Abstract
Preparatory motion (consciously or unconsciously moving a body part just before performing a task) enhances motor performance. Repeated body movements are roughly categorized as rhythmic and discrete. However, the most effective mode of preparatory motion remains unclear. The present study utilized both modes of preparatory motion and compared subsequent performances. Twelve participants executed dart throwing after performing rhythmic, discrete, or no preparatory motion. Performance was statistically evaluated by comparing the error components, assumed contributions of each mode of preparatory motion. The results revealed that the mean value of the error component for the rhythmic mode was significantly smaller than that for the discrete mode. This suggests that the rhythmic mode of preparatory motion produced better dart-throwing performance.
To achieve better motor performance, a person consciously or unconsciously moves all or part of the body repeatedly prior to performing the aiming task, which is called “preparatory motion.” Preparatory motion can be observed in a variety of situations requiring better performance, such as in sports activities, and is empirically known to enhance performance of the task. The effects of executing preparatory motion have been experimentally tested. Uzu, Shinya, and Oda. (2009) tested tennis players in a whole-body reaching task and demonstrated that the preparatory motion of a slight vertical jump shortened the time required to travel a certain distance. Similarly, another study showed that basketball players' preparatory motion of continuous fluctuating body movements diminished the time required to move a certain distance in a whole-body reaching task (Fujii, Yoshioka, Isaka, & Kouzaki, 2013). As shown by those two studies, the execution of preparatory motion is believed to improve the performance of aiming movements. In spite of this, the most effective mode of preparatory motion has yet to be shown.
The modes of preparatory motions used in the previous studies (Uzu, et al. 2009; Fujii, et al. 2013) are distinguished by the characteristics of the motion. In the study of Uzu, et al. (2009), the participants performed preparatory motion in a discrete manner, while the participants in Fujii, et al. (2013) rhythmically continued the fluctuating movements during the phase of the preparatory motion. Human movements executed discretely or intermittently are roughly categorized as “discrete movement,” while continuous movements are roughly categorized as “rhythmic movement.” The functional relationship between rhythmic and discrete motions has been studied previously (for review, Degallier & Ijspeert, 2010). Schaal, Sternad, Osu, and Kawato (2004) questioned if rhythmic movements were repeated discrete movements and studied brain activity during the flexion and extension of the wrist joint. They concluded that, in terms of brain activation, rhythmic movements were not concatenated discrete movements as observed in brain activation. In another classic study (Adam, Vanderbruggen, & Bekkering, 1993), the kinematic parameters of pointing movements with the upper limb were observed. There were differences in peak velocity and movement time between continuous motion conditions and trials where movement was stopped at the target position. From these results, although rhythmic and discrete movements appear similar (with the exception of stopping vs. continuous motion) they could be quite different in nature. Therefore, in this study the effect of rhythmic preparatory motion on motor performance was compared to the effect of discrete preparatory motion.
Dart throwing was the task selected to examine the different effects that each mode of preparatory motion had on motor performance. Dart throwing was chosen in part because of its simplicity. The movement of dart throwing is mostly executed by a single upper limb without involving whole body movements or translation and excludes the complexity created by motion of multiple joints. For the same reason, some previous studies have chosen dart throwing to evaluate factors affecting motor performance not only in the field of human motion but also in attention studies (Marchant, Crough, & Crawshaw, 2007; Mohsen, Geoffrey, & Abbas, 2013;. This study assumed that any differences in performance were attributable to the mode of the preparatory motion. The purpose of this study was to show how different modes of preparatory motion affect subsequent motor performance.
Hypothesis. The performance of dart throwing is hypothesized to be affected by the mode of preparatory motion.
Method
Participants
Twelve healthy university students (7 men, 5 women; M age = 22.3 yr., SD = 1.5), who were novices in dart throwing and were blind to the nature of the study, participated as volunteers in this study. All the participants were right-handed according to scores on the Edinburgh Handedness Inventory (Oldfield, 1971). Prior to starting the experiment, the procedures were orally explained to the participants, and all participants provided informed consent. The study was conducted in accordance with the revised Declaration of Helsinki and was approved by the appropriate ethics review board of the authors' institution.
Apparatus
A square wooden board (48 cm high × 42 cm wide) with a filled 3.3 cm diameter circle as the target was mounted on the wall with the target's center 173 cm above the ground. A 3 cm grid on the board was used to measure throwing performance. For each column of the grid, positive numbers (1 to 7) were incrementally assigned from the center intersecting point toward the right edge, while negative numbers (−1 to −7) were decrementally assigned leftward. In the same way, positive numbers (1 to 8) were incrementally assigned to each row from the center intersecting point toward the top edge, while negative numbers (−1 to −8) were decrementally assigned downward. A boundary line where the participants stood was marked on the floor 237 cm from the wall. A video camera was set about 3 m from the participants to record the throwing movement to count the repetitions of the preparatory motion in each trial.
Procedures
The participants stood on one side of the boundary line opposite to the target with the right foot on the line. They were each given a dart (15.7 cm in length, weighing 20.0 g, with a metal tip and body and plastic blade), which they held in their right hands in a comfortable way. The participants were asked to try to hit as close as possible to the target's center. Before throwing the dart, the participants aimed for the target while executing the preparatory motion of the right arm. They were asked to extend and flex their right elbow joint repeatedly, either rhythmically or discretely; or not to make any preparatory motion. If preparatory motions were required, the forearm moved back and forth along the direction of the throw (Fig. 1). The movement of each elbow extension and flexion was almost spatially identical in the rhythmic and discrete conditions, but the temporal property was configured differently by the insertion of stop and hold periods in discrete mode.
The arm movements in the two preparatory conditions: (a) rhythmic and (b) discrete motions.
For the rhythmic mode of preparatory motion (Fig. 1a), the participants were instructed to move the right arm back and forth continuously along the rhythm of a metronome beep at a frequency of 1 Hz (van der Wel, Sternad, & Rosenbaum, 2009). As a result, one cycle of the arm movement in rhythmic mode took 1 sec. For the discrete mode of preparatory motion (Fig. 1b), the participants were instructed to move the arm back and forth along the rhythm of the metronome at a frequency of 2 Hz while inserting a 0.5 sec. stop and hold at the distal and proximal arm positions. As the result, one cycle of the arm movement for the discrete mode of preparatory motion took 2 sec. This was double the duration of the rhythmic mode, but the actual time of the arm movement for one cycle was still 1 sec., identical to the rhythmic condition.
Prior to the actual performance, the participants were required to practice throwing in the three conditions (each 10 times). For the actual trials, the participant held the initial stable posture with the elbow joint submaximally extended for 2 sec. Subsequently, the participants executed one of the three conditions of preparatory motion while aiming at the target. When the participants felt ready, they threw the dart. The number of repetitions of preparatory motion was not regulated and was arbitrarily determined by the participants so they could concentrate on aiming at the target. The participants repeated 15 trials in one block in one condition, and then another block with the same condition was repeated. After two blocks, the participants performed two blocks of another one of the two other conditions, and finally performed two blocks of the third condition. As the result, the participants engaged in 90 trials in total (three conditions thirty times each). The order of the three conditions was counter-balanced across all participants. To avoid fatigue, 3 min. breaks were inserted between the blocks.
Analysis
To assess the performance of the dart throwing, the row and column numbers of the grid where the dart tip landed on the board were written down for every trial. The evaluation was done by calculating the distance from the center of the target to the center of the grid where the dart landed. The error distance (ED) was calculated as the square root of the sum of vertical and horizontal distances. The means and standard deviations of the ED for all trials in each condition for each participant were obtained. The mean ED in rhythmic, discrete, and no preparatory motion conditions were defined as ED-r, ED-d, and ED-n, respectively. ED-n was free from the effect of executing preparatory motion; i.e., ED-n was assumed to reflect the pure error derived from the throwing motion itself. On the other hand, ED-r and ED-d were assumed to contain the component of ED-n plus the error component due to the effect of executing rhythmic (R-error) or discrete (D-error) preparatory motion, respectively. To calculate R- and D-error, ED-n was subtracted from ED-r and ED-d, respectively, as shown in the following equations.
To determine the approximate time required for the preparatory motion (P-time) in both rhythmic and discrete conditions, the number of cycles repeated for the preparatory motion was counted by observing the video recording. Based on the time required for one cycle of the rhythmic condition (1 sec.), P-time for the rhythmic condition was calculated by number of preparatory movements multiplied by 1 sec. In the same way, P-time for the discrete condition was calculated as the number of preparatory movements multiplied by 2 sec.
For the statistical analysis, Kolmogorov-Smirnov tests were applied to test the normality of all variables' distributions; all variables measured and calculated were normally distributed. A paired-samples t test was used to compare the R-error and D-error. The level of significance was set to 0.05.
Results
Each participant completed 90 trials of dart throwing, and EDs were successfully obtained from all trials. Descriptive data for ED obtained across all participants are shown in Table 1; The smallest mean value of ED was observed in the condition of the rhythmic mode of preparatory motion, and the largest was seen in the discrete mode, while the mean value of ED with no preparatory motion was between the two preparatory motion conditions. The means of R-error and D-error, calculated by subtracting ED-n from ED-r and ED-d and assuming pure error components due to the preparatory motion, are shown in Table 1 and Fig. 2. The difference between R-error and D-error was moderate (t11 = −2.40, p<.05, d = 0.40). The result indicated that the R-error was significantly smaller than the D-error. Therefore, the rhythmic mode of preparatory motion produced a better dart-throwing performance than the discrete mode.
The mean and standard deviation of the error distance obtained from the dart throwing performance in three preparatory motion conditions, rhythmic, discrete, and control (no preparation)
Error components induced by rhythmic (R-error) and discrete (D-error) mode of the preparatory movements. Error bars represent standard deviations. *p <.05.
The P-time, indicating the time required for the preparatory motion, was successfully measured from videos of 705 out of 720 trials, while in the other 15 trials reliable data could not be obtained for technical reasons. Descriptive data for P-time for the conditions of rhythmic and discrete modes of the preparatory motion are shown in Table 1; the difference was not significantly different for the two preparatory motion conditions (t11 = −1.74, p =.11, d = 0.30).
Discussion
The purpose of this study was to show how the different modes of preparatory motion affect subsequent motor performance in dart throwing. The result showed that the value of the R-error due to rhythmic preparatory motion was significantly smaller than the value of the D-error due to discrete preparatory motion. This finding suggested that the mode of the preparatory motion affected the dart-throwing performance.
In order to understand why different effects were elicited by different modes of preparatory motion, several possible factors were taken into consideration. First, the result obtained could be partially, if not completely, explained by the cognitive function known as “attention.” Attention is roughly divided into two types, “internal” and “external,” according to the target of focus (Wulf, 2007; Maas, Robin, Austermann Hula, Freedman, Wulf, Ballard, & Schmidt, 2008). Several previous studies have actually shown that the place where attention was focused, internally or externally, indeed influenced the motor performance of dart throwing (Marchant, et al., 2007; Mohsen, et al., 2013). Along with the results from those studies, it is believed in the sports psychology field that too much attention paid internally to one's own body or to the production of the movement deteriorates performance, while externally focusing on the target is believed to be more advantageous. The mechanism underlying this phenomenon could be also applied to the dart-throwing performance in this study. In the rhythmic mode condition, the arm movement was more like a spontaneous movement, requiring less voluntary control compared to the discrete mode. As a result, the participants' attention could be more externally focused on the target during the rhythmic mode of the preparatory motion, while the focus shifted more internally to the body or to the production of the movement during the discrete mode.
Although the comparison between the R- and D-errors did reveal a significant difference (Fig. 2), the measured error distances both in the rhythmic (ED-r) and discrete (ED-d) modes were not significantly different from the control (ED-n) (data not shown). Since the execution of preparatory motion was expected to enhance the dart-throwing performance, possibly some hidden factor suppressed the effect. The exploration of the limiting factor led to consideration of the possibility that the rhythm of the arm movements counteracted the advantageous effects of the preparatory motion. The participants were instructed to synchronize the rhythm of their arm movements with the metronome in order to have identical conditions for obtaining reliable data of the effects of preparatory motion. Restricting arm movements to a certain rhythm might have unexpectedly required large attentional resources. That, in turn, may have shifted some attention from aiming at the target to following the rhythm of the metronome or to adjusting arm movements to the rhythm, which would have attenuated the external focus on the target.
It is well known that repetitive body movements, as seen in locomotion, involve activation of the spinal neural system called “central pattern generator” (for review, see Kiehn, 2006; McCrea & Rybak, 2008; Guertin, 2013). A typical functional property of the central pattern generator is the cyclic repetition of movements in a certain rhythm with reduced recruitment of voluntary controls. Therefore, the involvement of the central pattern generator potentially reduces the attentional requirements for executing intended movements. Previous studies showed that the central pattern generator controls not only lower limb movements, as seen in locomotion, but also upper limb movements (Schaal, et al., 2004; Zehr, Carroll, Chua, Collins, Frigon, & Haridas, 2004; Smits-Engelsman, Swinnen, & Duysens, 2006; White, Bleyenhent, Ronsse, Smith, & Thonnard, 2008). These results lead to an assumption that the rhythmic preparatory motion in this study would have involved activation of the central pattern generator to some extent. If this is true, less attention was directed to production of the preparatory motion, which in turn avoided the internal focusing of the attention that would have attenuated the performance. This might partially explain the reason why different effects were observed between the conditions of rhythmic and discrete modes of preparatory motion.
The aiming phase prior to dart throwing is empirically known to be necessary for smooth transition to the performance, and it was experimentally elucidated that the viewing time prior to the target-aiming task affected the performance in pistol shooting (Ravindra, Errol, & Wing, 2009). In that study, Ravindra, et al. examined the relationship between the viewing time of the target and performance and reported that the best performance was reached with 3 sec. or more viewing time in novice shooters. In this study, means of the P-times, representing time for the preparatory motion, were greater than 4 sec. (4.33 sec. and 5.19 sec. for rhythmic and discrete mode conditions, respectively). Therefore, the time required for the preparatory motion in this study was considered sufficient to induce the better performance and was assumed not to be the regulating factor of the performances.
In conclusion, this study demonstrated that preparatory motion affected the subsequent performance of dart throwing depending on its mode, rhythmic or discrete. The rhythmic preparatory motion induced better performance than the discrete mode. Since the kinematics of the upper limb movements were not explored in this study, observation of the direct relationship between preparatory motion and performance is encouraged in the future for the further understanding of this topic.