Not only because of its relation to brain organization and especially hemispheric specialization for spatial and verbal abilities handedness has been an interesting research topic for many years. Far less examined than handedness is the phenomenon of “footedness,” which is considered a purer kind of laterality, because footedness appears to be less influenced by external and social factors than handedness (Gabbard and Hart, 2000). Footedness like handedness can be linked to hemispheric asymmetry and hemispheric specialization, since the motor cortex of each hemisphere controls contralateral limb movements (Bryden, 2000). Measurement of footedness is more difficult than measurement of handedness, since most activities are conducted with both legs involving a mobilization and a stabilization component, while hand activities mostly are scaled one-handed and therefore the dominant hand is considered to be the preferred hand for most daily routine activities like writing (Nicholls et al., 2013). In contrast to this for tests of footedness usually a foot-preference for standard active tasks (e.g., kicking) compared to stabilizing tasks (e.g., one-leg stance) is determined (Stöckel and Carey, 2016). In foot-preference tests the preferential kicking foot seems to hold a similar position like the preferred hand for writing in hand preference tasks (Peters, 1988). 88% of all women and 83% of all men declare their right leg as the “mobile leg” and the left one as stabilizing element and for most subjects handedness and footedness are collateral (Chapman and Chapman, 1987).

In team and combat sports literature often propagates the prevalence of left-handers and left-footed athletes (Grouios, 2004; Loffing and Hagemann, 2012) because of strategic advantages and innate superior abilities that are beneficial for that kind of sports. Moreover the effects of mixed- and left-footedness on motor abilities are consistent with published results on better brain inter-hemispheric communication (Tran and Voracek, 2016). There are many studies concerning the influence of unilateral hand training on the other hand. Most studies concentrate on the impact of a motor training with the non-dominant hand (Porac, 2016). The result of this studies show that right- and left-handers are able to improve the fine motor coordination skills (like writing and finger tapping) of their non-dominant left hand in relatively brief training periods shorter than 1 month (Ackland and Hendrie, 2004; Teixeira and Okazaki, 2007; Walker and Henneberg, 2007; Stöckel and Weigelt, 2012). An EEG study of Lange et al. (2006) indicates a modification of extrinsic coordinates after drawing- training with the non-dominant hand wherefore inter-hemispheric connections transferred via the corpus callosum seem to play an important role.

Concerning the effects of laterality-specific training Healy et al. (1986) as well as Steenhuis and Bryden (1989) exhibited that laterality differences in the range of handedness increase through practicing unilateral motor coordination tasks.

A special result of laterality-specific training is a bilateral transfer whereby the gain of experience of a trained limb has an impact on the untrained contralateral limb. Training conducted with one foot can induce a positive effect on the performance of the untrained limb conducting the same task (Teixeira, 2000). Haaland and Hoff (2003) found a positive effect of non-dominant foot soccer training in 15- to 20-year-old soccer players on both the dominant and the non-dominant foot. According to Kumar and Mandal (2005) bilateral transfer is more distinct from non-preferred side to preferred side and greater with respect to speed, but not to accuracy. Teixeira et al. (2003) examined the influence of bilateral practice on the modification of lateral performance asymmetries in young soccer players and found a reduced asymmetry index for the group which mainly trained with the non-preferred leg. These results indicate that lateral asymmetries arose of previous unilateral training can be modified by bilateral exercise. Therefore especially for young soccer players it seems to be important to enhance motor coordination training with their non-dominant foot.

There is lot of research which shows the relation of motor tasks, particularly motor tasks, which are conducted with hands and different kinds of cognitive activity, especially the influence on mental rotation performance of children and adults (Wohlschläger and Wohlschläger, 1998; Wiedenbauer and Jansen-Osmann, 2008). Besides spatial visualization and spatial perception (Linn and Petersen, 1985) mental rotation, which is defined as the mental representation and rotation of objects (Shepard and Metzler, 1971) is one of the classical tasks for the measurement of visuospatial thinking and is considered to be a prototype for depicting gender differences in spatial imagery (Peters and Battista, 2008). More precisely mental rotation is classified as an intrinsic and dynamic spatial skill (Uttal et al., 2013), which enables the mental visualization and transformation of different classes of objects. In classic mental rotation tasks like Vandenberg and Kuse (1978) usually a target stimulus (cubes, letters, animals, hands) has to be compared with identical or mirror-reversed versions of the upright presented target figure rotated in picture-plane or in depth.

Funk et al. (2005) reason that the processes underlying mental rotation especially the rotation of figures in mind is not an exclusively mental effort, but that it depends on the individual moving body and therefore is interfered with motor processes. A supporting effect of motor gestures on mental rotation performance was found in adults (Chu and Kita, 2011) and children (Ehrlich et al., 2006). A meta-analysis of Uttal et al. (2013) proved stable and transferable improvement of visuo-spatial abilities via special forms of spatial training even if the post-tests were not conducted immediately after finishing the training period. Beside that, the increase of subjects with initially weaker visuo-spatial abilities was larger than for rather skilled subjects.

As mentioned above manual mental rotation training plays an important role in relation to mental rotation. Wexler et al. (1998) as well as Wohlschläger and Wohlschläger (1998) verified the influence of manual motor training on mental rotation performance. In agreement with the assumption of Frick et al. (2005), Wiedenbauer and Jansen-Osmann (2008) proved that for children the success of manual motor training is not only bound to trained object, but that the process of mental rotation itself can be enhanced.

The positive impact of manual motor training on mental rotation performance (Wiedenbauer and Jansen-Osmann, 2008) has been well examined and Voyer and Jansen (2017) revealed an overall advantage of motor experts in spatial tasks. In general athletes show a better mental rotation performance compared to non-athletic subjects (Ozel et al., 2004; Moreau et al., 2012; Pietsch and Jansen, 2012a; Voyer and Jansen, 2017). In detail experts in types of sport which contain mental manipulation like wrestlers and gymnasts have better mental rotation skills than sporting novices or athletes without special requirements in visuo-spatial abilities like runners (Moreau et al., 2011). Jansen and Heil (2010) as well as Jansen et al. (2011) provided first evidence for the positive impact of special motor training through learning to juggle with three balls, while Blüchel et al. (2013) as well as Pietsch et al. (2017) supplied evidence that special motor coordination training on children results in a significantly higher increase in mental rotation abilities compared to a control group with and without physical activity. This fits with the presumption, that the steady learning of new and challenging movements benefits sensor-motor components that are obviously reflected in spatial performance (Pietsch and Jansen, 2012b).

Besides the influence of motor expertise and motor coordination training many studies deal with the connection of mental rotation performance and laterality. Generally spatial and verbal abilities are lateralized to one of the two cerebral hemispheres in most human subjects, even if left- and right-handers show no difference in general cognitive performance (Szaflarski et al., 2006). Spatial and non-verbal skills are ascribed to the right hemisphere, while the left hemisphere is prevalent for verbal skills and language (Hugdahl, 2000). Kolb and Whishaw (2015) proved anatomical differences in brain morphology of right- and left-handers, particularly Jang et al. (2017) found larger basal ganglia in non-right-handers compared to right-handers. Basal ganglia, especially the putamen, seems to be involved in motor preparation and motor performance (Alexander and Crutcher, 1990; Marchand et al., 2008; Stocco et al., 2010). Right-handers’ visuo-spatial skills usually are controlled by the right hemisphere (Vogel et al., 2003), while for left-handers no such general preference or even a small left-hemispheric dominance in frontal and parietal lobes was determined (Shimoda et al., 2008). Concerning mental rotation as a special visuo-spatial skill, both hemispheres are involved in the transformation process (Mellet et al., 2014).

Further it seems to be important, that subjects with very well skilled visuo-spatial abilities show no hemispheric preference while solving mental rotation tasks whereas subjects with minor spatial abilities primarily use their right hemisphere (Vogel et al., 2003). Studies with FMRI of Kucian et al. (2007) revealed a more bilateral activation pattern of adults compared to children while solving mental rotation tasks, which seems to lead to a faster and more effective mental rotation performance.

Summarizing motor training especially motor coordination training which includes a steady learning of new and challenging movements (Pietsch and Jansen, 2012b) improves mental rotation performance and increases hemisphere lateralization (Shimoda et al., 2008). Even bilateral operating methods (Kucian et al., 2007) seem to support mental rotation ability. Further unilateral motor training benefits the development of a greater lateralization of the controlling hemisphere (Steenhuis and Bryden, 1989) and encourages an interhemispheric bilateral transfer (Teixeira, 2000).

Until now, there is no study, which investigated lateralized foot-training and the effect on mental rotation. We hypothesize that young soccer-players which run through a special motor training program for their non-dominant leg show a higher increase of mental rotation performance compared to soccer-players which complete the same training program with their dominant leg.

Twenty secondary school-aged children (1 girl, 19 boys, mean age = 10.60, SD = 0.503), 16 right footers and 4 left footers took part in this study. The children were recruited from a school soccer base camp from a secondary school in Germany and were randomly assigned to experimental (mean age = 10.50, SD = 0.527) and non-experimental conditions (mean age = 10.70, SD = 0.483). All participants are members of soccer clubs and had played soccer for four or more years (M = 5.00, SD = 0.97) and for at least 5 h per week (M = 6.15, SD = 1.04). Both, experimental (1 girl, 9 boys, one left footer) and control group (10 boys, three left footers) consisted of ten pupils. All participants and their parents gave their written consent for participation. The experiment was conducted according to the ethical guidelines of the Helsinki declaration. The mental rotation test and the ZVT we conducted are similar to the type of tasks the children have to solve during math classes, therefore these tests seem harmless and morally inoffensive to us (detailed information is given below). The training program was conducted by two secondary school teachers which are qualified soccer coaches and worked with the children already for more than half a year. Soccer tasks and soccer tests (please see below) are acknowledged and typical forms of soccer training. Ethical approval for this study was not required in accordance with conditions outlined in guidelines from German Research Society (dfg, Deutsche Forschungsgesellschaft) where research that carries no additional risk beyond daily activities does not require Research Ethics Board Approval. We communicated all considerations necessary to assess the question of ethical legitimacy of the study. We assure that our research approach is in line with national and international human research ethics policies and that we exposed and communicated all considerations necessary to assess the question of ethical legitimacy of the study.

The testing of the pupils took place during regular base camp time at school in their familiar classroom and gym. Both groups worked through a demographic questionnaire and the ZVT. After that the mental rotation test and on the following day the soccer tests were completed. During training period both groups had one 30-min practice session per week for 10 weeks during their base camp time. After these 10 weeks, the mental rotation test was conducted again 1 day after the post test soccer session. The mental rotation tests were applied in a group session.

At the beginning the amount of training in years and per week was investigated by two independent t-tests for the factor training group. Then, two univariate analysis of variance with the between-subject factor training group (training of the non-dominant foot, dominant foot) and the dependent variables ‘age’ and ‘ZVT’ were conducted. Furthermore, a 2 (group) ^∗ 2 (time of testing) analysis of variance was computed with the dependent variable ‘number of correct items in the MRT’ and six further 2 (group) ^∗ 2 (time of testing) analysis of variance with the dependent variables “ball control performance,” “dribbling performance” and “shot performance” for the dominant and the non-dominant foot.

The two groups did not differ in their amount of training in years, t(18) = 1.41, p = 0.174 and per week t(18) = 1.342, p = 0.196.

The two groups showed no difference in age, F(1,19) = 0.783, p = 0.388, ηp2 = 0.042, but in ZVT-performance, F(1,19) = 8.517, p < 0.01, ηp2 = 0.321. The ZVT value was higher in the control-group (M = 2.064, SD = 0.28) compared to the Experimental group (M = 1.74, SD = 0.18).

There was no difference between the groups (EG: M = 4.00, SD = 3.19; CG: M = 3.80, SD = 2.39) in mental rotation results in the pre-test, t(18) = 0.158, p = 0.878. The ANOVA revealed a significant main effect of ‘time’ on mental rotation performance, F(1,18) = 12.61, p < 0.01, ηp2 = 0.412, performance in the pre-test was lower (M = 3.90, SD = 2.75) than in the post-test (M = 5.40, SD = 3.23). The main effect was qualified by a significant interaction between the factors ‘time’ and ‘group,’ F(1,18) = 8.07, p < 0.05, ηp2 = 0.310 (Figure 2). The increase in mental rotation ability was larger (d = -0.86) in the non-dominant foot group (post-test M = 6.70, SD = 3.68) compared to the dominant foot group (post-test M = 4.10, SD = 2.18). There was even a higher effect-size (d = 1.27) concerning the comparison of the difference values between the pre- and post-test for each group (non-dominant foot group: M = 2.7, SD = 1.49; dominant foot group: M = 0.3, SD = 2.2). The significant interaction between time and group was also obtained, when the cognitive processing speed was integrated as a co-variate in the analysis of variance mentioned above, F(1,17) = 8.392, p = 0.01, ηp2 = 0.331.

There were no differences between the groups in all six pre-test measurements (all p < 0.01). Because the performance in the soccer test was not the main focus of the study we only present the data without any figures.

This study provides a first insight of the importance of specific training programs with the foot and leg on mental rotation performance. Further studies should comprise larger experimental groups even if a high effect size according to Cohen’s d could be demonstrated. With a larger sample size, the amount of right- and left-footers in each group could be balanced. The small sample size could be responsible for the difference in the ZVT-score, independent of time of testing but in favor for the control group. This difference is not explainable, and should be controlled in further studies with a larger sample size. Beside this, the phenomenon of left- and right-footedness should be considered in more detail, because Jang et al. (2017) assumed differences in motor control based on anatomical differences concerning the volume of basal ganglia and differential manifestation of neural communication (Gorynia and Egenter, 2000; Christman and Propper, 2001). Also the developmental perspective should be investigated in more detail. All children were about 10 years old and were trained in soccer for about 5 years. Therefore they already collected more motor experience with both feet compared to untrained children, which could even have reduced the determined effect.