Characterizing Colour-Word Contingency Learning

Abstract

Contingencies are constantly found in everyday situations and humans are extraordinarily adept at learning them, whether implicitly or explicitly. The learning of contingencies can benefit behaviour in many ways, some subtle and some much more apparent. But the question of how we can improve the efficiency of our performance and what factors influence a person’s learning of contingencies remains relatively limited in exploration. In this dissertation, I use a simple contingency learning paradigm, the colour-word contingency paradigm (Schmidt, Crump, Cheesman, & Besner, 2007) to address three issues regarding the learning of contingencies: (1) Is there a cost in performance—in addition to a benefit—when learning contingencies?; (2) Does instance frequency play a role in the learning of contingencies?; and (3) Can people use higher-order associations such as semantic associations in learning contingencies? In the first series of experiments (Experiments 1-4), I found that although there is benefit in responding to events with high contingencies, there also is a cost to events with low contingencies. In the second series (Experiments 5-7), I showed that event frequency does play a role in contingency learning but that role is dependent on factors such as whether there is perfect contingency between events and how balanced the frequencies are in the high- and low-contingency conditions. In the third series (Experiments 8-10), I observed that people are capable of using higher-order associations—in this case, semantics—to guide the efficiency of their responding, however the use of such associations is dependent on whether the task encourages their use. This dissertation thereby addresses three issues where research has been quite tentative and, in some cases, the present findings are in contrast to the current literature. The overall goal was for this research to contribute to the growing literature on fundamental factors that can influence how we acquire contingencies, and to help us understand how we optimize our efficiency in learning, even when that learning occurs without awareness.