Abstract
In a previous study a negative relationship between word frequency and ease of recognition of error in a proofreading task was noted. This finding was tentatively explained by the mediating role of subjective word probability, an hypothesis supported in subsequent work with a different proofreading task. In the present study 50 graduate students provided measures of subjective word probability for the 66 typographical errors employed in the first study. The mediating role of word probability was again supported by significant relationships between this variable and word frequency (t = 1.82, df = 64, p < .05) and recognition of error (t = −2.18, df = 62, p < .05) and by the disappearance of the relationship between the latter two variables when word probability was controlled (t = −1.36, df = 61).
Get full access to this article
View all access options for this article.
