Most recreational and competitive cyclists complete both indoor and outdoor training sessions over the course of a training season. The decision on whether to exercise indoors or outdoors may be made based on personal preference, weather, or the availability of indoor equipment or outdoor trails. The cycling community is divided over which type of training is best. Anecdotally, some cyclists may feel as if they work harder when training indoors than they do outdoors. It is difficult to tell whether this feeling is due to an actual difference in work rate or due to the occurrence of different physiological and psychological responses indoors vs. outdoors. This information could be useful for cyclists to help them determine how to optimize time spent on training. For example, if there is a difference in the work rate between the settings, cyclists may choose to train in the setting, which allows them to work at the highest rate to elicit greater training adaptations.
Most of the current research on cycling has been performed in laboratories. Very little research has been performed on outdoor cycling in general, and much of the existing research examines time trials rather than self-paced workouts (16,19). Previous research involving outdoor cycling has been specific to demonstrating the reproducibility and reliability of certain protocols, and do not elaborate on the physiology or psychology of the exercise bouts (16,19). A study by Brown and Banister (2) compared outdoor cycling with an indoor laboratory setting, and to 3 simulated outdoor conditions using fans and lamps within the laboratory. This study found similar sweat rates despite a 5–7° C difference in the environmental temperature and an increased core temperature during the indoor trial. Additionally, outdoor cycling produced a higher mean heart rate (HR) than did indoor cycling although the participants were exercising at similar work rates (2). This finding was surprising, because a higher core temperature would be expected to lead to a higher sweat rate. A higher sweat rate could lead to cardiac drift, and therefore, a higher HR (6). No psychological measures were evaluated in the study by Brown and Banister (2), and a comparison of work rates in each condition was not emphasized. However, dehydration, as may occur with a higher core temperature, has been shown to lead to decreases in cycling performance (1). The current literature lacks research that compares both physiological and psychological outcomes of laboratory vs. outdoor cycling.
Psychological perception of a training session is commonly monitored through Borg's (1970) Rating of Perceived Exertion (RPE), a scale of 6–20 that correlates well with physiological measures of HR, power output, and blood lactate levels (6,17). The Tammen attentional focus scale has also been used in research to examine the focus of participants during exercise. It is a 1-item 10-point scale ranging from 0 (complete dissociation) to 10 (complete association). Dissociative thoughts include thoughts not related to the exercise task, such as daydreaming, observing the environment, or external thoughts. Associative thoughts include paying close attention to breathing or exercise technique, somatic sensations, and internal thoughts (18). Research using this scale has found mixed results as to whether associative or dissociative thoughts can best aid in performance (10,13). To the authors' knowledge, no specific research has evaluated these measures in a comparison of laboratory and outdoor cycling.
The purpose of this study was to determine whether a laboratory vs. outdoor cycling session stimulates different physiological and psychological responses. More specifically, it was to evaluate whether power output, core temperature, sweat rate, HR, RPE, and attentional focus are different between laboratory and outdoor cycling. It is hypothesized that participants will show higher HRs in the laboratory setting because of the lack of convective cooling. The lack of convective cooling may lead to an increased sweat rate, and therefore to increased dehydration and subsequent cardiac drift. This thermal strain may be expected to lead to a decreased power output in the laboratory setting. It is also hypothesized that the participants will demonstrate more dissociative thought patterns in the outdoor session than in the laboratory session, because of the greater possibility for distraction in this environment.
Experimental Approach to the Problem
Participants completed three visits total: an initial descriptive testing session and two experimental trials. Data were collected in the Exercise Physiology Laboratory and on a set outdoor course along a local paved recreation trail. A repeated measures, randomized, and counterbalanced study design was implemented to minimize any possible order effect or between-participants effect.
Twelve (n = 12) recreationally trained male cyclists (Table 1) were recruited from the local cycling community to participate in this study. Each participant signed an institutional review board–approved informed consent. All participants were determined to be at low risk for exercise testing based on risk stratification from the American College of Sports Medicine guidelines.
Descriptive data were collected for all participants before any exercise testing. Height was measured using a stadiometer (Detecto, Webb City, MO, USA), and weight was measured using a digital body scale accurate to the nearest 0.1 lb (Withings, Lewes, DE, USA). Body composition was assessed by hydrostatic weighing using a calibrated electronic load cell system (Exertech, Dresbach, MN, USA) corrected for estimated residual lung volume (12). Body density from the hydrostatic weighing was converted to percent body fat using the Siri equation (15) for Caucasian participants (n = 11), and the Shutte equation (14) was used for African American participants (n = 1).
Participants performed a graded exercise V[Combining Dot Above]O2 peak test to determine the maximal aerobic capacity. The participants rode their own bicycle mounted on an electronically braked cycle trainer ergometer (Computrainer; Racermate, Seattle, WA, USA). All participants rode similar style of road bicycle and had previous experience riding both outdoors and indoors on a trainer. After a brief warm-up (10 minutes at 150 W), the test protocol began with participants cycling at a workload of 95 W. The workload was then increased by 35 W every 3 minutes until volitional fatigue was reached. Workload at V[Combining Dot Above]O2 peak was calculated by adding the power output of the last completed stage to the proportion of time in the final stage multiplied by 35 W. During this test, expired gases were analyzed using a TrueOne 2400 metabolic cart (ParvoMedics, Sandy, UT, USA), calibrated for flow and gas concentration.
Participants performed 2 separate experimental trials, with no fewer than 2 days and no longer than 2 weeks between the sessions, to minimize potential complicating factors such as fatigue or changes in fitness. Participants were informed that they should refrain from any strenuous physical activity during the 24 hours before a trial, and complete a 3-hour fast before both trials. Participants were asked to keep a 24-hour food/drink and exercise log before the initial experimental trial, and were asked to repeat this in the 24 hours before the second trial. If no variations were observed between the 2 logs, the second trial took place as scheduled. If there were large changes in diet and/or activity in the log, the second trial was rescheduled for a time when diet and activity could resemble the 24 hours before the first experimental trial. Each experimental trial consisted of cycling on a 40-km set course. Before beginning, the participants read the following statement: “Exert as much effort as you normally would in a 40-km training ride. Try to keep your effort consistent throughout the ride. For example, don't do intervals. This effort should be perceived as the same for both your indoor and outdoor trial.” Participants were blinded to HR, power output, time, and distance completed, but were allowed to manually shift gears as they deemed necessary throughout the 40-km ride. One experimental trial was completed indoors, whereas the other was completed outdoors. Six participants completed the laboratory trial first, and 6 participants completed the outdoor trial first.
The laboratory trial took place in the Exercise Physiology Laboratory, where environmental conditions remain relatively constant. Participants completed the trial on their own bicycle mounted on the electronically braked cycle trainer ergometer. The outdoor trail was recorded through global positioning system (GPS), and downloaded to the RacerMate One (RacerMate Inc., Seattle, WA, USA) ergometer software, which builds a simulated course with the same geography as the outdoor course. During the laboratory trial, participants were not allowed the use of a fan.
The outdoor trials were completed along a relatively flat, out, and back course on a paved recreation trail (Keystone Trail, Omaha, NE, USA). The trials occurred in late August to early October, at a time of day when the environmental conditions outside were closest to the general conditions of the laboratory (i.e., low wind, similar temperature, and relative humidity, etc.). Participants were required to wear the same clothing and helmet during both of their trials, including any desired sunglasses or gloves. The helmet was worn in both trials for safety reasons and to maintain consistency with the thermoregulatory effects.
Both trials began with a self-directed 10-minute warm-up. Identical data collections occurred during both the indoor and outdoor experimental trials. Variables measured included power output, HR, core temperature, skin temperature, nude body weight, urine specific gravity (USG), RPE, attentional focus, and environmental conditions.
Urine specific gravity was measured before both experimental trials using a digital refractometer (Atago, Tokyo, Japan). Before and immediately after each experimental trial, the nude body weight of each participant was taken in a private location. To maintain reliability of the results of the nude body weight, participants were instructed to only drink 473 ml (16 oz) of water during each trial and not to void their bladder or bowels until after the second weight was taken.
Core body temperature was read from the ingestible Jonah Core Body Temperature Capsule (CBTC; Phillips Respironics, Bend, OR, USA) and skin temperature was read from a corresponding thermistor patch placed on the left side of the chest just below the pectoral muscle. Both core and skin temperatures were recorded continuously throughout the trials using a VitalSense monitor (Philips Respironics, Bend, OR, USA). Participants were instructed to swallow the CBTC with water and then consume a Fiber One granola bar (General Mills, Minneapolis, MN, USA) 1 hour before the start of the trial, to ensure the capsule was in a constant and proper location of the gastrointestinal tract for accurate measurements to be taken. The VitalSense monitor was worn in a holder around the participants' waist throughout the trials, and results were downloaded after each trial.
Participants took a 1 minute break after the first 20-km of each trial. During this break, participants reported their RPE according to the 15-point Borg Scale (Borg, 1973) and their level of attentional focus using the Tammen attentional focus scale (18). The participants read the following statement: “Dissociative thoughts include thoughts not related to the exercise task, such as daydreaming, observing the environment, or external thoughts. Associative thoughts include paying close attention to breathing or exercise technique, somatic sensations, and internal thoughts” and then asked to point to the number corresponding to their level of associative or dissociative thoughts, with 0 meaning complete dissociation and 10 meaning complete association. During the laboratory trial, the investigator informed the participant when they had reached 20-km and then collected data. For the outdoor trials, an investigator was stationed at the 20-km point to collect the data. Rating of Perceived Exertion and attentional focus were also collected immediately on completion of the 40-km.
Time to completion and a continuous measuring of power output and HR (through a strap worn around the chest) were recorded with the CycleOps Power Tap power meter (Saris Cycling, Madison, WI, USA), which has previously been validated (3,5).
Environmental conditions of wind speed, temperature, relative humidity, heat index, dew point, wet bulb, and barometric pressure were monitored onsite at the beginning, middle (20-km), and end of each trial using the Kestrel 3500 portable weather station (Kestrel Meters, Birmingham, MI, USA). Readings were then averaged to represent the conditions during the ride.
An a priori power analysis was performed based on pilot data for power output, the criterion variable of the study, which demonstrated a mean difference of 20.3 W and a SD of difference of 14.7 W between the 2 conditions. The a priori power analysis determined that a sample size of at least 7 participants was required to achieve a power of 0.80. Power output, HR, skin temperature, core temperature, sweat rate, initial USG, RPE, attentional focus, and environmental conditions were all analyzed using dependent t-tests. Body weight was analyzed using a 2-way (trial × time) repeated-measures analysis of variance. When a significant F value was found, Fisher's protected least significant differences post hoc analysis was performed to detect where differences occurred. A probability of type I error of less than 5% was considered significant (p ≤ 0.05). All statistical analyses were performed using Microsoft Excel and the Statistical Package for Social Sciences software (SPSS 19.0).
Wind speed was higher in the outdoor trial than the laboratory trial (p = 0.02). There was no difference in environmental temperature (p = 0.932), relative humidity (p = 0.151), heat index (p = 0.120), dew point (p = 0.301), wet bulb (p = 0.887), or barometric pressure (p = 0.945) between trials (Table 2).
Power output was 30.00 ± 0.05% higher in the outdoor trial than in the indoor trial (p < 0.001, Figure 1). All participants had a higher-power output outdoors than in the laboratory, with a minimum increase of 11.2% and the largest increase being 69.6%. Participants took less time to complete the outdoor trial than the laboratory trial (83.0 ± 3.3 minutes and 96.3 ± 4.5 minutes, respectively, p < 0.001).
Average HR was higher in the outdoor trial than the laboratory trial (152 ± 4 and 143 ± 6 b·min−1, respectively, p ≤ 0.05).
Body Weight and Urine Specific Gravity
No difference in USG was noted at the beginning of the laboratory and outdoor sessions (1.016 ± 0.002 and 1.012 ± 0.003, respectively, p = 0.131), indicating that participants arrived for each trial in a similarly hydrated state. Body weight decreased from before exercise to the end of the exercise bouts (82.5 ± 4.8 kg and 80.9 ± 4.7 kg, respectively), regardless of trial (p < 0.001). The body weight loss and controlled water consumption corresponded to no difference in the estimated sweat rate between the laboratory and outdoor trials (1332.5 ± 109.6 and 1506.7 ± 152.3 ml·h−1, respectively, p = 0.28).
Skin and Core Body Temperature
Skin temperature was found to be higher during the laboratory trial than during the outdoor trial (33.0 ± 0.2° C and 31.4 ± 0.3° C, respectively, p < 0.001). Core body temperature was not different between the laboratory and outdoor trials (37.4 ± 0.1 and 37.7 ± 0.2, respectively, p = 0.07). Thus, there was an increased thermal gradient between the core and skin in the outdoor trial compared with the laboratory trial (6.5 ± 0.3 and 4.4 ± 0.2° C, respectively, p < 0.001).
Rating of Perceived Exertion and Attentional Focus
Rating of Perceived Exertion was not different between the laboratory and outdoor trials (13.7 ± 0.3 and 13.7 ± 0.4, respectively, p = 0.877). Likewise, attentional focus was similar between the laboratory and outdoor trials (6.2 ± 0.4 and 5.5 ± 0.4, respectively, p = 0.261).
Much of the current research on cycling has been performed indoors in laboratories, whereas little research has been performed on differences between indoor and outdoor cycling. The novel finding of this study was that power output and HR were higher during the outdoor cycling trial than the laboratory cycling trial despite no differences in perceived exertion. No difference in core body temperature was observed; however, skin temperature was lower during the outdoor trial than the laboratory trial, creating a larger thermal gradient for heat dissipation. Limitations of the study included a relatively small sample size, use of all male participants, and uncontrollable variables during the outdoor trial such as the number of other cyclists and pedestrians on the trail, as well as factors such as wind speed and direction.
The observed 30% increase in power output during outdoor cycling compared with laboratory cycling in this study is in contrast with previous research on indoor and outdoor cycling, which evaluated the reliability of indoor and outdoor 40-km time trial cycling performance (16). This previous study found no difference in the mean power output between 3 indoor trials and 3 outdoor trials (303 ± 35 W and 312 ± 23 W, respectively, p = 0.34). Possible explanations exist for the difference between this and the previous study. The previous study compared time-trial cycling performance, where the participants were asked to complete 40-km in the shortest time possible, achieving the highest average power output. Whereas, this studied focused on an applied approach to training, as participants were instructed to complete 40-km at a typical training ride intensity, rather than an all-out effort. In the previous study, participants were also allowed to view their HR response, elapsed time, and percentage of course completed during the trials. This study blinded the participants to all physiological variables and time elapsed during the trials in an effort to prevent participants from matching power outputs or HR between the 2 trials, as this could skew physiological comparisons. The mean HR in this study was higher in the outdoor trial. This is again in contrast to the previous work comparing 40-km time trial performance between indoor (172 ± 6 b·min−1) and outdoor conditions (173 ± 6 b·min−1), which showed no difference between HR during the indoor and outdoor trials (16). The observation that both power output and HR were higher in the outdoor trial than in the indoor trial in this study is not surprising because increases in the power output lead to increases in the HR. In cycling, the relationship between power output and HR is linear at low to submaximal power output, and curvilinear from submaximal to maximal power output (4). In this study, both the average RPE given by the cyclists (13.7) and the percentage of maximum power in which they completed the ride (57% in the laboratory and 73% outdoors) demonstrate that they were riding at a submaximal intensity. Therefore, the observed increase in HR during the outdoor trial is a consequence of the increased power production. When the present data are taken in context with previous investigations, it is possible that different dynamics between laboratory and outdoor cycling may exist at different exercise intensities.
Weather conditions during the outdoor trial were matched as closely as possible to the indoor conditions. Overall, the environmental conditions of both the laboratory and outdoor trials were similar in all areas (including temperature, relative humidity, heat index, dew point, wet bulb, and barometric pressure) with the exception of wind speed, which was higher in the outdoor environment and nonexistent in the laboratory. Wind direction was not recording during the trials. Although wind direction may have had an impact on the outdoor trial, the out-and-back nature of the course and relatively low wind speeds (2.5 ± 0.6 m·s−1) helped to partially offset this impact (because participants would ride both with and against the wind in equal amounts during the trial). The higher wind speed and physical movement of the cyclist led to more air circulation around the cyclists as they completed the outdoor 40-km ride than during their laboratory ride. This air movement likely contributed to the observed decreased skin temperature in the outdoor trial.
In this study, a greater power output was observed during the outdoor trial than during the indoor trial. This increased power production, and therefore energy production, would increase the rate of heat production in the body. Interestingly, no difference in core body temperature was seen between the indoor and outdoor trials. This finding may be explained by our skin temperature data and the difference in the airflow between the 2 environments. The fact that our cyclists had the same core temperature but different power production between trials implies a greater heat loss to the environment during the outdoor trial.
Skin temperature was cooler outdoors than in the laboratory, potentially because of the presence of the airflow moving over the skin, and thus the convective heat loss. It has been previously reported that male cyclists riding at 60% V[Combining Dot Above]O2max for 60 minutes in the wind (2.55 m·s−1) had skin temperatures that were reduced by about 0.6° C compared with trials completed in no wind (8). Additionally, convective heat transfer has been shown to increase with increased relative air velocity (9). Overall, the combined influence of the larger temperature gradient between the core and skin along with the increased convective cooling capacity in the outdoor trial may explain why our cyclists were able to produce much more power outdoors, while still maintaining a similar core temperature to their laboratory trial where they were exercising at a lower work rate.
It was hypothesized that participants would have a lower RPE and more dissociative thought outdoors than indoors. Although, to the best of our knowledge, these variables have not been investigated during outdoor cycling before, the hypothesis was based on the premise that the outdoor environment would serve as a distraction from exercise, more so than the non-stimulating indoor environment. Distraction during exercise, such as music or video, has often been shown to lead to a lower RPE during exercise sessions of the same intensity (11,19). The findings of this study did not support our hypothesis, as no difference was found between trials in the participants' RPE or attentional focus despite different environments and work rates. It seems that recreationally trained cyclists are familiar with personal levels of perceived exertion and are able to appropriately match levels of exertion on different occasions.
Both associative and dissociative thought can aid in physical activity performance through reduced perceptions of exertion through dissociative thought and the ability to regulate intensity to avoid injury and overexertion through associative thought (7). Our participants experienced a combination of both associative and dissociative thought that remained constant between trials. It is possible that our participants, as experienced exercisers, have each developed a level of associative and dissociative thought they use throughout their exercise as a way to avoid injury and regulate feelings of exertion. Overall, RPE and attentional focus are not able to explain the differences observed in power output between the indoor and outdoor trials.
By instructing our participants to perform these trials at a self-selected training intensity, it allows our results to be applied to typical training situations. Because of the observed higher power output, it may be assumed that our participants received a better training stimulus outdoors than indoors at the same RPE and attentional focus during the 40-km rides. Many athletes may not have access to or use a power meter or HR monitor during training, and therefore rely on perceived exertion to monitor exercise intensity. Because our participants showed no differences in their perceived exertion, yet performed at a higher work rate outdoors, we recommend that when a cyclist has the option to complete a 40-km ride indoors or outdoors, the outdoor environment may be superior to achieve a better training stimulus through increased intensity. Alternatively, if a cyclist must perform their 40-km training ride indoors, they may choose to ride at a higher perceived exertion to account for the potentially lower work rate that may take place, to receive the same benefits as they would completing the ride outdoors. Cyclists and coaches can use these results to assist in planning training schedules, such as planning for indoor workouts to be performed at a higher perceived exertion than outdoor workouts to help achieve the same benefits.
This study was supported by Gatorade Sport Science Institute's Student Grant.
1. Bardis CN, Kavouras SA, Arnaoutis G, Panaqiotakos DB, Sidossis LS. Mild dehydration and cycling performance during 5-kilometer hill climbing. J Athl Train 48: 741–747, 2013.
2. Brown SL, Banister EW. Thermoregulation during prolonged actual and laboratory-simulated bicycling. Eur J Appl Physiol Occup Physiol 54: 125–130, 1985.
3. Duc S, Villerius V, Bertucci W, Grappe F. Validity and reproducibility of the ErgomoPro power meter compared with the SRM and Powertap power meters. Int J Sports Physiol Perform 2: 270–281, 2007.
4. Gardner A, Stephens S, Martin D, Lawton E, Lee H, Jenkins D. Accuracy of SRM and power tap power monitoring systems for bicycling. Med Sci Sports Exerc 36:1252–1258, 2004.
5. Grazzi G, Alfieri N, Borsetto C, Casoni I, Manfredini F, Mazzoni G, Conconi F. The power output/heart rate relationship in cycling: Test standardization and repeatability. Med Sci Sports Exerc 31: 1478–1483, 1999.
6. Groslambert A, Mahon AD. Perceived exertion: Influence of age and cognitive development. Sports Med 36: 911–928, 2006.
7. Lind E, Welch AS, Ekkekakis P. Do “mind over muscle” strategies work? Examining the effects of attentional association and dissociation on exertional, affective and physiological responses to exercise. Sports Med 39: 743–764, 2009.
8. Mora-Rodriguez R, Del Coso J, Aguado-Jimenez R, Estevez E. Separate and combined effects of airflow and rehydration during exercise in the heat. Med Sci Sports Exerc 39: 1720–1726, 2007.
9. Nishi Y, Gagge AP. Direct evaluation of convective heat transfer coefficient by naphthalene sublimation. J Appl Physiol 29: 830–838, 1970.
10. Pennebaker JW, Lightner JM. Competition of internal and external information in an exercise setting. J Pers Soc Psychol 39: 165–174, 1980.
11. Potteiger JA, Schroeder JM, Goff KL. Influence of music on ratings of perceived exertion during 20 minutes of moderate intensity exercise. Percept Mot Skills 91: 848–854, 2000.
12. Quanjer PH, Tammeling GJ, Cotes JE, Pedersen OF, Peslin R, Yernault J. Lung volumes and forced ventilatory flows. Eur Respir J Suppl 6: 5–40, 1993.
13. Russell WD, Weeks DL. Attentional style in ratings of perceived exertion during physical exercise. Percept Mot Skills 78: 779–783, 1994.
14. Shutte J, Townsend E, Hugg J, Shoup R, Malina R, Blomqvist C. Density of lean body mass is greater in blacks than in whites. J Appl Physiol Respir Environ Exerc Physiol 56: 1647–1649, 1984.
15. Siri WE. Body composition from fluid spaces and density: Analysis of methods. 1961. Nutrition 9: 480–491, 1993; discussion 480–492.
16. Smith MF, Davison RC, Balmer J, Bird SR. Reliability of mean power recorded during indoor and outdoor self-paced 40 km cycling time-trials. Int J Sports Med 22: 270–274, 2001.
17. Steed J, Gaesser GA, Weltman A. Rating of perceived exertion and blood lactate concentration during submaximal running. Med Sci Sports Exerc 26: 797–803, 1994.
18. Tammen VV. Elite middle and long distance runners associative/dissociative coping. J Appl Sport Psychol 8: 1–8, 1996.
19. Tan FHY, Aziz AR. Reproducibility of outdoor flat and uphill cycling time trials and their performance correlates with peak power output in moderately trained cyclists. J Sports Sci Med 4: 278–284, 2005.