Exp Brain Res DOI 10.1007/s00221-016-4587-7
RESEARCH ARTICLE
The effect of 6 h of running on brain activity, mood, and cognitive performance Petra Wollseiffen1 · Stefan Schneider1,2 · Lisa Anne Martin2 · Hugo A. Kerhervé2 · Timo Klein1,2 · Colin Solomon2
Received: 14 October 2015 / Accepted: 30 January 2016 © Springer-Verlag Berlin Heidelberg 2016
Abstract Long-duration exercise has been linked with the psychological model of flow. It is expected that the flow experience is characterized by specific changes in cortical activity, especially a transient hypofrontality, which has recently been connected with an increase in cognitive performance post-exercise. Nevertheless, data on neuro-affective and neuro-cognitive effects during prolonged exercise are rare. The cognitive performance, mental state, flow experience, and brain cortical activity of 11 ultramarathon runners (6 female, 5 male) were assessed before, several times during, and after a 6-h run. A decrease in cortical activity (beta activity) was measured in the frontal cortex, whereas no changes were measured for global beta, frontal or global alpha activity. Perceived physical relaxation and flow state increased significantly after 1 h of running but decreased during the following 5 h. Perceived physical state and motivational state remained stable during the first hour of running but then decreased significantly. Cognitive performance as well as the underlying neurophysiological events (recorded as event-related potentials) remained stable across the 6-h run. Despite the fact that women reported significant higher levels of flow, no further gender effects were noticeable. Supporting the theory of a transient hypofrontality, a clear decrease in frontal cortex activity was noticeable. Interestingly, this had no effect on cognitive performance. The fact that self-reported flow * Stefan Schneider schneider@dshs‑koeln.de 1
Institute of Movement and Neurosciences, German Sport University Cologne, Am Sportpark Müngersdorf 6, 50933 Cologne, Germany
2
Faculty of Science, Health Education and Engineering, University of the Sunshine Coast, Maroochydore, QLD, Australia
experience only increased during the first hour of running before decreasing, leads us to assume that changes in cortical activity, and the experience of flow may not be linked as previously supposed. Keywords Prolonged exercise · Brain activity · Cognitive performance · Flow · Affective state
Introduction Ultramarathon (>42.2 km) running is performed in both recreational and competitive situations. Despite the physiological and psychological stress going along with prologued running and which requires an integrated response, it has been rarely used in scientific research. Accordingly, the specific changes in brain neural activity and the association of these changes with mood, perceived effort, and cognitive performance during ultra-distance exercise, remain unknown. Evolutionary pressures favoured an integration of different (sub-)cortical structures to handle the increasing flow of information, higher cognitive function is organized hierarchically. The frontal cortex plays a major role in integrating these different sub-systems and can be regarded as the constitutive entity for coordinating and integrating emotion, cognition, awareness, working memory, and action (Miller and Cohen 2001), and therefore, is an entity highly involved in processing explicit information (Dehaene and Naccache 2001; Ashby and Casale 2002; Dietrich 2003). In contrast, exercise routines can be regarded as implicitly processed tasks that do not require prefrontal cortex control. Previous research has shown that during moderate-tohigh intensity exercise of up to 60 min, there is a shift of cortical activity away from frontal and prefrontal cortex
13
Exp Brain Res
Table 1 Participant and run characteristics of total distance, average speed, and total test time Participant Gender
Age (year) Total distance (km)
Average speed (km/h)
1 2 3 4 5 6 7 8 9 10 11 Mean
40 49 43 33 39 36 38 40 30 22 37 36.5
62.23 62.96 67.57 61.79 57.39 59.19 56.79 54.57 53.07 67.12 68.55 61.02
11.12 10.82 11.54 10.34 9.80 10.00 9.52 9.31 9.30 11.74 11.56 10.46
7.0
5.31
0.94
SD
F M M F F F M F M M M F: 5 M: 6
Methods Participants and procedures This study was approved by the Ethics Committee of the University of the Sunshine Coast, Australia. Participants had the procedures explained to them and provided written informed consent. The participant group was 11 individuals, who were specifically trained and experienced in ultramarathon running (Table 1).
6 Hour run
regions [1, 2]. This is likely due to an increase in demand of cortical resources in the motor cortex and associated regions (Brummer et al. 2011). This phenomenon, firstly described as a transient hypofrontality (Dietrich 2004), seems to be similar to changes in brain activity described by the mental state termed flow (Dietrich 2004). Flow is a term used to describe a subjective internal state during which a person feels fully immersed in the process of an action (Csikszentmihalyi and Rathunde 1992). Although flow is used as a psychological term in exercise science (Jackson 1996), the linking of a transient hypofrontality with flow allows a neurophysiological approach in which the frontal cortex might play a major role (Dietrich 2004). It has been shown before that an exercise-induced decrease in prefrontal cortex activity remains stable at least 30 min post-exercise (Schneider et al. 2009b) and has a positive effect on mood (Schneider et al. 2009a) and cognitive performance (Schneider et al. 2013). Nevertheless, changes in brain cortical activity, mood, and cognitive performance have only ever been examined post-exercise and, so far, have not been related to a flow state. This study aimed to identify the effect of prolonged exercise, 6 h of running, on the progression of cognitive performance and mood. It was hypothesized that 6 h of Table 2 Study overview, time points of measurements
13
running will result in a decrease in prefrontal cortex activity and that these changes will be correlated with changes in mood, flow experience, and cognitive performance.
The 6-h running (6-h run) was completed on a 1.173-km flat loop within the University of the Sunshine Coast. The participants were instructed to complete the run at selfselected intensity and to drink and eat what and when they liked. The run consisted of six 1-h periods, between which the measurements of EEG, mood, and cognitive performance were performed (Table 2). The time for testing was not included in the 6-h run time and kept to a minimum. Before the commencement of the run, as well as after each hour of running, brain cortical activity was recorded at 500 Hz for 2 min at rest in a seated, eyes-closed position using a commercial dry EEG system (Brain Products, Gilching, Germany), with 16 electrodes on positions Fp1, Fp2, F3, Fz, F4, F7, F8, C3, C4, Cz, P3, Pz, P4, O1, Oz, and O2. The EEG cap was not worn during the running time but was repeatedly applied based on pen-marker positions applied on the forehead, in front of the ears and in the neck, which were applied after the first mounting of the cap. All electrodes were referenced to the right earlobe. A ground electrode was located on the left earlobe. A profile of mood state was recorded pre- and post-running, as well as after the first (1 h), third (3 h) and fifth (5 h) hours to assess individuals’ perceived physical, motivational, and psychological state. Flow state was also recorded at these same intervals. Cognitive performance was measured
Pre
1 h
Cognitive testing Mood assessment
X X
X
Flow state Rest EEG
X X
X X
ERP and RT
X
2 h
3 h
5 h
Post
X
X
X X
X X
X X
X X
X
X
4 h X
X
X
Exp Brain Res
pre- and post-running, and after the second (2 h) and fourth (4 h) hours by a mental arithmetic test (Table 2). This design was chosen in order to minimize the breaks in between the 6-h run. In addition, pre- and post-run a simple reaction time task involving the recording of event-related potentials was performed. Cognitive performance Following previous considerations concerning a decrease in motivation when performing standardized psychological tests within a repeated-measures design (Schneider et al. 2013), it was decided to integrate the Chalkboard Challenge, which is an off-the-shelf brain game performed on an iPod touch. Chalkboard Challenge is commercially available by lumosity.com and is dedicated to assess mental arithmetic (problem solving). Players are presented with two numbers/equations and then have to decide by tapping on the screen which of the two is larger or whether they are equal. With ongoing levels, arithmetic gets more and more difficult [e.g. 16 * (5 − 3) vs. 27]. Time is limited, but being accurate helps to earn more time. Each correct answer earns more points. Once the game is finished, the final score and level of accuracy are presented. To assess the underlying neuro-cognitive parameters, this game was adapted to a simple choice reaction time task to be performed on a PC. Here participants had to use the left and right arrow keys to indicate which of two numbers number the bigger one was. During this assessment, which was performed pre- and post-running, event-related potentials were recorded. Profile of mood state To assess whether any reductions in subject perceived physical, mental, or psychological state occurred during the 6-h run, the MoodMeter® (Kleinert 2006; Schneider et al. 2009a) was used to assess these variables. This methodology has been described in detail elsewhere (Schneider et al. 2008). In brief, the participant was presented with 32 adjectives and was required to indicate whether a given adjective described his/her physical or mental state on a six-point Likert scale from 0 (not at all) to 5 (totally). From the 32 MoodMeter® adjectives, three different dimensions were formed for assessment: perceived physical state (PEPS), perceived psychological relaxation (PSYCHO), and perceived motivational state (MOT). Flow state The nine-item short version of the Flow State Scale-2 [S FSS-2; (Jackson et al. 2008)] was administered before and throughout the 6-h run to assess intensity of flow and any
changes in flow state that may have occurred. The short version of the measure was chosen to reduce response burden upon participants. Each of the nine items is representative of each of the nine dimensions of flow outlined by Csikszentmihalyi (Csikszentmihalyi 1990). Items (e.g. “The way time passed seemed different from normal”) are presented on a five-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). The scale is associated with acceptable goodness-of-fit indices (Jackson et al. 2008), and scores from the scale have been shown to be internally consistent [α = 0.82; (Martin et al. 2006)]. The 36-item Long Flow State Scale-2 [FSS-2; (Jackson and Marsh 1996)] was administered as soon as possible post-run to assess the total flow experience across the full 6-h run. This scale is also a valid and reliable measure of flow and has a mean internal consistency of α = 0.84 (Jackson and Marsh 1996). Both the S FSS-2 and the FSS-2 can provide separate scores across the nine proposed dimensions of flow; however, the global scores for each measure can also be used (Jackson 1999). Although the L FSS-2 was conducted at the end of the run, the 9 Flow items administered throughout the run were extracted to get a post-run flow measure. EEG data processing EEG data were processed using Brain Vision Analyzer 2 (Brain Products, Gilching, Germany). After a first manual artefact detection Butterworth zero-phase filters were applied (Low Cut-off: 3 Hz, Time constant 0.0531 s, 48 dB/ oct High Cut-off: 70 Hz, 48 dB/oct Notch Filter: 50 Hz). Subsequently a systematic protocol for excluding artefacts was performed that included careful visual inspection of all EEG data and automated exclusion procedures, which were set to gradient threshold
RESEARCH ARTICLE
The effect of 6 h of running on brain activity, mood, and cognitive performance Petra Wollseiffen1 · Stefan Schneider1,2 · Lisa Anne Martin2 · Hugo A. Kerhervé2 · Timo Klein1,2 · Colin Solomon2
Received: 14 October 2015 / Accepted: 30 January 2016 © Springer-Verlag Berlin Heidelberg 2016
Abstract Long-duration exercise has been linked with the psychological model of flow. It is expected that the flow experience is characterized by specific changes in cortical activity, especially a transient hypofrontality, which has recently been connected with an increase in cognitive performance post-exercise. Nevertheless, data on neuro-affective and neuro-cognitive effects during prolonged exercise are rare. The cognitive performance, mental state, flow experience, and brain cortical activity of 11 ultramarathon runners (6 female, 5 male) were assessed before, several times during, and after a 6-h run. A decrease in cortical activity (beta activity) was measured in the frontal cortex, whereas no changes were measured for global beta, frontal or global alpha activity. Perceived physical relaxation and flow state increased significantly after 1 h of running but decreased during the following 5 h. Perceived physical state and motivational state remained stable during the first hour of running but then decreased significantly. Cognitive performance as well as the underlying neurophysiological events (recorded as event-related potentials) remained stable across the 6-h run. Despite the fact that women reported significant higher levels of flow, no further gender effects were noticeable. Supporting the theory of a transient hypofrontality, a clear decrease in frontal cortex activity was noticeable. Interestingly, this had no effect on cognitive performance. The fact that self-reported flow * Stefan Schneider schneider@dshs‑koeln.de 1
Institute of Movement and Neurosciences, German Sport University Cologne, Am Sportpark Müngersdorf 6, 50933 Cologne, Germany
2
Faculty of Science, Health Education and Engineering, University of the Sunshine Coast, Maroochydore, QLD, Australia
experience only increased during the first hour of running before decreasing, leads us to assume that changes in cortical activity, and the experience of flow may not be linked as previously supposed. Keywords Prolonged exercise · Brain activity · Cognitive performance · Flow · Affective state
Introduction Ultramarathon (>42.2 km) running is performed in both recreational and competitive situations. Despite the physiological and psychological stress going along with prologued running and which requires an integrated response, it has been rarely used in scientific research. Accordingly, the specific changes in brain neural activity and the association of these changes with mood, perceived effort, and cognitive performance during ultra-distance exercise, remain unknown. Evolutionary pressures favoured an integration of different (sub-)cortical structures to handle the increasing flow of information, higher cognitive function is organized hierarchically. The frontal cortex plays a major role in integrating these different sub-systems and can be regarded as the constitutive entity for coordinating and integrating emotion, cognition, awareness, working memory, and action (Miller and Cohen 2001), and therefore, is an entity highly involved in processing explicit information (Dehaene and Naccache 2001; Ashby and Casale 2002; Dietrich 2003). In contrast, exercise routines can be regarded as implicitly processed tasks that do not require prefrontal cortex control. Previous research has shown that during moderate-tohigh intensity exercise of up to 60 min, there is a shift of cortical activity away from frontal and prefrontal cortex
13
Exp Brain Res
Table 1 Participant and run characteristics of total distance, average speed, and total test time Participant Gender
Age (year) Total distance (km)
Average speed (km/h)
1 2 3 4 5 6 7 8 9 10 11 Mean
40 49 43 33 39 36 38 40 30 22 37 36.5
62.23 62.96 67.57 61.79 57.39 59.19 56.79 54.57 53.07 67.12 68.55 61.02
11.12 10.82 11.54 10.34 9.80 10.00 9.52 9.31 9.30 11.74 11.56 10.46
7.0
5.31
0.94
SD
F M M F F F M F M M M F: 5 M: 6
Methods Participants and procedures This study was approved by the Ethics Committee of the University of the Sunshine Coast, Australia. Participants had the procedures explained to them and provided written informed consent. The participant group was 11 individuals, who were specifically trained and experienced in ultramarathon running (Table 1).
6 Hour run
regions [1, 2]. This is likely due to an increase in demand of cortical resources in the motor cortex and associated regions (Brummer et al. 2011). This phenomenon, firstly described as a transient hypofrontality (Dietrich 2004), seems to be similar to changes in brain activity described by the mental state termed flow (Dietrich 2004). Flow is a term used to describe a subjective internal state during which a person feels fully immersed in the process of an action (Csikszentmihalyi and Rathunde 1992). Although flow is used as a psychological term in exercise science (Jackson 1996), the linking of a transient hypofrontality with flow allows a neurophysiological approach in which the frontal cortex might play a major role (Dietrich 2004). It has been shown before that an exercise-induced decrease in prefrontal cortex activity remains stable at least 30 min post-exercise (Schneider et al. 2009b) and has a positive effect on mood (Schneider et al. 2009a) and cognitive performance (Schneider et al. 2013). Nevertheless, changes in brain cortical activity, mood, and cognitive performance have only ever been examined post-exercise and, so far, have not been related to a flow state. This study aimed to identify the effect of prolonged exercise, 6 h of running, on the progression of cognitive performance and mood. It was hypothesized that 6 h of Table 2 Study overview, time points of measurements
13
running will result in a decrease in prefrontal cortex activity and that these changes will be correlated with changes in mood, flow experience, and cognitive performance.
The 6-h running (6-h run) was completed on a 1.173-km flat loop within the University of the Sunshine Coast. The participants were instructed to complete the run at selfselected intensity and to drink and eat what and when they liked. The run consisted of six 1-h periods, between which the measurements of EEG, mood, and cognitive performance were performed (Table 2). The time for testing was not included in the 6-h run time and kept to a minimum. Before the commencement of the run, as well as after each hour of running, brain cortical activity was recorded at 500 Hz for 2 min at rest in a seated, eyes-closed position using a commercial dry EEG system (Brain Products, Gilching, Germany), with 16 electrodes on positions Fp1, Fp2, F3, Fz, F4, F7, F8, C3, C4, Cz, P3, Pz, P4, O1, Oz, and O2. The EEG cap was not worn during the running time but was repeatedly applied based on pen-marker positions applied on the forehead, in front of the ears and in the neck, which were applied after the first mounting of the cap. All electrodes were referenced to the right earlobe. A ground electrode was located on the left earlobe. A profile of mood state was recorded pre- and post-running, as well as after the first (1 h), third (3 h) and fifth (5 h) hours to assess individuals’ perceived physical, motivational, and psychological state. Flow state was also recorded at these same intervals. Cognitive performance was measured
Pre
1 h
Cognitive testing Mood assessment
X X
X
Flow state Rest EEG
X X
X X
ERP and RT
X
2 h
3 h
5 h
Post
X
X
X X
X X
X X
X X
X
X
4 h X
X
X
Exp Brain Res
pre- and post-running, and after the second (2 h) and fourth (4 h) hours by a mental arithmetic test (Table 2). This design was chosen in order to minimize the breaks in between the 6-h run. In addition, pre- and post-run a simple reaction time task involving the recording of event-related potentials was performed. Cognitive performance Following previous considerations concerning a decrease in motivation when performing standardized psychological tests within a repeated-measures design (Schneider et al. 2013), it was decided to integrate the Chalkboard Challenge, which is an off-the-shelf brain game performed on an iPod touch. Chalkboard Challenge is commercially available by lumosity.com and is dedicated to assess mental arithmetic (problem solving). Players are presented with two numbers/equations and then have to decide by tapping on the screen which of the two is larger or whether they are equal. With ongoing levels, arithmetic gets more and more difficult [e.g. 16 * (5 − 3) vs. 27]. Time is limited, but being accurate helps to earn more time. Each correct answer earns more points. Once the game is finished, the final score and level of accuracy are presented. To assess the underlying neuro-cognitive parameters, this game was adapted to a simple choice reaction time task to be performed on a PC. Here participants had to use the left and right arrow keys to indicate which of two numbers number the bigger one was. During this assessment, which was performed pre- and post-running, event-related potentials were recorded. Profile of mood state To assess whether any reductions in subject perceived physical, mental, or psychological state occurred during the 6-h run, the MoodMeter® (Kleinert 2006; Schneider et al. 2009a) was used to assess these variables. This methodology has been described in detail elsewhere (Schneider et al. 2008). In brief, the participant was presented with 32 adjectives and was required to indicate whether a given adjective described his/her physical or mental state on a six-point Likert scale from 0 (not at all) to 5 (totally). From the 32 MoodMeter® adjectives, three different dimensions were formed for assessment: perceived physical state (PEPS), perceived psychological relaxation (PSYCHO), and perceived motivational state (MOT). Flow state The nine-item short version of the Flow State Scale-2 [S FSS-2; (Jackson et al. 2008)] was administered before and throughout the 6-h run to assess intensity of flow and any
changes in flow state that may have occurred. The short version of the measure was chosen to reduce response burden upon participants. Each of the nine items is representative of each of the nine dimensions of flow outlined by Csikszentmihalyi (Csikszentmihalyi 1990). Items (e.g. “The way time passed seemed different from normal”) are presented on a five-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). The scale is associated with acceptable goodness-of-fit indices (Jackson et al. 2008), and scores from the scale have been shown to be internally consistent [α = 0.82; (Martin et al. 2006)]. The 36-item Long Flow State Scale-2 [FSS-2; (Jackson and Marsh 1996)] was administered as soon as possible post-run to assess the total flow experience across the full 6-h run. This scale is also a valid and reliable measure of flow and has a mean internal consistency of α = 0.84 (Jackson and Marsh 1996). Both the S FSS-2 and the FSS-2 can provide separate scores across the nine proposed dimensions of flow; however, the global scores for each measure can also be used (Jackson 1999). Although the L FSS-2 was conducted at the end of the run, the 9 Flow items administered throughout the run were extracted to get a post-run flow measure. EEG data processing EEG data were processed using Brain Vision Analyzer 2 (Brain Products, Gilching, Germany). After a first manual artefact detection Butterworth zero-phase filters were applied (Low Cut-off: 3 Hz, Time constant 0.0531 s, 48 dB/ oct High Cut-off: 70 Hz, 48 dB/oct Notch Filter: 50 Hz). Subsequently a systematic protocol for excluding artefacts was performed that included careful visual inspection of all EEG data and automated exclusion procedures, which were set to gradient threshold
