An examination of the sentence task through in-person and online administrations

Abstract

The Sentence Task (Rachman et al., 1996) has respondents write “I hope ___ is in a car accident,” filling the blank with a loved one. Used in many laboratory-based studies of Thought-Action Fusion (TAF), it is unclear whether the induced distress is TAF-specific. Four studies (total N = 405 undergraduates) examined the extent to which Task-induced anxiety is empirically correlated with conceptually relevant self-report measures. We also explored using the Task in an online format; two studies were in-person, two were online. First, completion of the Task resulted in significant (ps < .001) increase in anxiety in all four studies (ds ranged from 0.52 to 0.86). However, in three of the four studies, there was no significant correlation (ps > .05) between the magnitude of pre-to post-Task anxiety change and baseline measures (Thought-Action Fusion Scale, Dimensional Obsessive-Compulsive Scale). This suggests that the Task-induced anxiety may not be specific to TAF. Second, there were no significant mean-level differences for any variables whether assessed in-person vs. online. However, Task refusal was significantly higher for online (21.2%, 24.1%) vs. in-person (5.4%, 6.0%) administrations. Findings contribute to a better understanding of the Task and offer mixed support for use in an online format.