The present study was designed to examine affect-related respiratory responses during film scenes, reaction time (RT) and cold pressor (CP) tasks. In addition, I investigated whether affect-related respiratory responses could be characterised by a number of underlying dimensions, such as positive versus negative affect and active versus passive coping. Respiratory output was indexed by a detailed analyses of various volumetric and timing components of the breathing cycle. There were fundamental differences in respiratory responses to different types of affective and cognitively demanding tasks. The emotionally loaded film stimuli showed clear effects on respiration whenever the films elicited amusement and disgust. That is, amusement (laughter) induced a decrease in inspiratory time and tidal volume, whereas disgust could be linked to a prolongation of inspiratory pauses (breath-holding). Relative to a pretask baseline, RT performance elicited a relatively fast, shallow and regular breathing pattern. The pattern of breathing in response to CP was in stark contrast to the RT task, in that a substantial increase in tidal volume, minute ventilation, mean inspiratory flow and breathing irregularity was accompanied by no change in total cycle duration. The implications of these results are discussed with regard to specificity and dimensional models. It is suggested that specificity models might apply to phasic respiratory activity such as breath-holding in disgust and pain. On the other hand, breathing irregularity and expiratory pause duration appears to correspond to emotion dimensions (positive versus negative affect) that cut across the boundaries of emotion categories. Lastly, there were the respiratory patterns that could be more readily interpreted in terms of a task-involvement or active versus passive coping attitude.