Bridging the Gap between Autonomous and Predetermined Paradigms: The Role of Sampling in Evaluative Learning
Datasets usually provide raw data for analysis. This raw data often comes in spreadsheet form, but can be any collection of data, on which analysis can be performed.
While most evaluative learning paradigms remove participants’ autonomy over the information they receive, other research traditions have demonstrated that information sampling has an important role in learning. We investigate the impact of information sampling on a central evaluative learning paradigm: evaluative conditioning. We compare a traditional evaluative conditioning paradigm with a paradigm in which participants have autonomy over the stimulus pairings they receive. Participants in the high-autonomy condition show a strong preference for positively paired CSs. Nevertheless, we obtained evaluative conditioning effects in all conditions. High-autonomy participants, but not their low-autonomy counterparts, also show effects of the sampling decision on evaluations. Specifically, sampled stimuli become more positive, whereas ignored stimuli become more negative over the learning phase. Thus, the present research provides a cornerstone for integrating several research traditions within and beyond the evaluative learning literature, providing a foundation for new insights and more comprehensive theories of learning.