While conducting experiments via the Internet has become quite popular recently, there is still an ongoing debate regarding the reliability of data obtained using this method, especially for subtle manipulations and measurements susceptible to minor changes (e.g., reaction times). In this series of two experiments employing the emotional Stroop task (using emotional word stimuli differing in their valence, arousal, and subjective significance levels), we compared the reaction times of participants taking part in experiments either in the laboratory (Experiment 1) or online (Experiment 2). In line with previous studies, there were no significant differences observed between the two experiments. Both modes of conducting studies yielded a similar pattern of results, namely interactions between valence and arousal, and a three-way interaction between valence, arousal, and subjective significance. We conclude that the pattern of disturbance in cognitive processing caused by affect is not susceptible to the setting that the subjects are in, which may be a significant argument for reliability of affect-related experiments conducted online.