Objective Thought-action fusion (TAF), which is a tendency to make the relationship between one’s thoughts and external consequences, is a dysfunctional belief linked to obsessive-compulsive disorder (OCD). While the TAF is commonly evaluated using the Thought-Action Fusion Scale (TAFS), it cannot fully reflect the actual experience of experimentally evoked TAF. In the present study, we applied a multiple-trial version of the classic TAF experiment and evaluate two variables, reaction time (RT) and emotional intensity (EI). Methods Ninety-three patients with OCD and 45 healthy controls (HCs) were recruited. The participants were asked to read the name of a close or neutral person embedded in different positive (PS) or negative (NS) TAF statements. During the experiments, RT and EI were gathered. Results The OCD patients presented with longer RT and lower EI in the NS condition than HCs. In each group, the HCs showed a significant relationship between RT in the NS condition and TAFS scores, whereas the patients did not, although they had higher TAFS scores than the HCs. In contrast, the patients showed a trend toward a correlation between RT in the NS condition and guilt. Conclusion These findings may indicate our multiple-trial version of the classical TAF showed reliable results of the two new variables, especially RT, in the task and allow to newly identify paradoxical patterns in which the TAFS scores are high but actual performance is impaired, that is, the inefficient activation of TAF in OCD.