Priming in word stem completion tasks: comparison with previous results in word fragment completion tasks

This study investigates priming in an implicit Word Stem Completion (WSC) task. A total of 305 participants performed a WSC task in two phases (study and test). The test phase included 63 unique-solution stems and 63 multiple-solution stems. After confirming the presence of priming (mean = 0.22), analysis revealed that it was stronger in the case of multiple-solution stems, indicating that the stems were not a homogeneous group of stimuli. Thus, further analyses were performed only for the data of the unique-solution stems. The correlations between priming and a set of conceptual (familiarity, frequency of use, number of meanings) and non-conceptual (letters/blanks ratio and difficulty of completed) variables showed significant relationships between two conceptual variables (familiarity and frequency) and priming. Difficulty was also significantly correlated with priming. The most familiar and frequent words were those that produced a greater magnitude of priming. At the same time, the most difficult stems were those generating more priming. A regression analysis showed that the difficulty of completing a stem was the strongest predictor of priming. When difficulty was the dependent variable in the regression analysis, the significant variables in the regression were familiarity and letters-blanks ratio. Finally, a comparison was made between these results and those obtained in a previous study of WFC by Soler et al. (2009) in which the same words and procedure were employed. A comparison of results from these two sets of data suggested that the only relevant difference between the two tasks was the influence of the variable letters-blanks ratio. This perceptual variable had a significant correlation with priming only in the WFC task. These results highlight the importance of controlling the characteristics (conceptual and non-conceptual) of stimuli used in WFC and WSC tasks when exploring the nature of priming.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.