UC San Diego

The present set of experiments tested the influence of events outside of the concurrent-chains procedure on choice proportions and other behavioral measures within the procedure. Context effects on the initial link relative response rates (i.e., choice proportions) were not found in each experiment. However, when a choice proportion context effect was found, the results generally indicated that choice proportions increased as a function of increasing rates of reinforcement in the context. This result is not predicted by the three currently popular quantitative theories of choice. However, because the effects found were relatively weak the current theories of choice do not require modification. It was further shown that reinforcement in the context must be response- dependent to have an impact on choice proportions. However, increases in initial and terminal link response rates as a function of increases in contextual reinforcement were found in conditions with both response-independent and dependent reinforcement. It appeared that the modulation of initial link response rates was due in large part to reinforcement rates in the context. In general, choice may not need to be conceptualized as a process independent of external factors. Just as other behavioral theories such as contrast and the quantitative law of effect have acknowledged and synthesized the impact of distal events on proximal behavior, general theories of choice may need to do the same if the results from some of the current experiments can be replicated