Fpsyg 08 00212
Fpsyg 08 00212
Fpsyg 08 00212
games and throwing games (invasion games). Among throwing Second Division. Experts showed significantly higher values
games, the technically correct throwing movement is relevant in velocity of throwing without stepping (23,8 ± 1.9 m·s−1
for, and comparable to many sport disciplines such as handball; vs. 21,8 ± 1.6 m·s−1 , p < 0.05) and three-step rhythm
yet it is also fundamental for learning the process of the javelin (25,3 ± 2.2 m·s−1 vs. 22,9 ± 1.4 m·s−1 , p < 0.05) than amateurs
throw. The importance of high level sport-specific motor skills using the same approach. Rivilla-García et al. (2010) found
in team sports have already been formulated by Schack and significant differences in velocity between senior players and U-
Bar-Eli (2007, S. 62). “The success of an individual athlete or a 18 players in different throwing situations: (1) heavy medicine
team is highly dependent on how well the essential techniques ball throw, (2) light medicine ball throw, (3) throwing velocity
of the sport are applied and mastered.” Typical handball specific without opposition, and (4) throwing velocity with opposition.
situations in which the overarm throwing movement comes into Senior players were found to perform significantly better than
effect are 7-m throws, free-throws, backcourt throws, and the the U-18 players in all four throwing situations (p < 0.001;
initiation of fast breaks (Kolodziej, 2010). The overarm throwing t1 = 6.958; t2 = 8.244; t3 = 8.059; t4 = 5.399; df = 92). Studies
movement is also a form related to the jump shot, which is the of quantitative characteristics of throwing within active handball
most important and most commonly used throwing technique in players showed a usually better performance of male athletes than
handball (Wiener, 2011). Hence, this motor skill is an important from comparable female athletes.
part of school and university education, as well as belonging to Qualitative assessments of throwing movement are commonly
club sport. documented by photographs or videos in controlled or game
The accuracy and the velocity of throwing are often used settings, and were often based on the component analysis of
to investigate the differences in quantitative characteristics of Halverson and Roberton (1984). They designed a component
throwing movements. Gender differences in throwing velocity approach, which analyzed the development of the full body
have been identified by Roberton and Konczak (2001). From movement. Therefore the full body movement was segmented in
the ages of 6 to 13, boys performed better compared with partial body movement (see Table 1). A partial body movement is
girls of the same age. In addition, Roberton and Konczak one particular component or joint action of the body (Goodway
(2001) see an increasing amount of difference in the throwing and Lorson, 2008). This approach allows evaluating relevant
performance. At a certain level of development only boys nodes of the throwing movement and a separated classification.
improve their skills, while the girls stagnated at their level, or
became worse. A large meta-analysis of Thomas and French
(1985) recognizes gender differences in throwing accuracy, and TABLE 1 | Component approach.
the velocity already at the age of three. Boys exceed girls in Component Grade Characteristic
throwing velocity from 4 to 7 years of age and throwing distance
at 2 to 4 years of age. Morris et al. (1982) examined the (1) Trunk T0 Varying movement in trunk action
influence of age and gender on the performance of seven basic T1 No trunk action or forward-flexion
motor skills. Two of them (tennis ball throw for distance, and T2 Upper trunk rotation or block rotation
softball throw for distance) were related to throwing movements. T3 Differentiated rotation
Regarding throwing velocity and throwing accuracy, boys at the (2) Humerus H0 Varying movement in humerus action
age of 3 to 6 years were superior to girls of the same age. H1 Humerus oblique
Similar results are described by Vogt (1978). His studies show H2 Humerus aligned but independent
significantly better performance in the throwing accuracy in favor H3 Humerus lags
of boys between 5 and 6 years of age. Within the quantitative
characteristics of throwing movement reported studies (Vogt, (3) Forearm F0 Varying movement in forearm action
1978; Morris et al., 1982; Thomas and French, 1985; Roberton F1 No forearm lag
and Konczak, 2001) consistently emphasize better performances F2 Forearm lag
of male novices. F3 Delayed forearm lag
The quantitative characteristics of throwing movements of (4) Stepping S0 Varying movement in stepping
experts and beginners in handball were analyzed by van den S1 No step
Tillaar and Ettema (2004). They evaluated Norwegian handball S2 Ipsilateral step
players at the age of 20 to 24 years. The task was to throw a ball S3 Contralateral, long step
as hard as possible to a 50 cm × 50 cm target. van den Tillaar and
Ettema (2004) found gender differences in the throwing velocity (5) Backswing B0 Varying movement in backswing
in favor of men, when compared to women (23.2 m·s−1 for B1 Backswing behind or beside the head
men and 19.2 m·s−1 for women). The studies of Rousanoglou B2 Backswing behind or beside the head
Insufficient backward extension
et al. (2015) clarify quantitative differences between experts
B3 Nearly complete backward extension
and amateurs. Amateurs performed with significantly lower
throwing velocity and worse throwing accuracy (p < 0.001). The five components according to the modified component approach (Halverson
and Roberton, 1984) and the main developmentally characteristic levels (T, trunk;
Gorostiaga et al. (2005) found similar results when comparing
H, humerus; F, forearm; S, stepping; B, backswing). Grades from worst (0)
two handball male teams: expert team, one of the worlds to optimal (3) movements. Modified Component Approach by Halverson and
leading team and amateur team playing in the Spanish National Roberton (1984).
The component approach is the most extensive and the most used et al. (2005), Rivilla-García et al. (2010), and Rousanoglou et al.
analysis of throwing movements, (e.g., Roberton and Konczak, (2015). But in the latter studies, only velocity and accuracy
2001; Barrett and Burton, 2002; Hamilton and Tate, 2002; were analyzed. The relationship of quantitative and qualitative
Langendorfer and Roberton, 2002; Ehl et al., 2005; Goodway throwing performance, age, and gender among athletes was
and Lorson, 2008). A more detailed characteristic is described in not evaluated. The present cross-sectional study analyzes the
Halverson and Roberton (1984). qualitative and quantitative throwing performance of male and
Published studies refer to gender differences in component female athletes throughout three age bands.
levels within each age group. For example, Lorson et al. (2013) The goal of this study was to assess whether there
found gender differences among adolescent and young adults; were gender based qualitative and quantitative differences in
Goodway and Lorson (2008) recognized qualitative gender throwing performance of young athletes. It is hypothesized that
differences in game situations of children aged 6 to 8 years. They the qualitative and quantitative performance of the overarm
found significant (p < 0.017) gender differences for step, trunk throwing movement differs between male and female athletes.
and forearm components. In a 3-year longitudinal study, Nelson It was predicted that the quantitative and the qualitative
et al. (1991) investigate 100 children from 5 years of age, and the performance of the throwing movement of male athletes would
gender differences in qualitative and quantitative terms. Nelson be better than female athletes. With focus on five essential
et al. (1991) also reviewed the quality of trunk and the stepping components of action: trunk, forearm, humerus, stepping,
actions and found differences in the trunk and the step in favor and backswing, we furthermore aimed to explore whether all
of the boys. Menzel (1999) analyzes highly skilled female javelin components are equally affected by gender differences.
thrower to identify changes of individual movement patterns
in the javelin throw. He found different movement patterns in
the course characteristics of the knee angle and in the angles MATERIALS AND METHODS
which define the trunk movement, which is associated with
performance level. Results of previous studies mostly suggest that Participants
the qualitative performance of females developed slower than In this cross-sectional study, 96 skilled participants between the
their male counterparts. age of 6 and16 years participated (M age = 11.93, SDage = 3.190;
37 female; 59 male). At the time of examination, all athletes
were members of sports clubs and active handball players.
Instruction to Investigate Kinematics of The handball teams were competitive, sport-oriented clubs.
Throwing Male and female athletes played in the highest leagues of
Different instructions were used to determine quantitative their age groups. Throughout the motor testing all participants
characteristics such as velocity, hardness or accuracy, i.e., “throw were healthy and in good conditions. To conduct our study,
hard” (Halverson et al., 1982, p. 199; Roberton and Konczak, written, full informed consent of the parents (or, alternatively,
2001, p. 94), “each participant performed maximal-effort throws” of the adult participants) were obtained, in accordance with
(Lorson et al., 2013, p. 240), “maximum velocity” (Karadenizlil the latest revision of the Helsinki Declaration of 1975 (World
et al., 2014, p. 21), “throw normally” (Etnyre, 1998, p. 1211) Medical Association [WMA], 2013). Our study was also approved
or “throw the ball softly, increasing the speed until [. . .] near- by the ethics committee (“Ethik-Kommission”) at Bielefeld
maximum speed(s)” (Intermill and Husak, 1984). van den Tillaar University and adhered to the ethical standards of the latest
and Ettema (2006) compared the velocity and accuracy outcome revision of the Declaration of Helsinki. Additionally, and also
of the overarm throw by using different instructions. They found, in accordance with the Helsinki Declaration of 1975 (World
that “no trade-off between speed and accuracy was found for Medical Association [WMA], 2013), was the gathering of
novices [. . .] or experts [. . .] and accuracy did not change personal information such as body size, sex, age, handedness, club
by instruction” (van den Tillaar and Ettema, 2006, p. 503). membership, training age, and training frequency per week. In
Contrary to the quantitative measurements, the instructions are the childhood the handball athletes already had a high amount of
relevant for qualitative measurements (Etnyre, 1998). Burton and training years (M = 3.5, SD = 1.94) and training hours per week
Rodgerson (2003) denote most studies as non-functional when (M = 2.1, SD = 0.969). The average training age of all handball
people are instructed to throw as hard as possible without paying athletes was 5.8 years, and the average training frequency was
attention to the throwing accuracy (Schott, 2010). Therefore, in 5.2 h per week. For data analysis, we formed three different
the present study throwing accuracy and velocity are combined age bands according to levels of development in the motor
in one task. ontogenesis (Meinel and Schnabel, 1998, p. 240): (a) childhood
Summarized gender-specific differences can be seen in (age of female 7 – 12, age of male 7 – 13), (b) pubescence, (age of
throwing velocity, throwing accuracy, and movement quality female 12,1 – 14; age of male 13,1 – 14,5), (c) adolescence (age of
characteristics. In addition, the motor performance increases female 14,15 – 16, age of male 14,6 – 16.
with age and training experience. Most of the studies that
analyzed the throwing movements refer to the novices, (e.g., Task and Procedure
Halverson et al., 1982; Bingham et al., 1989; Hamilton and Tate, After a warm-up of about 15 min, which consisted of different
2002; Lorson et al., 2013). A comparison of experts and novices exercises to prepare throwing movements, the participants were
was made by van den Tillaar and Ettema (2003), Gorostiaga individually tested. The participants who were still being tested
or have already been tested took part in further training. If as being successful. The proof of reliability for the results from
possible, a visual delimination had been included in order to the three different distances (see Figure 1) revealed an internal
avoid participants potentially distracting each other. To assess consistency of α = 0.657. The item-scale-statistic shows a low
the quality and the quantity of movement, the throwing accuracy positive discriminatory power for the first distance (two times
and the throwing velocity was combined in one task. The body height). In addition, the reliability improved (α = 0.741), if
instruction was to ‘throw as hard and accurate as possible.’ After item deleted. Therefore the quantitative performance measured
the movement task was explained, the participants threw out from two times body height was excluded and not used in the
of a 2.5 to 2.5 m tall corridor. The participants aimed at a data analysis.
square target (75 cm × 75 cm) with a self-selected ball. The The qualitative performance was calculated as the mean of
choice between different ball sizes was in accordance to the three independent expert ratings. At the time of the study,
quality standard of the International Handball Federation (00, the experts were current handball coaches with an A- or
0, 1, 2, and 3) and also a tennis ball. It was important that the B-license, and additionally, one expert had been a former
examiner made sure that the participants were able to hold the international handball player. By using the video recording the
ball safely with one hand. The target was placed at shoulder experts had made evaluations in all five components (trunk,
height, and the participants threw three times from three different humerus, forearm, stepping, and backswing) for each participant.
distances, for a combined total of nine throws. The distances Based on Robertson’s component approach (Halverson and
of the participant’s throws were determined by multiplying their Roberton, 1984; see Table 1) for each trial the participant’s
body-height by two, four, and six (see Figure 1). The movement movement was computed for the different components in grades
testing took 5 min. The test set-up was similar to handball in that from worst (0) to optimal (3) movement. This resulted in
the throwing distances are in the range of the free-throw line (7- a performance score for each component. Subsequently, the
m line). The gradations matched the height of the participants reliability of these ratings were calculated. Using Cronbach’s
which allowed for individual testing. This enabled more precise alpha, the internal consistency of the observer agreement was
statements about throwing performances, and compensated for assessed. The internal consistency of the observer is in an
possible deficits, such as strength requirements. acceptable and good range (α = 0.782 (Expert I), α = 0.795
The movement analysis of each trial was recorded with a (Expert II), and α = 0.746 (Expert III)). Then the performance
Sony HDR-CX410VE wide angle lens camera with a high-speed scores of an expert were grouped in one qualitative performance
1/10000 shutter. A side view placed at a 90◦ angle from a distance score. One qualitative performance score represents an expert’s
of 5 m was used. rating across all five components. So, for each participant there
exist three qualitative performance scores. Finally, the qualitative
Data Collection performance was formed by all three qualitative performance
To analyze the quantitative performance, any attempt which hit scores. The evaluation of the qualitative movement refers to this
the target or which touched the mark of the target was considered qualitative performance.
FIGURE 1 | Schematic drawing of the experimental setup of the motor testing and movement analysis from a bird’s-eye view. (x = body height, target
size (a) 75 cm × 75 cm in shoulder height (h), camera at a 90◦ angle at a distance of 5 m to the participant).
Data Analysis F(2,90) = 23.848, p < 0.001. The interaction effect was not
In this cross sectional study, a 3 × 2 age (childhood, pubescence, significant, F(1,90) = 0.582, p = 0.561.
and adolescence) and gender (male and female) one-way analysis To follow-up the main effect of age, separate t-tests
of variance (ANOVA) and a significance level of 0.05 was used were performed (using Bonferroni Holm correction). T-tests
to identify effects of age, effects of gender, and interaction effects for independent samples showed a significant age effect on
in quantitative performance of the throwing movement. For post quantitative performance between childhood and pubescence,
hoc analysis, t-tests for independent samples were conducted. t(67) = −3.738, p < 0.001, between pubescence and adolescence,
A Bonferroni Holm correction was employed to adjust the t(38.67) = −2.586, p = 0.014. Consistently, adolescence
significance level in order to locate significant differences among yielded also better quantitative performances than childhood,
age bands (Holm, 1979). t(69.633) = −7.782, p < 0.001.
FIGURE 5 | Changes of qualitative performance of the humerus action FIGURE 6 | Changes of qualitative performance of the forearm action
(minimum 0 – maximum 3) for male and female athletes as a function (minimum 0 – maximum 3) for male and female athletes as a function
of motor ontogenese (childhood, pubescence, and adolescence). The of motor ontogenese (childhood, pubescence, and adolescence). The
solid line (circles) illustrates the qualitative performance of male athletes. The solid line (circles) illustrates the qualitative performance of male athletes. The
dashed line (triangles) illustrates the qualitative performance of female dashed line (triangles) illustrates the qualitative performance of female
athletes. Standard error marked by the error bars. athletes. Standard error marked by the error bars.
FIGURE 7 | Changes of qualitative performance of the stepping action FIGURE 8 | Changes of qualitative performance of the backswing
(minimum 0 – maximum 3) for male and female athletes as a function action (minimum 0 – maximum 3) for male and female athletes as a
of motor ontogenese (childhood, pubescence, and adolescence). The function of motor ontogenese (childhood, pubescence, and
solid line (circles) illustrates the qualitative performance of male athletes. The adolescence). The solid line (circles) illustrates the qualitative performance of
dashed line (triangles) illustrates the qualitative performance of female male athletes. The dashed line (triangles) illustrates the qualitative
athletes. Standard error marked by the error bars. performance of female athletes. Standard error marked by the error bars.
TABLE 3 | T-tests for independent samples showed effects of gender on five components (trunk, humerus, forearm, stepping, and backswing actions)
according to levels of development in the motor ontogenesis (childhood, pubescence, and adolescence).
(1) Trunk −1.522 (43) p = 0.135 −3.961 (22) p = 0.001 −2.942 (24.897) p = 0.007
(2) Humerus −1.574 (43) p = 0.123 −1.221 (18.268) p = 0.238 −0.257 (25) p = 0.799
(3) Forearm −0.331 (43) p = 0.742 −0.158 (22) p = 0.876 −0.321 (25) p = 0.751
(4) Stepping −2.634 (43) p = 0.012 −3.817 (22) p = 0.001 −1.374 (24.703) p = 0.182
(5) Backswing 0.000 (43) p = 1.000 −2.419 (17.148) p = 0.027 0.000 (25) p = 1.000
TABLE 4 | T-tests for independent samples showed effects of age (childhood to pubescence, childhood to adolescence, and pubescence to
adolescence) on five components (trunk, humerus, forearm, stepping, and backswing actions) involving male athletes.
Component t (df)
(1) Trunk 0.058 (46) p = 0.954 −3.079 (45) p = 0.004 −2.664 (21) p = 0.015
(2) Humerus −3.878 (13.696) p = 0.002 −6.708 (45) p < 0.001 −0.892 (21) p = 0.383
(3) Forearm −8.351 (46) p < 0.001 −14.918 (40.472) p < 0.001 −1.038 (21) p = 0.311
(4) Stepping −3.177 (46) p = 0.003 −3.870 (45) p < 0.001 −0.751 (21) p = 0.461
(5) Backswing −3.432 (14.523) p = 0.004 −3.432 (13.211) p = 0.004 0.000 (21) p = 1.000
TABLE 5 | T-tests for independent samples showed effects of age (childhood to pubescence, childhood to adolescence, and pubescence to
adolescence) on five components (trunk, humerus, forearm, stepping, and backswing actions) involving female athletes.
Component t (df)
(1) Trunk −2.722 (19) p = 0.014 −0.752 (22.930) p = 0.460 −3.011 (26) p = 0.006
(2) Humerus −4.294 (19) p < 0.001 −5.362 (23) p < 0.001 −2.215 (26) p = 0.036
(3) Forearm −5.959 (19) p < 0.001 −7.018 (23) p < 0.001 −0.952 (26) p = 0.350
(4) Stepping −2.141 (10.445) p = 0.057 −3.791 (23) p = 0.001 −2.695 (23.542) p = 0.013
(5) Backswing −1.072 (19) p = 0.297 3.400 (22.921) p = 0.002 −2.712 (24.342) p = 0.012
a humerus action over shoulder height. Within the childhood In fact, in real game situations or other competitions similar
and the pubescence the stepping action of male and female performances are often the results of different motor strategies
athletes were different. Male athletes showed significantly better (Bauer and Schöllhorn, 1997; Preatoni, 2010). In the case of
qualitative movement in stepping action. Female athletes mostly our study, the disparate development of throwing components
showed an ipsilateral step and a contralateral long step. Male between male and female athletes may be justified by this non-
athletes mainly demonstrated a contralateral step. In adolescence, linear point of view. Wagner et al. (2011) analyzed the movement
gender differences are compensated. Male and female athletes variability in various throwing techniques and skill levels in team-
show different developmental progressions in backswing action, handball. Results indicated a decrease in movement variability in
which is in a line with Wagner et al. (2012). During childhood, skilled and highly skilled handball athletes. Furthermore, Button
both showed similar throwing patterns in backswing action. et al. (2003) analyzed the variability of basketball free-throw
In the pubescence, the male athletes developed significantly, action and found that improvement in performance level was
whereas the female athletes stagnated at the level of childhood. correlated with increased movement variability to the end of the
From pubescence to adolescence, only the females improved throwing arm.
their throwing patterns in backswing action. Male athletes, on A closer look at the qualitative movement of each component
the other hand, stagnated at the level of pubescence. During shows that male and female athletes demonstrated different
adolescence, male and female athletes perform the backswing in movement patterns in trunk action throughout all age bands.
the same way. Within the childhood and the pubescence there are also
From a dynamical systems perspective in which a large differences in the course of development in backswing and
number of factors determined the complexity of movement, we stepping actions. From a functional point of view, it is important
may have to rethink the idea of common throwing techniques, to see that the main function phase (humerus and forearm
in particular as regards the development of throwing techniques. actions) of the throwing movement is not affected by gender
differences throughout the different age bands. The present et al. (1994). The representation of gender differences in
results show that only the secondary and primary auxiliary some publications provoked prejudices and gender-stereotypes,
functional phases (stepping, trunk, and backswing actions) are and thus, the expectations in different sports and disciplines
affected. The aim of preparatory phases is to optimize the main (Alfermann, 2009, p. 3). Training intervention and school and
function phase, which is directly related to the overall action university education should equally encourage and support
goal (Göhner, 1992). In game situations, in which handball both sexes in learning complex motor skills, such as throwing
athletes respond quickly and use opportunities to throw, there movements. Also, as Menzel (1999) pointed out, theory
is very little time for placing the right stepping position or and practice should refer to the respective developmental
managing a large backswing. To be successful in a throw on level of children and adolescents which should then lead to
goal, in contrast to the main function phase, these auxiliary individualized instructions.
functional phases play a subordinate role. In addition, the main These results are limited by the controlled setting in which the
function phase of the overarm throwing movement is identical testing was performed. In this well-defined environment, there
to main function phases of various kinds of throwing which were no typical game situations, such as action of opponents,
are used in different game situations, e.g., jump throw or tactical guidance, temporal and spatial pressure situations, or
throwing with run-up. This indicates that the main functional psychological pressure by the audience. The participants could
phase is well established in motor skills of athletes. From concentrate on their attempts with little disturbing influence
a biomechanical perspective throwing, kicking and striking from the outside. Obviously, the throwing situation that was
movements are characterized by the principle of temporal conducted in this study is very different from situations in team-
coordination of single components, which is a proximal-to-distal handball competitions. It is possible that throwing techniques
linkage system. The particular difficulty of throwing movements in a controlled setting may be different in field settings or real
is the optimal coordination of body segments (Hochmuth, 1981; game situations. It remains to be shown, how far fundamental
Hamilton and Tate, 2002). These statements argue in favor of motor skills can be transferred to game situations. This issue is
the cognitive movement representation, in which Basic Action similar to recent studies (van den Tillaar and Ettema, 2004, 2006;
Concepts (BACs) can be associated with main phase or auxiliary Wagner et al., 2011, 2012). Whereas Lorson (2003) suggested
phases and biomechanical aspects (Göhner, 1992; Schack, 2012). that body component levels for the step, trunk, and forearm
actions demonstrated in a real game situation were correlated
with the body component levels demonstrated during practice.
CONCLUSION Therefore subsequent studies could be conducted in team-
handball competitions in a longitudinal design. This may provide
In this study, it can be seen, that with increasing age, valuable insights, but need also many resources. Explorations of
qualitative and quantitative performance of male and female the present study in practice situations and control situations in
athletes improves. The results suggest that there are gender- physical education in the schools are entirely conceivable. For
specific differences in qualitative throwing performance, but not individualized coaching (Menzel, 1999), further research should
necessarily in quantitative throwing performance. A closer look also focus on cognitive measurements of throwing movements.
shows that differences in the qualitative throwing performance The Dimension Analysis-Motoric diagnostics (Schack, 2012)
can only be seen in specific components: trunk, stepping and could provide a possible approach that helps to assess the
backswing actions, and on certain levels of development. Male cognitive movement representations of function phase, and their
and female athletes demonstrated similar movement patterns in role for accuracy of movement execution.
humerus and forearm actions. The forearm and humerus actions
form the forearm whip, which is the main function phase and the
most important component in the overarm throwing movement. AUTHOR CONTRIBUTIONS
Looking to the performance, we recognized that male and female
handball athletes lead to a proficient mover, which in turn speaks MG, DK, and TS developed and designed the study. MG collected
for a purposeful intervention. In addition, it is very important to the data. MG and DK analysed the data. MG, DK, and TS
see that developmental differences are apparently not immutable. interpreted the findings and wrote up the manuscript.
The trunk and the stepping actions represent main faults in the
development of the throwing movement for female athletes. The
main fault amongst male athletes is also the trunk action. FUNDING
Although the results were cross-sectional, this data highlights
the possibility to compensate deficits in throwing performance. This research/work was supported by the Cluster of Excellence
A specific training can apparently alter the throwing pattern of Cognitive Interaction Technology ‘CITEC’ (EXC 277) at Bielefeld
males and females and improve the quantitative and qualitative University, which is funded by the German Research Foundation
throwing outcome, which contradicts the findings of Thomas (DFG).
World Medical Association [WMA] (2013). World medical association declaration Copyright © 2017 Gromeier, Koester and Schack. This is an open-access article
of helsinki ethical principles for medical research involving human subjects. distributed under the terms of the Creative Commons Attribution License (CC BY).
J. Am. Med. Assoc. 310, 2191–2194. The use, distribution or reproduction in other forums is permitted, provided
the original author(s) or licensor are credited and that the original publication
Conflict of Interest Statement: The authors declare that the research was in this journal is cited, in accordance with accepted academic practice. No
conducted in the absence of any commercial or financial relationships that could use, distribution or reproduction is permitted which does not comply with these
be construed as a potential conflict of interest. terms.