In two previous experiments, we studied how stimulus motion affects both the self-report of emotion experience and the physiological sequelae of emotion. In both studies, image motion intensified emotional responding, and the effect of motion was relatively specific to the arousal dimension of the emotion; there was little evidence that image motion altered the valence of the image. Moving images also appeared to sustain the attention of the participants for a longer period of time than did the still images. In these two experiments, however, image motion was manipulated within participants. In the present experiment, we used a between-subjects manipulation of image motion and found a nearly identical pattern of results. These data indicate that motion inherently increments the arousal value of an image and that this increment is not dependent on the context in which motion is introduced.