water, or drugs that can influence behavior without any prior training. Secondary reinforcers are those that impact behavior through their trained relationship with a primary reinforcer, like money and tokens. For this reason, they are also known as conditioned reinforcers. Conditioned reinforcers are used in therapies and training designed to increase desirable behaviors, but they are also involved in the development and maintenance of maladaptive behaviors such as drug use and overeating. To both increase effectiveness of behavioral therapies and reduce maladaptive behaviors, we must fully understand the factors that result in the development of effective conditioned reinforcers. To that end, the experiments in this dissertation examined a wide range of temporal variables during training of the relationship between the intended conditioned reinforcer and primary reinforcer thought to be critical in the development of strong conditioned reinforcers. Chapter 2 examined the relationship between the duration of time between training trials and conditioned reinforcer efficacy. It was previously thought that this relationship should be linear, with longer intervals resulting in better conditioned reinforcer efficacy. Instead, the results showed that the relationship broke down at very long intervals. A mathematical model was developed to account for this breakdown. Chapter 3 tested a different interval, the time between presentation of the intended conditioned reinforcer and the primary reinforcer. A breakdown was also observed at intervals predicted to produce strong reinforcers, which was not accounted for by the model proposed in Chapter 2. These results have important implications for those who wish to use conditioned reinforcers in behavioral treatments and present new questions about the ideal conditions in which a conditioned reinforcer is trained.