The Impact of Cell Phone Use on the Intensity and Liking of a Bout of Treadmill Exercise

This study used a within-subjects design to assess the effect of three common cellular telephone (cell phone) functions (texting, talking, listening to music) on planned exercise. Forty-four young adults (n = 33 females, 21.8 ± 1.3 years) each participated in four, separate, 30-minute exercise conditions on a treadmill in a random order. During each condition, the treadmill speed display was covered and grade was fixed at zero. However, participants were able to alter treadmill speed as desired. Throughout the texting and talking conditions, research personnel used a pre-determined script to simulate cell phone conversations. During the music condition, participants used their cell phone to listen to music of their choice. Finally, participants completed a control condition with no cell phone access. For each condition, average treadmill speed, heart rate and liking (via visual analog scale) were assessed. Treadmill speed (3.4 ± 1.3 miles∙hour-1), heart rate (122.3 ± 24.3 beats∙min-1) and liking (7.5 ± 1.5 cm) in the music condition were significantly (p ≤ 0.014) greater than all other conditions. Treadmill speed in the control condition (3.1 ± 1.2 miles∙hour-1) was significantly (p = 0.04) greater than both texting and talking (2.8 ± 1.1 miles∙hour-1 each). Heart rate during the control condition (115.4 ± 22.8 beats∙min-1) was significantly (p = 0.04) greater than texting (109.9 ± 16.4 beats∙min-1) but not talking (112.6 ± 16.1 beats∙min-1). Finally, liking during the talking condition (5.4 ± 2.2 cm) was greater (p = 0.05) than the control (4.3 ± 2.2 cm) but not the texting (5.1 ± 2.2 cm) conditions. In conclusion, using a cell phone for listening to music can increase the intensity (speed and heart rate) and liking of a bout of treadmill exercise. However, other common cell phone uses (texting and talking) can interfere with treadmill exercise and reduce intensity.


Introduction
According to the Pew Research Center (2013), 91% of American adults and 97% of young adults (age [18][19][20][21][22][23][24][25][26][27][28][29] own cellular telephones (cell phones). Indeed, the device is ubiquitous and experimental design provides evidence that a distraction effect from cell phone use may be the cause of this negative relationship [21]. As more studies indicating relationships between cell phone use and various behaviors emerge, there is a need for studies able to assess the potential causal impact of these devices.
Recently our group began to explore the potential relationship between cell phone use and physical fitness [22]. With its increased functionality, the cell phone provides constant access to entertainment activities traditionally associated with increased sedentary behavior (e.g. surfing the internet, watching videos, playing video games) and sedentary behavior is negatively associated with physical activity and fitness [23][24][25][26]. Conversely, the cell phone is a mobile device and therefore can be used while standing, walking, and during some forms of moderateintensity physical activity (e.g., pedaling a stationary exercise bike). Furthermore, an abundance of cell phone applications have been designed to increase physical activity and fitness. Thus, the relationship between cell phone use and physical activity/fitness may be different than the use of traditional sedentary devices (e.g., watching television). In our previous study, we assessed the relationship between cell phone use and cardiorespiratory fitness (i.e., peak oxygen consumption or VO 2 peak ml kg -1 min -1 ) in a sample of typical college students. Results indicated that VO 2 peak, which was assessed via indirect calorimetry during a progressive treadmill exercise test to exhaustion, was negatively associated with all measures of cell phone use (texting, calls made, and total minutes of use) [22]. This relationship was significant independent of several known correlates of VO 2 peak: sex, percentage of body fat relative to lean mass, and self-efficacy for physical activity [23][24][25][26]. A better understanding of this relationship is important because poor cardiorespiratory fitness is a predictor for a number of health concerns [27], [28].
In an effort to explain this negative relationship between cell phone use and fitness, all participants were interviewed about their daily physical activities, sedentary behaviors, and cell phone use habits. Participants' responses revealed two potential explanations for why cell phone use was negatively associated with cardiorespiratory fitness. First, participants tended to describe cell phone use as a sedentary behavior. Furthermore, high frequency cell phone users, relative to low frequency users, reported engaging in a greater number of sedentary behaviors (e.g. playing video games, watching TV, etc.). Therefore, cell phone use may be a marker for a wider array of sedentary behaviors which research suggests typically occurs at the expense of physical activity [23][24][25][26]. Second, participants identified cell phone use as a potential distraction during planned exercise and described how it may disrupt a given exercise bout. This idea is illustrated in the following statements from two different participants in response to the question: "Thinking about your daily life, would you say that your cell phone use increases or decreases your physical activity?" The participants replied: Maybe decrease. I'm not sure. I mean, sometimes I'll be walking and texting on the treadmill, but it's not like I'm stopping to send a text, I'm still walking. It is the same on the elliptical if I have to send a text. I've kind of gotten used to, like, balancing so that I can text if I need to. Or, if it's too long, like with my mom, then I'll call her and I'll still be going on the treadmill, I might not be going as fast but I'm still going.
It decreases physical activity because, for instance, the other day, one of my friends called me during my work out, and like, I haven't talked to her in a while and I had to tell her a lot of stuff. So it kind of distracted me from my work out.
As these statements suggest, cell phone use during a bout of planned exercise may cause a distraction and potentially reduce exercise intensity. If so, this could limit the ability of an exercise bout to enhance fitness and provide a partial explanation for the negative relationship between cell phone use and cardiorespiratory fitness described above.
Thus, the purpose of this study was to assess the effect of cell phone use on a bout of 30-minutes of treadmill exercise. The study utilized a within-subjects design to compare the intensity (average speed and heart rate) and enjoyment (i.e., liking) of 30-minute bouts of self-selected treadmill exercise during the following conditions: texting on a cell phone (texting), talking on a cell phone (talking), using a cell phone to listen to music (music), and no cell phone use (control). It was hypothesized that using a cell phone to listen to music would result in the greatest average speed and liking of all conditions. This was hypothesized as listening to music has been previously shown to increase exercise intensity and enjoyment [29][30][31][32][33]. Conversely, we hypothesized that the other cell phone use conditions (texting and talking) would distract from treadmill exercise and result in decreased exercise intensity (i.e., reduced speed and heart rate) relative to the control condition. As such, this study is the first to utilize an experimental design to assess the effect of cell phone use on exercise behavior.

Participants
Forty-four college students (n = 33 females, n = 11 males, age 22.0 ± 1.4 years, Table 1) each participated in four separate 30-minute exercise conditions (texting, talking, music, and control) on a treadmill on separate days. The order of these conditions was randomized and each participant completed all four conditions (i.e., within-subjects design). Participants were excluded if they did not own a cell phone or had a cell phone without the ability to send text messages or replay music, and if they had any contraindications to exercise (i.e., orthopedic injuries). Prior to participation subjects were instructed on the benefits and risks of the study and signed informed consent and medical history forms. This study was approved by the Kent State University Institutional Review Board.

Procedures
Participants reported to the Exercise Physiology Laboratory on four separate days. During each visit they completed one of the four exercise conditions (texting, talking, music, and control). This was a within-subjects design as each subject completed all four exercise conditions. Prior to initiation of each 30-minute exercise condition participants were familiarized with the treadmill (Quinton MedTrack CR60, Bothell, WA) and instructed that they would select the initial treadmill speed and were free to alter that speed at any time during the 30-minute session. In other words, if they wished to increase or decrease speed, at any point, participants were free to do so as often as they wished. It was necessary to allow participants to alter treadmill speed as assessing the effect of cell phone use on average speed was a primary purpose of this study. The treadmill grade was set at zero degrees and it was not to be altered. During each of the four exercise conditions time and speed displays on the treadmill were shielded from the participants using an opaque piece of paper. This was done in order to allow the participants to be able to alter their speed, but at the same time be blinded to the display. Heart rate was measured every five minutes during all 30-minute exercise conditions (texting, talking, music, and control). Heart rate was recorded using a heart rate monitor (Polar, Kempele, Finland) and average heart rate was reported in beats min -1 . Upon completion of each condition, liking of the condition was assessed by having participants place a mark on a 10 cm visual analog scale (VAS) anchored by "do not like it at all" on the left and "like it very much" on the right. Liking or hedonics is an affective rating of a behavior that directly correlates with physical activity participation. Average speed was calculated using the following formula: rate (mph) = distance (miles)/time (hours).
During the talking and texting conditions participants were interviewed over the phone (either via text message or conversation) by a member of the research staff. The interview consisted of the following seven topics: movies, songs, food, places to go on vacation, high school classes, hobbies, and places to go with friends on the weekend. For each of these topics the following questions were asked: 10. Would you like to add anything else regarding your favorite (insert topic here)?
The questions were repeated for each of the seven topics until the 30-minute treadmill condition ended. If a participant answered all 10 questions for each of the seven topics before the 30-minute treadmill condition concluded, then the research staff would repeat the questions for each topic until the 30-minute session expired. The research staff did not record the answers to these responses as these responses were considered inconsequential to the purpose of this study. The sole purpose of the interview was to simulate phone conservation (texting and talking). Research staff and participants were placed in separate rooms during the phone interview.
During the music condition participants used their cell phone to listen to music. Participants were instructed that they could listen to any kind of music that they desired. Participants were allowed to self-select their music as we believed this best simulated a real-world use of this cell phone function during exercise. Furthermore, we were concerned that a prescribed playlist of music would not be universally enjoyed by all participants and this lack of enjoyment could negatively affect exercise during that condition. Lastly, while there is evidence that the type (e.g., fast-paced versus slow-paced) of music listened to may differentially affect exercise, multiple studies have indicated that listening to music, regardless of type, has a positive effect on exercise behavior and/or perception [34][35][36][37].
During the control condition participants did not have access to their cell phone nor any interaction with other individuals or electronics and exercised in a quiet, distraction-free room with only research personnel present for supervision. Research personnel present in the room did not interact with the participant during treadmill exercise.

Analytic plan
All data were analyzed with SPSS version 17.0 (SPSS Incorporated, Chicago, IL) with an a-priori α level of 0.05. Males and females physical characteristics (height, weight, age) were compared using independent samples-t-tests. Because there were no differences between males and females and there were no hypotheses based upon sex, it was not included as an independent variable in all subsequent analysis of variance (ANOVA) models. Four condition (texting, talking, music, control) repeated-measures ANOVAs were utilized to examine differences in average treadmill speed, heart rate and liking. Post-hoc analyses for all significant main effects were completed using t-tests with the Benjamini and Hochberg False Discovery Rate correction [39]. NOTE: Differences in estimated relative VO 2 , METs, and kilocalories across all four exercise conditions were not analyzed as each of these variables was ultimately calculated from average treadmill speed. Therefore, analyzing relative VO 2 , METs, and kilocalories using the above ANOVA model would yield the exact same results as the analysis of average treadmill speed. The values calculated from the metabolic calculations (i.e., relative VO 2 , METs, kilocalories) from each exercise condition (texting, talking, music, control) are presented in Table 2.

Physical characteristics
Independent samples-t-tests revealed no significant differences in males and females physical characteristics for height, weight, and age (Table 1). Treadmill speed There was a significant (p < 0.001, effect size (eta-squared (η 2 )) = 0.20) main effect of condition for treadmill speed. Average treadmill speed was significantly greater (p 0.008) in the music condition (3.4 ± 1.3 miles hour -1 ) than all other conditions. Average speed during the control condition (3.1 ± 1.2 miles hour -1 ) was significantly (p 0.04) greater than the texting (2.8 ± 1.1 miles hour -1 ) and talking (2.8 ± 1.1 miles hour -1 ) conditions. Average speeds in the texting and talking conditions were not significantly (p = 0.9) different.

Heart rate
There was a significant (p < 0.001, η 2 = 0.18) main effect of condition for heart rate. Average heart rate was significantly greater (p 0.014) in the music condition (122.3 ± 24.3 beats min -1 ) than all other conditions. Average heart rate in the control condition (115.4 ± 22.8 beats min -1 ) was significantly (p = 0.04) greater than the texting condition (109.9 ± 16.4 beats min -1 ) but not (p = 0.28) the talking condition (112.6 ± 16.1 beats min -1 ). There was a trend (p = 0.07) towards a difference in heart rate between the texting and talking conditions.

Liking
There was a significant (p < 0.001, η 2 = 0.35) main effect of condition for liking. Average liking was significantly greater (p < 0.001) in the music condition (7.5 ± 1.5 cm) than all other conditions. Average liking in the talking condition (5.4 ± 2.2 cm) was significantly (p = 0.05) greater than the control condition (4.3 ± 2.2 cm). Liking in the texting condition (5.1 ± 2.2 cm) was not different (p ! 0.10) than the control or talking conditions.

Discussion
This study utilized a within-subjects design to compare the intensity (average speed and heart rate) and liking of 30-minute bouts of treadmill exercise during four conditions: texting on a cell phone (texting), talking on a cell phone (talking), using a cell phone to listen to music (music), and no cell phone use (control). This design allowed us to test several effects that common cell phone functions may have on a bout of planned exercise. In our previous research, we identified a negative relationship between cell phone use and cardiorespiratory fitness [22]. Interview data collected during that same study suggested a cell phone distraction effect during exercise may partially explain the negative relationship. Present results demonstrated that average treadmill speed and heart rate was significantly greater in the music condition compared to all other conditions. This was an expected outcome as previous research has indicated that listening to music may increase exercise intensity [29][30][31][32][33]. Interestingly, average speed in both the talking and texting conditions was significantly less than the control condition. In addition, average heart rate was significantly lower in the texting condition relative to the control. There was no significant difference between heart rate in the talking and control conditions. These findings suggests that commonly used cell phone functions (texting, talking) have the ability to decrease exercise workload (i.e., average treadmill speed), which could negatively impact fitness. Conversely, two of the cell phone functions tested (talking and music) increased the liking of a bout of treadmill exercise. Results from this study demonstrated that average liking in the music condition was significantly greater than all other conditions. Liking in the talking condition was significantly greater than the control condition, while liking in the texting condition was not significantly different than the talking or control conditions. Thus, using a cell phone only for music increased both intensity (average speed and heart rate) and liking of a bout of treadmill exercise. Talking on a cell phone increased the liking of treadmill exercise but reduced walking speed. Texting, relative to the control condition, reduced speed and heart rate and did not increase liking. This reduction in speed was hypothesized and supports previous findings that used motion capture cameras to assess ground walking (i.e., not on a treadmill) mechanics and speed during cell phone use [15]. In this previous study, the adverse impact of cell phone use on gait mechanics was exacerbated when the task became more difficult, which helps explain why treadmill speed was reduced in both the talking and texting conditions in the present study. The present findings are also in-line with earlier studies which have identified a distraction effect from cell phone talking and texting which negatively impacts well practiced pedestrian behaviors such as crossing the street [7][8][9][10][11][12][13][14]. Taken together, there appear to be positives and negatives to cell phone use during planned exercise. Because liking has been positively correlated with duration and frequency of exercise as well as long-term commitment towards physical activity [40], [41], results suggest that using the cell phone strictly for listening to music, and to a lesser degree talking, may have benefits on the duration and/or frequency of exercise behavior. Conversely, a lack of time is the most commonly identified barrier to exercise participation in modern society [42][43][44]. As a result, exercise often occurs in planned bouts of a fixed duration (e.g., 30-minutes). If the duration of a bout of exercise was fixed, using a cell phone for texting and talking during exercise may reduce workload (i.e., speed) without the possibility of increasing duration. This may provide a partial explanation for the negative link between cell phone use and cardiorespiratory fitness that our group has previously identified.
In addition to the potential implications cell phone use during exercise may have on intensity and ultimately cardiorespiratory fitness, these altered exercise intensities may have important implications for caloric balance. Many individuals utilize exercise as a means of preventing weight gain [45]. Based upon average treadmill speed, caloric expenditure was greater in the music condition (125 kcals 30 min -1 ) compared to all other conditions in the present study. Caloric expenditure in both the texting and talking conditions were the same (110 kcals 30 min -1 each condition), which was significantly less than the control condition (116 kcals 30 min -1 ). Assuming an individual participates in the American College of Sports Medicine's recommended 150 min week -1 of moderate intensity exercise for an entire year and that 3,500 kcal = one pound bodyweight [38], the energy expenditure from each condition for a year would be equivalent to 9.3 lb for the music condition, 8.6 lb for the control condition and 8.2 lb for the talking and texting conditions. In other words, using a cell phone to listen to music while exercising may result in 0.7 lb or 1.1 lb greater energy expenditure per year than not using a cell phone or talking/texting, respectively. Not using a phone at all during exercise may result in 0.4 lb greater energy expenditure per year than talking/texting. These differences may seem small however, weight gain for most individuals is a slow process. American adults gain an average one pound of body weight per year [45]. Over time this slow weight gain can lead to the development of obesity. Therefore, making simple, small changes such as using your cell phone to listen to music and/or avoiding texting or talking on the phone during exercise may help increase exercise intensity and caloric expenditure which could combat this slow weight gain.
One important caveat to these potential differences in caloric expenditure is the fact that despite a significantly slower average speed in the talking condition participants exhibited a heart rate that was not lower from the control condition. This is somewhat surprising as we would expect a reduced heart rate to accompany a reduced workload (i.e., lower speed) as heart rate is a known correlate to exercise intensity [46]. This was the case in the texting condition as participants exhibited both a significantly slower speed and lower average heart rate relative to the control condition. It may be that that the energy requirements for talking and holding the phone to their ear (speaker-phone mode was not allowed) further increased participants' heart rate making the overall intensity of the talking condition similar to the control condition despite a reduced average speed. If this was the case, the metabolic cost of the control and talking conditions may not be dissimilar despite the differences in average treadmill speed.
While we believe this study yields novel and useful information, it is not without limitations. Presently we only examined treadmill exercise, which may not be the exercise mode of choice for all participants, thereby limiting our ability to generalize the current results to other modes of exercise. Requiring participants to exercise on a treadmill may have reduced intensity and liking if this was not the preferred mode of exercise. Additionally, we utilized interviews during the talking and texting conditions that asked questions which were unrelated to the activity the participants were undertaking: exercise. Recent research has indicated that if a cell phone user is participating in an action while interacting with another individual via cell phone that interaction may be less distracting if both cell phone users are aware of the action being participated in [47]. For example, if a coach/trainer were to interact with an athlete via cell phone during exercise and that interaction is relevant to the exercise (e.g., providing instruction/encouragement) that interaction may not distract the athlete from their task. This type of cell phone interaction may have a different effect on exercise behavior than was noted in the present study and is an idea that warrants further research. Another potential limitation of the present study may be related to the music condition. While listening to a self-selected playlist of music likely best mimics a real-world exercise environment, we did not attempt to assess the type (genre, tempo, loudness, etc.) of the music participants listened to. This may have been useful data as music tempo and loudness have been shown to have an effect on exercise intensity [34][35][36]. The sample also consisted of only college students which have been raised entirely in the digital age, while other populations (middle-aged and older adults) may have differing results due to the possibility of having less experience and comfort associated with digital technology, thus limiting our ability to generalize these results to other populations. That said, previous research has measured a significant cell phone distraction effect in early adolescents [11], college students [10], and older adults [7]. Of the various populations studied, the negative impacts appear greatest with older adults as the cognitive decline associated with aging may exacerbate the dual-task effect [7]. Lastly, we estimated caloric expenditure from treadmill speed. This may underestimate actual energy expenditure in the talking condition as it does not account for the participants' caloric cost of talking and holding the phone to their ear. Future studies may wish to utilize other methods (e.g., oxygen consumption via indirect calorimetry) to obtain a more accurate caloric cost of the various conditions.

Conclusions
Cell phones are ubiquitous among today's young adults [48], [49]. However, we are only just now beginning to discover some of the behavioral effects that the use of these devices may have. Because modern cell phones allow individuals to access activities (e.g., browsing the internet, playing video games, watching videos, etc.) that have traditionally been negatively associated with physical activity behavior it is possible that the use of these devices may distract from or interfere with physical activity behavior and planned exercise. Presently, we have demonstrated that using a cell phone for talking or texting during treadmill exercise significantly reduces the workload (i.e., average speed) of that exercise relative to a condition where no cell phone was used. Conversely, using a cell phone to listen to a self-selected playlist of music significantly increased exercise intensity (both speed and heart rate) relative to all other conditions. Relative to the control condition, both listening to music and talking on the phone increased liking of the exercise bout which may have positive effects on exercise duration and/ or frequency assuming the individual has the flexibility in their daily lives to exercise for a longer duration and/or more frequently. In conclusion, relative to a control condition different cell phone functions appear to differentially affect exercise behavior. Listening to music increased the intensity and liking of the exercise bout. Talking increased liking, maintained heart rate but reduced workload. Texting reduced intensity (both speed and heart rate) and did not alter liking. Therefore, the relationship between cell phone use and exercise intensity appears to be specific to the function being utilized on the phone. Individuals should take note of these differences when utilizing their cell phones during their planned exercise time.