Respiration as a source of intrasubject variation in simple information processing tasks

Abstract

Intrasubject variation (ISV) may contribute unpredictably to the results of tasks requiring momentary processing of information and may arise from several sources, such as motivation, fatigue, or internal physiological changes. Research has indicated that one of the possible internal physiological sources of ISV may be associated with respiration. The purpose of the present research was to examine the relationship between respiratory phase and two levels of information processing. In the first experiment, pneumographic respiratory records were collected from 10 female Ss during visual and auditory detection tasks. The results revealed that in both tasks Ss detected more threshold level signals presented during the exhalation phase of respiration than during the inhalation phase. Reaction times, however, were not related to the phase of respiration. Respiratory records were also obtained in the second experiment during a slightly more complex information processing task in which Posner's (1964) paradigm for classification of letters was utilized in a physical match task and a name match task. Ten female Ss performed each task and the results suggested that respiration phase was not related to this type of information processing. However, the results supported the information processing model suggested by Posner and Mitchell (1967) and extended it to include word classifications. The results of each experiment were discussed in relationship to possible neurophysiological explanations as well as other psychological factors. The respiratory records in both experiments were further scrutinized in an exploratory manner to examine the consistency of the six respiratory measures over trials and their relationships with task performance measures such as detection probabilities and reaction times. These results indicated that the respiration measures tended to be stable over trials and tasks, but their relationship to performance measures was unreliable. Suggestions for future research were discussed.