For a long time classical and operant conditioning were considered distinct categories of learning requiring distinct pathways in the brain. Some observations, however, seem to indicate that most learning situations contain operant and classical components at various degrees.
Pavlov's hungry dog, for example, will show appetitive behavior towards the bell, if it has seen it while it was ringing during classical training even if it was immobilized at that time. More spectacularly, Chimpanzee males, trained operantly to insert coins into a food dispenser will trade the coins to equally trained females for sex, as they normally do with food.
A key to resolving the dilemma whether operant and classical conditioning can indeed be conceived as separate entities, might be to compare the different types of associations the subjects form between the various stimuli, the response and the reinforcer. According to Pavlov (1927), Kandel and his coworkers (Kandel et al. 1983; Hawkins et al. 1983; Carew et al. 1983; Carew and Sahley, 1986) and Hammer (1993), stimulus-substitution seems to account for the capacity of the CS to elicit the CR in classical conditioning. The association is assumed to be stimulus-reinforcer in nature. According to Skinner (1938), Mackintosh (1975) and Wolf and Heisenberg (1991), associations in operant conditioning are formed between the behavioral output of the organism and its stimulus situation. The association is assumed to be response-reinforcer in nature.