Université de Poitiers
Centre National de la Recherche Scientifique
Centre de Recherche Sur la Cognition et l’Apprentissage
Rapport technique : 2008/03/M.AUD
Acute aerobic exercise and information processing: Energizing motor processes during a choice reaction time task Rémi L. Capa, Michel Audiffren, Université de Poitiers, Centre de Recherche sur la Cognition et l’Apprentissage, CNRS, MSHS, 99, Avenue du Recteur Pineau, 86000 Poitiers, France ; Phillip D. Tomporowski and James Zagrodniky, University of Georgia, Department of Kinesiology, Georgia, USA
A paraître dans / To appear in : Audiffren, M., Tomporowski, P.D., & Zagrodnik (2008). Acute aerobic exercise and information processing : energizing motor processes during a choice reaction time task. Acta Psychologica.
Address correspondence to: Michel Audiffren Laboratoire Langage, Mémoire & Développement Cognitif Maison des Sciences de l’Homme et de la Société 99 avenue du recteur Pineau 86000 Poitiers FRANCE Phone: + 33 (0)5 49 45 46 02 Fax: + 33 (0)5 49 45 46 16 e-mail:
[email protected] 1
KEYWORDS:
Electromyography, motor processes, motor time, physical activity, physiological arousal, premotor time, reaction time, stimulus intensity.
ABSTRACT The immediate and short-term after effects of a bout of aerobic exercise on young adults’ information processing were investigated. Seventeen participants performed an auditory two-choice reaction time (RT) task before, during, and after 40 min of ergometer cycling. In a separate session, the same sequence of testing was completed while seated on an ergometer without pedalling. Results indicate that exercise (1) improves the speed ofreactions by energizing motor outputs; (2) interacts with the arousing effect of a loud auditory signal suggesting a direct link between arousal and activation; (3) gradually reduces RT and peaks between 15 and 20 min; (4) effects on RT disappear very quickly after exercise cessation; and (5) effects on motor processes cannot be explained by increases in body temperature caused by exercise. Taken together, these results support a selective influence of acute aerobic exercise on motor adjustment stage.
INTRODUCTION Since 1990, more than 20 studies have clearly shown an improvement of information processing during sustained ergometer cycling at an intensity ranging from 40% to 70% of VO2max (e.g., Adam, Teeken, Ypelaar, Verstappen, & Paas, 1997; Arcelin, Delignières, & Brisswalter, 1998; Chmura, Krysztofiak, Ziemba, Nazar, & Kaciuba-Uscilko, 1998; Davranche, Audiffren, & Denjean, 2006; McMorris & Graydon, 1996; Paas & Adam, 1991; Pesce, Capranica, Tessitore, & Figura, 2002; Pesce, Casella, & Capranica, 2004). The facilitating effect of information processing observed in young adults during or immediately after a bout of moderate aerobic exercise is generally explained by an increase in physiological arousal induced by the physical activity (e.g., Davey, 1973; McMorris & Graydon, 2000; Näätänen, 1973; Thayer, 1987). In spite of the strength of the descriptive evidence, lacking is an understanding of the mechanisms by which exercise influences the components of the information processing system. Sanders (1983, 1998) proposed a cognitive energetical model in which three mechanisms (i.e., arousal, activation, and effort) influence specific stages of information processing. Sanders argued, on the basis of a large number of reaction time (RT) experiments using the additive factors method of Sternberg (1969, 1998), that arousal is linked to the feature extraction stage, activation influences the motor adjustment stage, and effort is involved in response selection. Within this theoretical framework, the manipulation of physical activity has been used to elucidate the roles of arousal and activation on sensory and motor stages of information processing. The results of the studies conducted to isolate the locus of exercise-induced arousal on specific stage of processing have been inconsistent. Two studies (Arcelin et al., 1998; Davranche & Audiffren, 2004) using Sternberg’s additive factors method (AFM) (1969, 1998) were unable to localize the facilitating effect of moderate aerobic steady-state exercise on RT. Studies that have employed electromyography (EMG) have been revealing, however. In these experiments, RT is fractionated into two components: (1) premotor time, which is the interval between the onset of the response signal and the onset of EMG activity of the response muscle and (2) motor time, which is the time interval between the onset of EMG activity and the onset of the required motor response (Botwinick & Thompson, 1966). Motor time (MT) reflects the duration of the electromechanical transduction within muscular fibres, whereas premotor time (PMT) reflects the duration of all earlier stages of information processing. Sanders’ model (1990) views MT as a part of the motor adjustment stage of processing. Analysis of PMT and MT provides the means to determine whether the facilitating effect of acute bouts of exercise on RT occurs prior to or after the onset of EMG activity. Measurement of PMT and MT isolates early cortical-integration processes from later motor-movement processes (Hasbroucq, Burle, Bonnet, Possamaï, & Vidal, 2001). Two studies using EMG fractionation of RT provide evidence that acute aerobic exercise decreases MT but not PMT (Davranche, Burle, Audiffren, & Hasbroucq, 2005, 2006). These results suggest that exercise selectively influences activation but not arousal. The purpose of the present study was to clarify the association between acute aerobic exercise and components of RT, and to determine the role of arousal and activation in the facilitating effect of exercise on reaction processes. 2
Davranche and her co-workers employed a visual choice reaction time (CRT) and observed a small but significant interaction between the facilitating effect of exercise and the effect of intensity of visual stimuli. This interaction is important as it suggests that exercise may influence stimulus-driven sensory processes via the arousal mechanism. We elected to investigate the interactive relation between exercise and information processing via an auditory CRT. Several studies have demonstrated that a loud auditory imperative signal (>70 dB) not only provides information but also exerts an immediate arousing effect (e.g., Sanders & Andriessen, 1978). The AFM logic predicts that the combination of the immediate arousing effect produced by a loud auditory signal together with the facilitating effect of acute aerobic exercise would produce an overadditive interaction on CRT if both effects involve the same energetical system. It is also plausible that the decreases in young adults’ MT observed during exercise are due simply to an elevation of body temperature that increases the conduction velocity of both muscle fibres and peripheral nerves (Van der Hoeven & Lange, 1994). Physical activity is known to increase muscle temperature including those muscles that control hand muscles during cycling (Halar, Hammond, & Dirks, 1985). We assessed this possibility by correlating the skin temperature of the hands involved in the CRT and MT. Most of the research conducted to assess the effects of exercise on cognitive function has used experimental designs, in which cognitive function is measured prior to and again immediately following a bout of exercise; relatively few studies have measured cognitive function at time points throughout an exercise bout and during recovery from exercise (Tomporowski, 2003). As a result, little is known of the temporal relation between exercise and cognitive function. The present experiment was developed to provide information critical to understanding the relation between specific exercise conditions and cognitive function. The intensity and duration play an important role in determining the manner in which exercise influences cognitive performance (Tomporowski, 2003; Tomporowski & Ellis, 1986). The exercise intensity and duration employed in the present experiment were carefully chosen and controlled to increase the likelihood that a positive effect of exercise on cognitive processes would be observed. A 35 min steady-state exercise at an intensity of 90% of the participants’ ventila- tory threshold (VT) was selected for two reasons: (a) exercise intensity above the VT is associated with rapid and continuous increases in blood lactate level and failure to maintain a steady-state blood lactate concentration (e.g., Yamamoto et al., 1991) and (b) VT is a more critical determinant for evaluating sub maximum fitness than the maximum oxygen uptake (VO2max) (Kumagai et al., 1982; Weltman, Katch, Sady, & Freedson, 1978). To sum up, the present experiment was designed to (1) assess the effects of an acute bout of aerobic exercise on a CRT task; (2) determine whether a facilitating effect of exercise influences both PMT and MT; (3) explore a predicted interaction between the effect of exercise and the effect of the auditory signal intensity on CRT and PMT; (4) examine the correlation between MT and skin temperature; and (5) study the temporal changes in CRT performance during and following a bout of exercise. We expected (a) a reduction of CRT, PMT and MT during aerobic exercise in comparison to rest; (b) a smaller size of this effect in the loud auditory signal condition; (c) a progressive disappearance of the facilitating effect of exercise as soon as exercise stops; and (d) a modest correlation between skin temperature and MT during exercise.
METHOD PARTICIPANTS Twenty-eight participants were recruited through classes in the Department of Kinesiology of the University of Georgia and via posted flyers. Selection/exclusion criteria for participation included: (a) being between 18 and 25 years of age; (b) having no contraindications to strenuous exercise or injury as described by a medical history questionnaire; and (c) having a VO2max greater than 36 ml min1 kg1 for males and 30 ml min1 kg1 for females. Participants’ level of physical fitness is known to moderate the effects of an acute bout of aerobic exercise on RT (Brisswalter, Collardeau, & Arcelin, 2002; Brisswalter & Legros, 1996; Tomporowski & Ellis, 1986). Participants were thus selected on the basis of their physical fitness in order to reduce inter-individual variability. The choice of the criteria was made according to the norms provided by the Institute for Aerobics Research (1994), available in the American College of Sports Medicine (2000, 6th edition). The values reported by ACSM for young adults were adjusted downwards by 15% because a preliminary experiment conducted in our laboratory indicated that higher VO2max values were taxing for our sample of voluntary participants. Six males and three females did not reach the fitness criterion. The mean VO2max for these six 3
rejected participants was 31.73 ml min-1 kg-1 (SD = 2.90 ml min-1 kg-1) for the men, and 28.10 ml min-1 kg-1 (SD = 1.59 ml min-1 kg-1) for the women. One participant withdrew after the completion of the second session. From the 28 participants recruited, 18 (nine males and nine females) completed the experiment. The data of one participant were discarded as she did not comply with the accuracy instruction during the CRT. Anthropometrical and physiological characteristics of the 17 remaining selected participants are displayed in Table 1. Participants received $50 upon completion of the study, or if they withdrew or were not selected for the study, $10 for attending session 1 and $20 each for attending the two other sessions.
Table 1 Anthropometrical and physiological characteristics of selected participants of experiment 1 (mean ± SD)
Note: VO2max, maximal oxygen uptake; PVO2 max, power at maximal oxygen uptake; HRmax, heart rate at maximal oxygen uptake; VT, ventilatory threshold; PVT, power at ventilatory threshold; 90% VT, exercise intensity at 90% of the ventilatory threshold expressed as a percentage of VO2max; P90%VT, power at 90% of the ventilatory threshold.
PROTOCOL Each participant completed three experimental sessions, each lasting approximately 95 min: (1) an evaluation and practice session; (2) an exercise session; and (3) a rest session. Each session was separated by approximately 8 days (range 5–16). The evaluation and practice session consisted of five phases. During the first phase, each participant read and signed the informed consent which was approved by the Institutional Review Board (IRB) of the University of Georgia and then completed the medical history and physical activity history questionnaires. During the second phase, each participant learned the auditory CRT task, following instructions displayed on the screen of a computer. They were seated comfortably on a chair, and provided responses on the same key pads as those fixed on the handle bar of the bike. Upon completion of the instruction phase, each participant performed a block of 24 trials in which he or she was asked to respond as quickly and accurately as possible to the auditory stimuli presented via headphones by pressing the appropriate key with the thumb. The participant then performed between three and 10 practice blocks of 40 trials to stabilize performance. At the end of each block of trials, feedback concerning the average RT and the percentage of errors was given to the participant. They were instructed to improve upon the mean RT obtained in the previous block of trials as much as possible but without making more than two errors per block. The practice phase concluded when the participant reached a criterion of an error rate of less than 5% and a coefficient of variation of CRT less than 15% in two blocks of trials, with a minimum of three practice blocks of trials and a maximum of ten. The third phase involved learning to use the rating of perceived exertion (RPE) scale, which is a standardized rating instrument that allows subjects to rate how hard they feel when they are working. A numerical rating ranging from 6 (very, very light effort) to 20 (maximal effort) corresponds to the perceived relative intensity of the exercise. During the fourth phase, participants completed a graded exercise test (GXT), which provided a measure of VO 2max. Prior to performing the graded exercise test, the participant was instructed to put a heart rate monitor on his/her chest. Heart rate (HR) was 4
measured every 30 s using a Polar T61 chest strap transmitter. Then, the participant positioned himself/herself on an electronically braked cor- rival cycle ergometer (Model No. 906900, LODE BV). Seat and handlebar height were adjusted for each participant for proper position and comfort. A research assistant then described the VO2max test procedure to the participant and answered any questions. The test began with a 5 min warm-up at 25 W. Workload increased continuously at a rate of 24 W per minute thereafter. Respiratory gases were continuously sampled using open-circuit spirometry (PARVO Medics True One 2400 metabolic measurement system). The participant exercised at room temperature (21–23 °C). Verbal encouragement was given to the participant throughout the graded exercise test, but especially after the participant reported a RPE of 17 or higher. VO 2max (ml kg-1 min-1) was defined by the attainment of four criteria: (a) a respiratory exchange ratio (RER) ≥1.15; (b) a rating of perceived exertion (RPE) ≥ 18; (c) a failure to maintain the minimum cadence of 40 rotations per minute with further exercise intensity; and (d) a HR ≥ 220 minus the age of the participant. The test was terminated when the participant voluntarily stopped because he/she could not maintain the current work load. RPE and HR were obtained every one minute during the VO 2max test. Ratings were obtained by asking the participant to point a thumb up or down to indicate his/her level of exertion while we called out the scale numbers. The same apparatus was used during the ‘exercise’ and ‘rest’ sessions to control the workload and record HR and the breathing parameters. The ventilatory threshold was calculated using the V-slope method (Beaver, Wasserman, & Whipp, 1986). If the participant reached the designated fitness criterion, he/ she was asked to perform two blocks of 104 CRT trials (Phase five). The first block was performed sitting on the cycle ergometer at rest and the second block while pedalling at 90% of his/her ventilatory threshold. The electromyographic (EMG) activity of the left and right flexor pollicis brevis was recorded during these two blocks of trials. These two additional blocks were included in the protocol to familiarize the participant with the experimental conditions and to teach him/her how to relax his/her hand muscles between two muscle contractions in order to reduce the baseline EMG activity. At the end of the first session, the participant was instructed to (a) drink liberally the day before the next session; (b) drink a 237 ml glass of water an hour before the next session; (c) not alter his/her routine intake of stimulant beverages; and (d) not engage in exhausting exercise the day prior to testing. During the exercise session, the participant first completed a 24 h history questionnaire and put on a polar HR monitor. Next, EMG electrodes were placed on his/her hands. The test session began with the participant performing a block of 40 practice CRT trials while sitting on a chair in front of the computer screen. At the end of the practice block, the participant was instructed to mount the cycle ergometer, and the seat height and peddle straps were adjusted. Once the comfort and proper positioning of the participant were checked, a temperature probe was fixed on each palm. The exercise protocol involved: (1) sitting on the cycle ergometer for 5 min without exercising; (2) a 5 min warm-up period of cycling at 30% of the participant’s VO2max; (3) a 35 min exercise bout at 90% of the participant’s ventilatory threshold; and (4) a 35 min post-exercise period. The participant performed nine blocks of 104 CRT trials; respectively, 5 min before beginning exercise, and at minutes 8, 14, 22, 28, and 34 of the 40 min exercise protocol. Post-exercise CRT tasks were performed 1 min, 15 min and 30 min following the termination of exercise. During the blocks of CRT trials, participants were seated on the cycle ergometer with their front arms resting on a platform which allowed them to relax their hand and forearms. RPE was obtained during the session by asking the participant to point a thumb up or down to indicate his/her level of exertion while scale numbers were called out. RPE and HR were recorded, respectively, at min 4, 13, 19, 27, 33, 39 and 46. Skin temperature was recorded only while the participant performed the CRT task. A respiration sampling was carried out 19 min after the beginning of the exercise. The workload was adjusted if the RPE, the HR, and ventilatory parameters did not match the values recorded during the VO2max test for the same intensity of exercise. In the course of the session, the participant was supplied with water and, on request, a towel was given to him/her and a fan was directed toward him/her. The three postexercise CRT tasks were performed while the participant was seated on the cycle ergometer. At the end of the seventh and eighth block of trials, the participant was instructed to dismount. He/she had the option of sitting down on a comfortable chair located beside the stationary cycle or standing up, while waiting for the next block of trials. Participants also had the choice of reading or not reading magazines during their waiting periods. The non-exercise session was conducted using the same protocol as used during the exercise session, with the exception that the participant sat on the cycle ergometer but did not pedal. Exercise and rest conditions were counterbalanced across participants. At the end of testing, a research assistant debriefed the participant, described the intent of the study, and provided answers to any questions.
5
To minimize the effect of energetical variables, such as sleep deprivation, time of day, and stimulant social drugs such as caffeine, participants completed the three experimental sessions during different days but at the same moment of the day and were instructed to refrain from drinking stimulant beverages (coffee, tea, alcohol, and coca-cola) just before each session, to sleep normally the night preceding each session, and not to practice exhausting exercise the previous day and the day of each session.
THE AUDITORY CRT TASK Participants were asked to respond as quickly and accurately as possible to two auditory stimuli by pressing a designated key with the thumb (7.5 N). The left key was to be pressed in response to a stimulus presented in left earphone speaker and the right key in response to a stimulus presented in the right earphone speaker. Auditory stimuli were computer-generated tones presented to each participant via headphones. The auditory tones were generated by a Sound-Blaster AWE 64 sound card (Model CT4500). A trial began by the occurrence of a 100 ms auditory warning signal (70 dB, 500 Hz). The response signal followed the onset of the warning signal (WS) at a constant interval of 0.5 s. The response signal was a pure tone of 2500 Hz with an intensity of 45 or 80 dB and with a duration of 200 ms delivered in the right or left padded earphone. The two sound levels and the two ears stimulated (left/right) were randomized within a block of trials. The inter-trial interval was 2 s. The total time of one practice block of 40 trials was approximately 3 min, whereas for a block of 104 trials was approximately 5 min. During each practice block of 40 trials, the participant received qualitative feedback concerning the correctness or incorrectness of his/her response; the messages ‘correct’, ‘too slow’, ‘anticipation’ and ‘wrong key’ were displayed on the screen. The four first trials of each block of 104 CRT trials were warm-up trials and were omitted from analysis. Very slow (RT >800 ms during the practice phase, RT >500 ms during exercise and rest sessions), very rapid (RT
