The thesis puts forward the proposition that a reinforcing stimulus also has the characteristic of being an unconditioned inhibitor. The inhibitory effect is manifested in postreinforcement pauses. Thus, duration of postreinforcement pauses is an increasing function of the magnitude of a reinforcer.
The proposal was examined in two ways: (i) previous literature on the postreinforcement pauses and on the effects of reinforcer magnitude was reviewed in order to see whether findings that have appeared contradictory could be accounted for; and (ii) in a series of 15 experiments the relationship between reinforcer magnitude and postreinforcement pause was investigated. Results of both inquiries strongly supported the proposition.
The experiments were conducted with rats, responding under a continuous reinforcement schedule. The main findings were as follows:
1. Duration of postreinforcement pauses was an increasing function of reinforcer magnitude.
2. The pause duration after a reinforcer with a given magnitude was an increasing function of the magnitude of penultimate reinforcer.
3. When 50% of reinforcements were omitted randomly, following an omission the second of two successive postreinforcement pauses was longer than the first.
4. The positive pause - magnitude relationship was also observed with a response other than the lever press. However, when the reinforced movement was rather extensive, the positive relationship was not present.
5. In extinction more responses occurred following training with a high magnitude than with a low reinforcer magnitude. Thus the above findings did not indicate, in any general sense, that responses established with a reinforcer of low magnitude were "stronger".
Finally it was suggested that although it may seem paradoxical for a reinforcer to inhibit responses, the effect is likely to have biological utility. Post-reinforcement
pauses provide an organism with opportunity for switching to other types of behaviour, and thus to interact with other aspects of the environment.