Both Master s thesis students had an equal contribution in the accomplishment of this Master s thesis.

Transcriptie

1 Acknowledgement We would like to extend our thanks to our promotor Prof. dr. Raf Meesen and our copromotor dr. Koen Cuypers for their help with achieving this Master s thesis. We would also like to thank the PhD student drs. Asif Jamil to help us conducting the experiment together with KULeuven and the Uhasselt. We are also very grateful for the support of our family. Trevierenlei 6, 2900 Schoten, België A. D. Tervoortstraat 7A, 3830 Wellen, België L. L. June 6th, 2017

2

3 Research context The content of this duo-master s thesis part 2 of Master students of Rehabilitation Sciences and Physiotherapy is situated in several research domains. Firstly, it is situated on a combined domain of neurological and musculoskeletal research. The reason for this is the implementation of a bimanual motor coordination (BMC) task, where stimulation is given to specific brain areas. Secondly, this duo-thesis focuses on the geriatric research domain as well, because young adults were compared to older adults. This duo-master s thesis part 2 is part of a research study that was conducted by Prof. dr. Raf Meesen, dr. Koen Cuypers and drs. Asif Jamil. This research was realized on the campus KULeuven at Heverlee in the building De Nayer. The apparatus, where the measurements were carried out, was also developed in the building De Nayer. Our duo-master s Thesis part 1 was a literary study in which the role of the white matter structure corpus callosum (CC) during BMC tasks in patients with Multiple Sclerosis (MS) was investigated. The CC specifically was investigated because of its crucial role concerning interhemispheric communication, which in turn had a great part in BMC. The result of our duo-master s thesis part 1 was that BMC task performance was affected by degeneration of the CC. However, this degeneration of the CC is not only present in patients with MS, but also in older adults (Fling et al., 2011). Therefore, within the capacity of our research we were keen to come to a consensus concerning the BMC reduction in humans of elder age, in our duo-master s thesis part 2. Among BMC tasks belong the various tasks of daily life such as tying your shoelaces, buttoning your shirt, eating with knife and fork, etc. The relevance of this research is nearly entirely self-explanatory as there is an increase in aging in the society. Therefore, it is understandable that there is an increased need for care. Older adults, who have more problems with motor function, are in more need of care (Heuninckx, Wenderoth, & Swinnen, 2010). In light of this, one must therefore state it is in the best interest to not only investigate the motor functions of the older adults, but to indeed show regards for retaining motor function possibilities. Electro-stimulation could inherently be seen as one of such. 1

4 This is where the involvement of the physiotherapeutic matter becomes adamant. Anticipation and activation could therefore very well be the foundation of the physiotherapist s contribution to an aging community, resulting in socially independent elder beings. In specific terms, we can conclude that in this duo-master s thesis part 2 the difference in BMC task performance between younger and older adults and the effect of stimulation will be investigated, as well as the difference between non-feedback and feedback performances. The contribution of the Master s thesis students is as follows: The research is part of an ongoing project. The research design and methodology have been conducted by Prof. dr. Raf Meesen, dr. Koen Cuypers and drs. Asif Jamil, which was not a part of our own investigatory capacity. The recruitment had been performed entirely by the Master s thesis students, with the support of dr. Koen Cuypers at the start of the recruitment. During the experiment, the subjects were also accompanied by the Master s thesis students. The data-acquisition was conducted by Prof. dr. Raf Meesen and drs. Asif Jamil, without the cooperation of the Master s thesis students. With the help of dr. Koen Cuypers, the Master s thesis students were able to thoroughly analyze the acquired data. The academic writing process was completely carried out independently by the Master s thesis students as well. Both Master s thesis students had an equal contribution in the accomplishment of this Master s thesis. 2

5 Reference List Fling, B. W., Walsh, C. M., Bangert, A. S., Reuter-Lorenz, P. A., Welsh, R. C., & Seidler, R. D. (2011). Differential callosal contributions to bimanual control in young and older adults. J Cogn Neurosci, 23(9), doi: /jocn Heuninckx, S., Wenderoth, N., & Swinnen, S. P. (2010). Age-related reduction in the differential pathways involved in internal and external movement generation. Neurobiol Aging, 31(2), doi: /j.neurobiolaging

6 4

7 Bimanual motor coordination in younger and older adults A physiotherapeutic view

8 6

9 Abstract Background: Evidence has revealed that in older adults motor function declines are due to several ageing processes in the brain. The independency of older adults depends on their ability to perform activities of daily life (ADL) and since bimanual motor coordination (BMC) is a crucial component for ADL, such as buttoning your shirt, it is a remarkable part to investigate. As physiotherapists, it is interesting to know whether or not electro-stimulation has an influence on the performance of a complex bimanual task. Also, the effect of adding visual augmented feedback could be of importance during physiotherapy sessions. Objectives: The aim of this study was to investigate the difference between younger and older adults on the performance of a complex BMC task. Also, the effect of stimulation, the influence of feedback and the influence of complexity was investigated. Participants: Twenty healthy younger subjects (mean age: years, 10 females) and 20 healthy older subjects (mean age: years, 7 females) participated in this study. Measurements: The point angle difference (PA) and the track target difference (TT) were considered as outcome measurements to evaluate the performance. Results: Concerning the difference between younger and older adults on the performance of a complex BMC task it is discovered that older adults perform significantly worse on both parameters than younger adults. Stimulation has a significant positive influence on the performance concerning parameter TT. In both parameters PA and TT, a significant positive influence of feedback is revealed. There is also an effect of difficulty in both parameters. Conclusion: Taken together, the results suggest that older adults performed the BMC task generally significantly worse than younger adults, which is an indication to perform physiotherapy with older adults. 7

10 8

11 Introduction It is generally known that motor function declines during the lifespan (Jimenez-Jimenez et al., 2011; Lin et al., 2014; Panouilleres, Joundi, Brittain, & Jenkinson, 2015; Shetty, Shankar, & Annamalai, 2014; Ward, 2006). The underlying mechanisms that play a key role during the ageing process are: (1) atrophy of the grey and white matter volume in movement related cortical areas (Maes, Gooijers, Orban de Xivry, Swinnen, & Boisgontier, 2017), (2) quantity and quality diminishment of the white matter, such as a disruption of the white matter integrity (Ward, 2006). The largest white matter structure in the brain is the corpus callosum (CC) and is essential for interhemispheric communication, which is of importance during bimanual coordination. However, in elderly is known that aging is associated with a decreased size and integrity of the CC (Fling et al., 2011). (3) Changes in corticospinal excitability and inhibition (Corti et al., 2017). Of particular interest is the decline in bimanual coordination during healthy aging as intact coordination is crucial for independence during daily living (Heuninckx, Wenderoth, Debaere, Peeters, & Swinnen, 2005; Hummel et al., 2010). Motor function, in general, can be divided into fine and gross motor skills. Bimanual motor coordination (BMC) is an example of fine motor skills and is crucial for successfully performing activities of daily living (ADL), such as for example tying your shoelaces, buttoning your shirt, eating with knife and fork. Not only the previous examples of ADL tasks are important in daily life but also more complex coordination tasks, which require a cognitive aspect, such as gardening, manual shifting during driving and gaming (Hummel et al., 2010; Zimerman et al., 2013). Because the percentage of the aging population is increasing, healthcare cost will increase and independence in daily life is strongly determined by the ability to rely on intact BMC, it is desirable to focus our research on older adults (Maes et al., 2017; Zimerman et al., 2013). Previous work already demonstrated that if a more complex motor task was performed by older adults, an age-related motor problem became visible (Heuninckx, Wenderoth, & Swinnen, 2010; Zimerman et al., 2013). For this purpose, a bimanual tracking task with varying difficulties was selected. 9

12 Moreover, as compared to younger adults, healthy elderly showed over-activation of certain brain areas (Heuninckx et al., 2005). Two possible explanations for over-activation of certain brain areas during a specific task are proposed by the literature, namely compensation or dedifferentiation hypothesis. In the case of compensation, the task performance is improved by activating more or different brain areas. If the task performance is worsened by overactivation of certain brain areas, the over-activation is due to dedifferentiation (Heuninckx et al., 2010). Currently, it is acknowledged that the compensation hypothesis is applicable in older adults compared to younger adults (Ward, 2006). The goal of the current study was to investigate whether or not these age-related motor problems could be alleviated by applying non-invasive brain stimulation to the motor cortex while performing a more complex fine motor task. Hence, in this study transcranial Direct Current Stimulation (tdcs) was used. Besides evaluating the effect of tdcs during motor performance, the effect of feedback will also be considered. A distinction can be made between externally and internally guided movements, whereby e.g.: during visually and somatosensory guided tasks different areas are activated (Heuninckx et al., 2010). Amongst augmented feedback, we comprehend offering external information during the motor performance and might cause an influence on the future behavior (Beets et al., 2015). It is known that healthy subjects have a better motor performance by adding visual feedback, furthermore older adults gained even more benefits from augmented visual feedback compared to younger adults (Heuninckx et al., 2010). This specific topic is related to physiotherapy because a physiotherapist often provides feedback to the patient. If we can confirm this statement of Heuninckx et al. (2010), the physiotherapist should consider whether it is possible to provide this information when giving a home exercise to the older adult. 10

13 In addition to the vast amount of research that s already available on the BMC performances in older adults, the following specific questions were addressed. First, we verified the literature if our task was appropriate to detect motor performance deficits in older adults in general. In view of this confirmation, we hypothesize that older adults will show decreased BMC compared to young adults. Additionally, an easier and a more difficult coordination pattern was provided to investigate the influence of task complexity. If the cause of a motor deficit is present during the task, anticipation is possible through an approach of evaluation and rehabilitation (Shiffman, 1992). As demonstrated by the study of Sisti et al. (2011) these motor deficits are more prominent with a higher task complexity level. Using this specific approach, we investigated whether or not a difference between young and older adults is present considering this task complexity. Secondly, we investigated whether or not stimulation on the right primary motor cortex (M1) had an influence on the performance and if elderly, in particular, would benefit from this stimulation. We hypothesize that, based on the compensation hypothesis (Heuninckx et al., 2010), the real stimulation will have a significant positive influence on the performance in older adults compared with younger adults relative to sham stimulation. In lieu of the ceiling effect, a positive tendency is possible in younger adults (Hummel et al., 2010; Panouilleres et al., 2015; Vancleef, Meesen, Swinnen, & Fujiyama, 2016). Lastly, we explored whether or not there would be a difference in performance between younger and older adults with and without feedback. We gauge, based on the guidance hypothesis, that older adults will have a significantly better performance with augmented feedback (Beets et al., 2015). The study of Salmoni, Schmidt, and Walter (1984) supports this hypothesis, which predicts that performance would deteriorate without feedback. 11

14 12

15 Methods Participants Forty healthy subjects were recruited from a list with contact details of subjects who participated in a previous study about motor coordination. Participants were also recruited by means of social media or through family members and friends. Flyers were hung in the building de Nayer for recruitment (see appendix 1). All subjects were personally called and received an explanation about the study. We informed the subjects about their compliance as to the participation in this study and their availability concerning 2 times during 2 hours with a spread of days. During this phone call, the inclusion and exclusion criteria were discussed with the subjects. After criterion upholding the collection of personal information was headed. Afterwards, the participants received an with the important details of the study and a description of the location. Twenty healthy younger subjects (mean age: years, 10 females) and 20 healthy older subjects (mean age: years, 7 females) were included in this study. Inclusion criteria were healthy female and male participants of age between and years old. See table 1 for an overview of characteristics of the subjects. Inquiries about the dominance of hand and capacity of eyesight were made. Concerning transcranial Direct Current Stimulation (tdcs) defined exclusion criteria were applied: metal present in the head, pacemaker, wires, implantable defibrillator, bio-stimulator or TENS-apparatus, aneurysm clips, intracranial clips, implanted medication pump, shrapnel (anywhere in the body), being a metal worker in the past and/or having one of the following illnesses in the past: cerebral thrombosis, cerebrovascular accident, head trauma, meningitis, long period of unconsciousness (>1h), migraine, epilepsy), prehistory of brain surgery, family history of therapy-resistant epilepsy, history of active use of a defined matter during last year (tattooed eyeliner, pregnancy). Table 1: overview of the distribution between males and females with their mean age Male Female Young 10 (23.6) 10 (23.1 years) Old 13 (71.31 years) 7 (69.71 years) 13

16 All the subjects were from Flanders, Belgium. Most of the older subjects were from Leuven and environs, while most of the younger subjects were from Antwerp and Limburg. All the younger subjects were still a student or were highly educated. All the subjects were Dutch speaking. Contrarily, there was no language hiccup as translators were present throughout this phase. Procedure This study was a double blinded crossover clinical trial to investigate the effect of tdcs on the performance of a bimanual tracking task in younger and older subjects and the difference between the younger and older subjects. This study protocol was approved by the local Ethics Committee of Biomedical Research at Katholieke Universiteit Leuven (KUL) on May 26 th, The experiment was performed at Building de Nayer (KUL), in accordance with the ethical standards of the 1964 Declaration of Helsinki (see appendix 2). Before initiating the study, all participants gave their formal approval after they read and signed a written informed consent, which was carefully clarified to the subjects (see appendix 2). Experimental setup The study was a cross-over design with 2 sessions. One session was an active stimulation session and the other session was a sham stimulation session. Between each session, there were minimum 10 days. The order of the 2 sessions was pseudo-randomised. Each session followed the same procedure. Firstly, the subjects were asked to complete a questionnaire about their attention and their fatigue (see appendix 3). Secondly, the subjects could practice the task first, but this part was not compulsory. Initially, the subjects were oblivious to the task, practicing with a phase of trial and error. This was followed by 4 blocks of trials. With regards to the bimanual tracking task, the subjects were seated in front of a screen with their hands shielded. The hands were shielded by a table top bench and were supported. The purpose of the task was to track a moving white dot along a blue target line on the screen by rotating 2 dials with their hands simultaneously. The subjects could rotate the dials by using the thumb and index finger as accurate as possible in both space and time. 14

17 The left hand controlled movements of the red cursor on the vertical axis, while the right hand controlled movements of the red cursor on the horizontal axis. Figure 1: Setup (A = Setup motor task, B = Subject practicing, C = Subject performance with setup tdcs, D = Schematic possible lines representing different bimanual coordination patterns (Beets et al., 2015)) The performance of the hand movements indicated the coordination pattern. These movements were clockwise or counterclockwise, but also inwards or outwards. By performing these coordination patterns a particular ratio was obtained according to the rotation speed of one hand relative to the rotation speed of the other hand. Because there were simple and more difficult slopes projected on the screen, the complexity of each trial varied. The slopes related with certain ratios, which consisted of 1:1, 2:1, 3:1, 1:3, 2:5 and 5:2. The first number indicated the speed of the left hand relative to the second number, which corresponded to the speed of the right hand. A frequency ratio of 1:1 indicated the 2 hands had to move simultaneously at the same speed. A frequency ratio of 1:2 indicated the 15

18 right hand had to move twice as fast as the left hand. The ratios corresponded each trial with a particular slope: 1:1 (45 ), 2:1 (63.4 ), 3:1 (71.6 ), 2:5 (21.8 ) (Maes et al., 2017; Vancleef et al., 2016). Each one of these ratios has their own difficulties with a corresponding slope. As was demonstrated in the study of Sisti et al. (2011), more errors were made during the ratios which differed of the ratio 1:1. This result was confirmed by the study of Beets et al. (2015). Also, the study of Sisti et al. (2011) indicates that ratio 3:1 is more challenging than ratio 1:3 for right-handed subjects, because your non-dominant hand had to rotate 3 times faster than your dominant hand. This results in significantly more errors. About these results of Sisti et al. (2011) some disagreement exists, because the study of Beets et al. (2015) demonstrates that there is no significant difference between these more complex ratios. The results of the study Beets et al. (2015) does indicate that ratio 1:1 is easier and that the other ratios (non- 1:1) are more complex. For this reason, we evaluated the performance on ratio 1:1 and ratio 3:1, because in daily life too there are simple and more difficult tasks. In this way, we can transfer our results to real life context (Vancleef et al., 2016). Figure 2: motor task (A = ratio 1:1, B = ratio 1:1, C = ratio 3:1, D = ratio 1:3) 16

19 The first and second block consisted of a bimanual tracking task with real or sham tdcs for 40 trials. The third block consisted of a bimanual tracking task with EEG recording for 40 trials. Final, the fourth block consisted of a bimanual tracking task with EEG recording without visual feedback for 14 trials. A computer screen provided visual feedback. In the absence of visual feedback, the computer screen displayed the blue target line with the white dot but not the red cursor of the subject. The subjects could view their performance after each trial in the no feedback condition, but not during the trial. One trial lasted 6 seconds, so 1 block lasted 4 minutes. BLOCK 0 Complete questionnaire and possible to practice the task Break + preparing tdcs BLOCK 1 Bimanual tracking task with real or sham tdcs (40 trials) Break + complete questionnaire BLOCK 2 Bimanual tracking task with real or sham tdcs (40 trials) Break + complete questionnaire + preparing EEG cap BLOCK 3 Bimanual tracking task with EEG recording (40 trials) Break + complete questionnaire BLOCK 4 Bimanual tracking task with EEG recording without visual feedback (14 trials) Figure 3: procedure of 1 session 17

20 During the active stimulation in block 1 and 2, the subjects received anodal tdcs over their right primary motor cortex (M1) or received sham stimulation. Most of the time this type of stimulation is used in an experimental setting, tdcs is a non-invasive and easy-to-use technique that has the potential to be applied in different clinical settings over the coming years (Hummel et al., 2010; Zimerman et al., 2013). The stimulator that was used was the neuroconn tdcs stimulator. The device consisted of a battery and electrodes. Between the electrodes, a weak electric current is applied. This induces a modulation of the membrane potential of the neurons. The anode is placed upon a specific brain area on the skull, causing a depolarization of the soma. As a result, the cells will have a higher excitability (Nitsche & Paulus, 2000). In this study, the anode was placed on the right M1 and the cathode on the contralateral prefrontal area. The stimulation had the following parameters: 1.0 ma tdcs for 20 minutes with 15 seconds ramp up and 15 seconds ramp down. During the sham stimulation in block 1 and 2, the subjects received a placebo stimulation, which means 1.0 ma tdcs for 30 seconds with 15 seconds ramp up and 15 seconds ramp down. Between block 2 and 3 there was a break to apply the EEG cap. In order to properly conduct the wires, Parker Lab signal gel was applied. 18

21 36 potential older subjects NaN potential younger subjects à Called by the recruiters to ask for participation Inclusion: - Healthy female and male subjects - Age and years old - Right handed - Good to normal sight Exclusion: - Predefined criteria concerning tdcs 20 younger adults and 20 older adults met the in- and exclusion criteria Pseudo-randomized allocation Intervention (anodal tdcs + task) Control intervention (sham tdcs + task) Wash-out period (10-14 days) Control intervention (sham tdcs + task) Intervention (anodal tdcs + task) Figure 4: cross-over design At the end of block 1, 2 and 3 the subjects were asked to complete a questionnaire about their attention, fatigue and their perception of the stimulation. The subjects were oblivious concerning the type of intervention (real stimulation or sham stimulation). 19

22 Outcome measures To evaluate the performance, we considered the track target difference and point angle difference. As the measurements are an average value per block, a lower score (near to zero) indicates a higher performance. Point angle difference error The point angle difference error (PA) is the subtraction of 2 angles, as in the average angle of the target track (blue line) and the average angle of the track of the subject (red dotted line). The average angle of the slope is calculated for each ratio. This slope reflects the average point angle. The angle of the slope of the subject (red dotted line) is subtracted from the angle of the slope of the target track (blue line). For example: figure 5 visualizes a target slope of 45 (blue line) and a slope of the subject of 20 (red dotted line) which results in a point angle difference error of 25 (45-20 ). This outcome measure represents the frequency of the hands, which corresponds with the bimanual coordination. Track + target difference error The track + target difference error (TT) represents the sum of 2 distances, which is the distance of the endpoint track and the endpoint cursor. The endpoint track is the distance between the red cursor of the subject and the shortest way to the required track (blue line), as visualized in figure 5. The endpoint cursor is the shortest way from the subjects red cursor to the white dot, as visualized in figure 5. This outcome measure represents both the speed and the ratio of the subject s track. 20

23 Figure 5: visualization of the outcome measures (A = point angle difference error (PA), B = track + target difference error (TT)) Statistical analyses JMP Pro 12 was used for the statistical analyses of the data. The level of significance was assessed at p < As the registered effect are confined to 4 fixed categorical, 1 random categorical effect and 2 continuous y variables, we used the Mixed Model to properly analyse the received data. The fixed categorical effects are group, block, ratio and stimulation. Subjects are a random effect because it is a randomised sample. Adding to this, we can safely state our correlated measures because of the repetitive nature of our cross-over study. Mixed Model could also be applied in case of missing data, this is possibly applicable in this study because we utilise averages of 40 trials per block. We imported the values into JMP among different columns, namely subjects (1-40), group (Young/Old), stim (Real/Sham), ratio (1:1/3:1), block (1/2/3/4), PA value (continuous) and TT value (continuous). Hereafter we considered the assumptions of Mixed Model, namely the normality. This demonstrated that for both continuous values the residuals were not normally distributed. See below for visualization of the assumptions in figure 6. 21

24 Figure 6: H 0 assumption of normality rejected It was deemed sensible to perform a non-linear transgression of the continuous values as an optimal acquirement is normally distributed data to perform Mixed Model analyses. This method proved to the assumption attainment. Usage of the logarithm of the continuous value, the mutual distances change between the components and with that the shape of the distribution. The consequence of such a change states that the conclusions relate to the transformed data (Slotboom, 2008). As the values were right skewed divided, there was a preference for the logarithm transformation (see figure 7). Figure 7: right skewed divided data 22

25 Following, the logarithm transformation on PA and TT values was performed so normality of the logarithm residuals could be perceived, as visualized in figure 8. This demonstrated that both continuous values after transformation were distributed normally, because the data is situated in between the red lines and the goodness of fit test is non-significant, which means the H 0 hypotheses could not be rejected. Figure 8: Normality distribution of the logarithm residuals Next, the non-significant interaction effects were eliminated of both models, concerning the industry s best practice. Starting with considering the four-way interactions, if this was nonsignificant the interaction was deleted and the model was run again. Then, the three-way interactions were minded, each time the interaction effect with the highest p-value was deleted and then the model was run again. The same procedure was used for the two-way interactions and the main interactions. The BIC value was considered too, between every newly model run, since if this value decreased, it means that the model became better. Moreover, the BIC value was vastly different between the non-transformed values and the logarithmical values. 23

26 After fitting and building the model it appeared that the model suits the data and reaches the assumption of normality. However, there are some interaction effects that need to be interpreted, which were afterwards investigated by using the Tukey s Post Hoc Test. To this extent it is possible to correct for multiple comparisons as Mixed Model reveals only if the effect is significant, but not where the differences are exactly situated. This Post Hoc Test can only be performed when the effect is significant in the Mixed Model. The Tukey s Post Hoc Test compares all possible pairs of means. 24

27 Results The kinematic data was recorded during the performance of the BMC task. Mean values of (log)pa and (log)tt for Sham/Real, Young/Old and 1:1/3:1 are shown in figure 9, as lower score meant more errors. We didn t complete the data analyses of the EEG values because they weren t processed yet. Log PA Y 1:1 O 1:1 Y 3:1 O 3:1 Real 1, , , , Sham 2, , , , Sham Real Real Sham Log TT Y 1:1 O 1:1 Y 3:1 O 3:1 Real 2, , , , Sham 2, , , , Real Sham Figure 9: Mean values of (log)pa and (log)tt In general, 2 different outcome measures were evaluated: TT difference error and PA difference error. For each outcome measurement, the difference between younger and older adults, the influence of intervention, the influence of feedback and the effect of task complexity were assessed. 25

28 Point Angle Difference In appendix 4 the main and interaction effects of PA are demonstrated (Whole model + Tukey s Post Hoc). Difference between young and old The hypothesis that there conceivably would be a difference between younger and older adults in the performance of a BMC task, where older adults would perform worse than younger adults was confirmed. There is a main effect of GROUP F(1, 38) = , p< However, the interpretation of these results should be carefully administered because GROUP is in 1 significant interaction effect, namely GROUP*RATIO F(1, 591) = 3.893, p<0.05. Approximately, it is clear that older adults are less accurate then younger adults concerning PA. It is also revealed that the logarithm values of older adults are higher than those of younger adults, as demonstrated in table 4A (fixed effects parameter estimates). Group*Ratio Concerning GROUP*RATIO it is visible in the boxplot (figure 10) that there is a difference for 1:1, as older adults perform worse than younger adults. This is confirmed after performing the Tukey s Post Hoc Test (t(50.802) = , p <0.001). This is also visible for ratio 3:1 and is confirmed by Tukey s Post Hoc Test (t(50.802) = , p=0.002). For further details see appendix 4B. Figure 10: boxplot GROUP*RATIO; first impression: younger adults make more mistakes than older adults from ratio 1:1 to ratio 3:1 (this possibly because older adults find ratio 1:1 already harder than younger adults) 26

29 Influence of the stimulation Towards the influence of stimulation on the performance of the BMC task, there was no significant effect for PA found because there was no main effect of STIM F(1, 591) = 2.008, p= In this case, the supposition could not be confirmed. Influence of feedback The main effect of BLOCK F(3, 591) = , p<0.001 confirms the hypothesis that there is an effect of feedback. With the Tukey s Post Hoc Test a significant difference between block 3 and block 4 appears. We were also interested in this, because in block 3 feedback was given and in block 4 feedback wasn t given. Block 4 (t(81.119) = , p<0.001) was generally perceived as more difficult (see appendix 4C), however no inequity could be made between younger and older adults. Ergo, the premise towards younger versus older adults regarding feedback could not be confirmed, in addition, it can be confirmed that there is an effect of feedback. But this clarification should be taken carefully because BLOCK is also in an significant interaction effect, namely RATIO*BLOCK F(3, 591) = , p< Ratio*Block See infra (task complexity) Influence of the task complexity There is a main effect of RATIO F(1, 591) = , p<0.001, which means that there is an influence of task complexity. The interpretation of these results should be carefully administered because RATIO is in 2 significant interaction effects, namely GROUP*RATIO F(1, 591) = 3.893, p=0.049 and RATIO*BLOCK F(3, 591) = , p< Group*ratio Previously it was already confirmed that based on the Tukey s Post Hoc Test, for ratio 1:1 as for ratio 3:1 there is a significant difference between older and younger adults too (see appendix 4B). This is visually demonstrated in figure

30 Ratio*Block RATIO*BLOCK demonstrates that through the Tukey s Post Hoc test, ratio 3:1 is significant harder during block 4. During the other blocks, it is revealed that during ratio 3:1 more mistakes were made but these are not significant (see appendix 4D). This is also visualized in figure 11. Block 1; 1:1 vs 3:1, t(155.49) = , p>0.05 Block 2; 1:1 vs 3:1, t(155.49) = , p>0.05 Block 3; 1:1 vs 3:1, t(155.49) = , p>0.05 Block 4; 1:1 vs 3:1, t(155.49) = , p<0.001 Figure 11: boxplot RATIO*BLOCK; first impression: ratio 1:1 seems to get easier in time, while ratio 3:1 also gets easier but without feedback (block 4) it gets harder yet again for the subjects 28

31 Track + Target Difference In appendix 5, the main and interaction effects of TT are visualised (whole model + Tukey s Post Hoc Test). Difference between young and old Our first premise stated that there would be a difference between younger and older adults in the performance of bimanual coordination, of which we thought older adults would perform lesser than younger adults. This hypothesis was confirmed by our data-analyses. There was a main effect in GROUP F(1, 38) = , p<0.001 (see appendix 5A). However, we need to take caution about this interpretation, because GROUP is part of 2 significant interaction effects, namely GROUP*RATIO F(1, 588) = , p<0.001 and GROUP*BLOCK F(3, 588) = 3.124, p< To consider these interaction effects, we performed a Tukey's Post Hoc Test. Generally, the older adults are less accurate in performance regarding TT. We noticed that the logarithm values of the older adults are higher, as visualised in table 5A (fixed effects parameters estimates). Group*Ratio Regarding GROUP*RATIO, there is a difference visible in figure 12 for ratio 1.1, with the older adults performing penurious against younger adults. This was also confirmed after performing the Tukey's Post Hoc Test (t(46.054) = , p<0.001). At first, this does not appear to be the case for ratio 3:1. However, after performing the Tukey's Post Hoc Test, there is a significant distinction (t(46.054) = , p<0.001), see appendix 5B. This immediately gave us an answer as to whether or not there is an effect of difficulty. This will be further explained. 29

32 Figure 12: Boxplot GROUP*RATIO; first impression: it is less observable that younger adults may make more mistakes than older adults from ratio 1:1 to ratio 3:1, compared to figure 10 of PA. However, it is clear that both younger and older adults may make more errors in ratio 3:1 Group*Block See infra section feedback. Influence of the stimulation In our second hypothesis, we suggested that stimulation could possibly affect the performance of the task. There is a main effect of STIM F(1, 588) = , p < By real tdcs stimulation, the group scored (log TT) lower than when the sham tdcs group was administered. This is revealed by the Tukey's Post Hoc Test (see appendix 5C). Influence of feedback There was a main effect of BLOCK F(3, 588) = , p < Based on Tukey's Post Hoc Test, we perceived that there is a significant effect of feedback between block 3 and 4. We were also interested in this, as feedback was given in block 3 and not in block 4. Block 4 (t (64.209) = , p <0.001) was generally experienced as more difficult (see appendix 5D). However, we must be careful about this perception, as BLOCK is part of 2 significant interaction effects. Namely GROUP*BLOCK F(3, 588) = 3.124, p < And RATIO*BLOCK F(3,588) = 7.541, p <

33 Group*Block Figure 13: boxplot GROUP*BLOCK; first impression: here, the motive given is that in block 4 more errors are made, for older as well as for younger adults. It also is given that for all subjects there is no considerable difference between block 1 & block 4, it remains clear that as one gets closer to block 3, fewer mistakes are made. To investigate the difference between the younger and older adults, we started with visually considering the boxplot, and afterwards we performed a Tukey's Post Hoc Test between block 3 and 4. Both for younger and older adults there appears to be a significant effect, block 4 was perceived as more difficult than block 3. Younger adults: (t(64.209) = , p<0.001), older adults: (t(64.209) = , p<0.001) (see appendix 5E with the yellow mark). We can also explore the difference between the young adults and the older adults and the effect of feedback between block 3 and 4. Younger adults turn out to perform better than older adults, both during block 3 as block 4 (see appendix 5E with the green mark). Block 3: there is a significant difference, namely younger adults perform better (t(64.209) = , p<0.001). Block 4: there is a significant difference, whereby younger adults also perform better (t(64.209) = , p<0.001). This confirmed our premise that there would be an effect of feedback and that younger adults would perform better, even without feedback. 31

34 Influence of the task complexity The final hypothesis was that there would be an effect of difficulty. We stated that the ratio 3:1 would be more difficult than ratio 1:1. There is a main effect of RATIO F(1, 588) = , p < However, we need to take this interpretation cautiously, because RATIO is part of 2 significant interaction-effects. GROUP*RATIO F(1, 588) = , p <0.001 and RATIO*BLOCK F(3, 588) = 7.541, p < Group*Ratio See supra for the difference between younger and older adults, where it was confirmed that there is a significant interaction effect of GROUP*RATIO. Thus, it was found that older adults performed worse to younger adults, both for ratio 1:1 and for ratio 3:1 (see appendix 5B). Ratio*Block With this regard to RATIO*BLOCK, it appears in the Tukey's Post Hoc test that ratio 3:1 is more difficult than ratio 1:1 for each block (see appendix 5F). Block 1; 1:1 vs 3:1, t(107.47) = , p<0.001 Block 2; 1:1 vs 3:1, t(107.47) = , p<0.001 Block 3; 1:1 vs 3:1, t(107.47) = , p<0.001 Block 4; 1:1 vs 3:1, t(107.47) = , p<0.001 Figure 14: boxplot RATIO*BLOCK; first impression: ratio 1:1 seems easier and easier, while ratio 3:1 also becomes a little easier but without feedback (block 4) it becomes much more difficult for the subjects 32

35 Discussion The main results of our study were: (1) older adults perform significantly worse on both parameters than younger adults, (2) stimulation had a positive significant influence on parameter TT, but not on parameter PA, (3) there is a significant positive influence of feedback, both in parameter PA and in parameter TT, (4) in both parameters, a difference is discovered between ratio 3:1 and ratio 1:1, so there is certainly an effect of difficulty. In our study, a direct comparison between younger and older adults was made. In this section, the findings will be further discussed. The results, concerning PA, outline the frequency in which hands move, this is necessary in bimanual coordination. The results, related to TT, are a somewhat more complex measurement as speed and ratio are involved. Speed is a concerted factor as the white dot should be tracked, but one should also mind the influence of the ratio as the divergence in slopes adds a degree of difficulty. TT has a high accuracy factor as well. Regarding the results of the first hypothesis, concerning older adults versus younger adults, we discovered that the older adults perform significantly worse than the younger adults on both parameters, which corresponds to the results found in Corti et al. (2017). For the results embracing the parameter PA, this may be due to impecunious coordination and this conceivably because older adults perform these activities not as often, since for example younger adults like to game. Or maybe because the integrity of the white mass has decreased, the importance of which is denoted as the CC provides interhemispheric communication during actions requiring bimanual coordination. The study of Shetty et al. (2014) indicates that there is a gender inequity concerning the CC, by demonstrating that men are better in performing BMC tasks. Table 1 demonstrated that our study had approximately equal groups regarding gender. Also, the study of Shiffman (1992) demonstrates that aging affects the hand function. Healthy older adults should have a stable hand function until the age of 65, followed by a slow decline in the following years. Most notable were the age divergences in performances beyond the threshold of 75 years of age. In our study, the average age of the elderly was below the age of 75, which can explain why 33

36 there were no striking differences. Other possible causes, according to the literature, may be dementia, Alzheimer's or MCI (Kiyama, Kunimi, Iidaka, & Nakai, 2014). Concerning the TT component, it is also possible that the integrity of the white mass has decreased and leads to a less rapid response time in elderly, because older persons are often slower and less accurate (Bangert, Reuter-Lorenz, Walsh, Schachter, & Seidler, 2010; Hummel et al., 2010; Shetty et al., 2014). Thus, the response rate might have been a useful outcome, but this could not be performed because we used average values. In both parameters, the link with physiotherapy can be made by investigating whether or not degeneration can be decelerated by exercise therapy. As is, we wonder if there is room for improvement after degeneration. To our knowledge there is no corresponding study and should be further investigated. According to the study of Shetty et al. (2014), age-related decrease in coordination can be explained by changes in the cortical and subcortical motor structures, by example the decrease in muscle mass and strength in older adults. As physiotherapists, we can interfere here. As our results reveal that older adults make more mistakes than younger adults during the BMC task, it can be stated that elderly will experience more difficulties during fine motor tasks and bimanual tasks in everyday life (Corti et al., 2017). Here, a physiotherapist can also have an important role. For this purpose, we examined whether or not stimulation could have an effect and whether visual feedback would be useful. If this is the case, it may be applied in practice, especially in the initial stage of motor learning, as feedback plays a major role. According to the study of Beets et al. (2015), learning effects obtained by giving feedback, will not only affect performance when feedback is given, but can also be extended to situations with no feedback. However, it is very important to take into consideration that subjects, who only train with visual feedback, become dependent on it and that performance deteriorates in situations without feedback. While training without feedback limits learning. As a result, training in situations with both feedback and no feedback is the best training strategy (Beets et al., 2015). 34

37 Concerning the second hypothesis, the stimulation intervention reveals that for the TT parameter stimulation had a significant positive influence on the performance, but for the PA parameter this was not the case. Approximately, these results coincide with the findings of the study of Hummel et al. (2010), which demonstrates that there is a positive effect of stimulation on the performance in younger and older adults as well. From the study of Hummel et al. (2010) it seems that if the subjects became older, the effect of the stimulation was more severe on the performance. Since stimulation doesn t have a significant interaction effect, we could not examine if there was an effect in older and/or younger adults. Initially, we scrutinized why M1 was stimulated. The study of Fleming and Newham (2016) demonstrates that the coordination of the upper limbs relies on the communication between the 2 cerebral hemispheres, as each M1 interacts with the contrary M1 so bimanual movements can be facilitated. This communication is mediated on cortical level by the pathways in the CC. M1 has as a function to produce the complex sequential movement and has a direct connection with the spinal cord through 1 synapse. This synapse is important because aging not only has an influence on cortical level, but also spinal level, which can give a contribution to a decrease in coordination in older adults (Shetty et al., 2014). Another important remark is also that for tdcs it is not exclusive that only M1 gets to be stimulated, the reason for this being the spatial resolution of tdcs, which possibly stimulates also other areas surrounding M1 (Hummel et al., 2010). We also investigated why M1 was stimulated ipsilateral (right). We state that because all the subjects were right-handed, the anode was placed on the right primary cortex. Thusly, the left-hand would be commensurate to the right-hand, which causes a better performance with the help of stimulation. This should be visible in both ratio 1:1 and ratio 1:3; as the lefthand must work threefold as hard as the left hand. However, in various study s, we came across other kinds of stimulation on M1. Specifically, in the study of Panouilleres et al. (2015), Hummel et al. (2010) and Vancleef et al. (2016) the left M1 was stimulated. In the study of Panouilleres et al. (2015), the subjects are left-handed, so it is acumen that the reasoning was antithetical to this study. However, in the study of Hummel et al. (2010) the 35

38 subjects were right-handed and the study of Vancleef et al. (2016) did not demonstrate what hand dominance the subjects had. Also in both studies, it was not explained why the stimulation was applied on the left M1 so it is until now still not clear for us. Inexorably, we queried why the stimulation of M1 ipsilateral doesn t have an effect in our study, specifically on parameter PA. In our study, the stimulation is given on the right M1 (left hand), however Ward (2006) has demonstrated that aging leads to reduced transcallosal inhibition of the ipsilateral motor cortex. For example, in the task with ratio 3:1, the left-hand must turn threefold as fast as the right-hand, so inhibition of the left motor cortex (right-hand) should be reduced in older adults. Thus, there is a mirror movement of the right-hand on the left-hand possible. If for older adults the stimulation was not effective, this should be a possible explanation, but we could not examine this since stimulation did not have a significant interaction effect. In the interest of not having a significant main effect of stimulation, we inferred this could be credited to the intricacy of parameter PA. It is possible that because of the stimulation, the subjects were going to use another strategy to perform the task better, but this can also result in a decrease in performance of PA. Regarding the averages of both results, it even becomes visible that for parameter PA the subjects perform the task significantly worse during the interventional stimulation. Also, it is possible that only stimulation of the primary motor cortex was not sufficient to improve PA, this because possibly other brain areas contribute to the attainment of this complex coordination task, for example the dorsolateral prefrontal cortex (DLPFC). The DLPFC is a cognitive area that is related to the motor area, corresponding to the motor cortex. Antecedent studies have demonstrated that too the DLPFC and M1 are involved in complex bimanual tasks (Vancleef et al., 2016). According to the study of Corti et al. (2017), movements in older adults are associated with increased recruitment of the prefrontal cortex (cognitive area) because coordination in older adults are less automatic and more conscious (Heuninckx et al., 2005), which requires more cognitive control. It is already demonstrated that the fronto-parietal areas are crucial for visually guided bimanual coordination (Kiyama et al., 2014). Within these fronto-parietal areas, M1 serves primarily 36

39 for the execution of the bimanual finger movements, but the supplementary motor area (SMA) and the pre-motor area (PMA) together are responsible for planning the preparation of the motor actions. Consequently, these areas are important for bimanual coordination (Beets et al., 2015), according as the cerebellum also has a large influence on bimanual coordination (Beets et al., 2015; Panouilleres et al., 2015). SMA has also an influence on the activity of M1 during the performance of a visual bimanual movement (Beets et al., 2015; Fling et al., 2011). If older adults made more use of conscious controlled movements, then the stimulation should also be given on a different brain area, so not just stimulation on M1 but also pre-frontal (DLPFC for example). A critical reflection upon these studies, in which is retracted which activation patterns are normally active during the task, is that these findings are only an example of possibly activated areas because the specific task was not completely the same. To be certain of what areas are activated during our task, we should observe the EEG activation patterns during the BMC task. Also, over-activation areas of other studies should not be copied from other studies because these are also task specific. Since in older adults over-activation of brain areas occurs, it means that older adults use a whole bunch of additional brain areas. If these other areas are not being stimulated, there is no improvement possible. The compensation theory of Ward (2006) supports this statement. Ward (2006) demonstrates that the activated networks are task specific, and that the concomitant age changes differ by task, whereby the largest difference is visible in the M1 area. The third hypothesis, regarding augmented visual feedback, was answered as follows: block 4 is significantly more difficult in both parameter PA and TT, because block 4 did not provide augmented visual feedback. In this context, the important role of the physiotherapist is being discussed. For example, if it is useful to provide augmented visual feedback during therapy. First, we can query if visual feedback can be equaled to external-controlled feedback by a physiotherapist. Based on the stages of motor learning, in which the subject during the automation phase relies on internal feedback, we considered that internal feedback is the most suitable to use in daily life. If this internal feedback is not given, the patient must relapse on external feedback (Sisti et al., 37

40 2011). It is very important that you reduce gradually the amount of feedback during the phases of motor learning. However, in our study, the transition from feedback conditions to non-feedback conditions was abrupt. As so, we cannot expect the subjects to perform the task properly without feedback. In order to achieve transition of exercises during rehabilitation to real-life context, it is important to work with a mix of feedback and nonfeedback situations during the training (Beets et al., 2015). We gave external augmented feedback to our subjects, which is possibly the reason why our subjects have not been able to build somatosensory internal representations. As the consequence, all subjects achieved significantly worse results in block 4. It is known that elderly attach great importance to feedback (Heuninckx et al., 2010), this could be a reason why they perform worse, compared with younger adults, as found in parameter TT. Unfortunately, we could not take this finding into account for parameter PA, because there was no significant interaction effect. We cannot determine whether or not there would be an effect of stimulation in block 4, as there is no significant interaction effect of block with stimulation. If there was a decreased effect of stimulation in the last block, it could have been a feasible explanation why the subjects achieved reduced results in block 4. However, there is also a disagreement in literature about the duration of stimulation (Hummel et al., 2010; Nitsche & Paulus, 2000; Panouilleres et al., 2015). Acknowledged as well from Beets et al. (2015) is the fact that other brain areas are active during augmented feedback and non-augmented feedback conditions, which is an underlying reason why the subjects perform worse without feedback. It may be conceivable that these other brain regions are more complex and require more interaction. Also, these regions weren't active previously during the trials, and in addition the number of trials in block 4 was only 14, which makes it less feasible to learn the task. 38

41 As the final hypothesis, we regarded the effect of difficulty. We discovered a difference between ratio 3:1 and ratio 1:1 in both PA and TT. This was also confirmed by the study of Sisti et al. (2011) which established the same findings, namely that ratio 3:1 is a more difficult ratio, which can be clarified because in ratio 3:1 the relative angular velocity is the utmost (Sisti et al., 2011). We expected, as previously indicated, that if the left (non-dominant) hand was stimulated it would not make a vast difference if the non-dominant hand had to run threefold as fast as the dominant hand. This could have been analysed by comparing ratio 3:1 with ratio 1:3. We established that older adults perform destitute to their younger counterparts in both parameters during ratio 3:1. So there are solid differences between older and younger adults, despite the more demanding task. For parameter TT, this finding is applicable for each block. For parameter PA however, this applies only during block 4. According to Corti et al. (2017), this result concerning older adults would have a compelling impact on the ability to perform daily tasks. Also, Corti et al. (2017) declares that elderly make more use of the compensation theory than younger adults. These compensations are valuable for unimanual and simple bimanual tasks, but from the moment the task becomes a complex bimanual task, the compensation is insufficient. There even is a ceiling effect. Thus, in complex tasks (requiring cognition) the performance will not improve abundant, this could also be a reason the elderly perform less well. Additionally, the prefrontal areas are the most sensitive to age related atrophy, to which we may relate the ceiling effect. However, these are still speculations and further research is needed. 39

42 Weaknesses - The response rate would have been a useful outcome measure. But since the results are average values, we weren t able to construe these results. - One has to be cognizant of the interpretation of the results because averages have been taken from start- and endpoints and no different time points are considered. - The outcome measure TT consists of the sum of 2 different measures, because of this it is difficult to distinguish what the issue is. Considering the measure endpoint track, you might interpret that the subject has correctly followed the target track (blue line), but not the target dot (white dot). Conversely when you consider the endpoint cursor, you might interpret that the subject has correctly followed the white dot, but didn t follow the target track accurately. - During the experiment, it wasn t always clear to the subjects that they had to follow the white dot as accurate as possible, for this reason the ignorance of the subjects also has an influence on the parameter TT. - Speed wasn t considered in the statistical analyses, because it wasn t indicated to the subjects that they had to follow the white dot at the same speed. Nevertheless, speed would have been an interesting outcome measure as it could be related to rehabilitation and physiotherapy. - Because of the logarithm transformation, it is impossible to make a prediction. Since the values are formed in a consistent manner, the conclusion will be the same, but a prediction can t be made. Hence, we need the true values. - In our study, it would have been interesting to investigate the EEG in order to learn something more about brain activity over the different conditions. - In this statistical analysis, we haven t considered the confounding factors of attention and fatigue. This was questioned during the experiment based on the VAS scale, but the data wasn t considered. We could have analysed this data by analysing within the group. We could also have specifically analysed the data of subjects whom didn t perform well in order to investigate which confounding factor played a major role. - We could also explore whether or not there was a placebo effect of sham stimulation on the performance by studying the questionnaires. - In the initial phase of the experiment the subjects were asked if they were right-handed, but this wasn t confirmed empirical. 40

43 - An Mini Mental State Examination (MMSE) could have been taken, since mild cognitive impairment (such as dementia and Alzheimer) are correlated with a reduced hand coordination (Kiyama et al., 2014). Also, the Geriatric Depression Scale (GDS) could have been taken, since patients with depression are slower in general. - The subjects were asked if they use medication, but this wasn t taken into account during the interpretation. - It isn t clear why the decision was made to use the ratio 5:2. To date, we found no evidence in the literature. - Considering the influence of feedback, it is important to consider that during the condition without augmented feedback there are fewer trials than during the conditions with augmented feedback. Because of this the data are less comparable and could even be a confounding factor. Strengths - In light of the combination of PA and TT, a broader understanding is relayed. If you wouldn t consider both parameters, the representation of the performance of the task would not be complete. - Using Mixed Model is an advantage, because all the values are averages and maybe there is some missing data. Mixed Model can be used in the case of repeated measurements, correlated measurements and missing data. The subjects were categorized as random because younger and older adults have different coordination patterns within group, this is also the advantage for using cross-over model, so there is no influence of the level of BMC of the subjects. - The groups were equal, since both groups included 20 subjects. Also, the rate of female to male was reasonably equal. This is important since there are sex differences in CC and bimanual coordination could be predicted by age and gender (Shetty et al., 2014). - The reasoning behind the inclusion of older adults in this study from the age of 65 is that from the age of 60 an obvious decline in motor performance could be visible (Ward, 2006). - The performance of all components of the task was the same for every subject, since the order of the 4 blocks was fixed. The only differences were some small deviations in the 41

44 breaks every subject took. Each subject also had practised the task in advance, before the beginning of the first block. - By having a break of minimum 10 days in between the sessions, we were able to exclude the subjects learning curve due to it being so small it could be neglected. Maybe a learning curve was existent from block 1 to 3, but only temporary during the session. - All trials have been run approximately the same amount of times by making sure the sequence of the trials was chosen randomly. By using this pseudo randomisation as a form of control, any possible negative influence of the outcome, by not having all patterns included in 1 trial, was avoided. - The study of Ward (2006) also demonstrated that age related changes are task specific, for this reason ratio 1:1 and 3:1 was chosen. In this way, the task difficulty could be transferred to daily life. The study of Ward (2006) reveals that a simple task corresponds with a motor task that is already known, and that with a complex task more cognitive areas are activated. So de facto, this study is a behavioral study in the grounds that the consideration was established to determine the divergence in performance for simple and complex tasks. - BMC is also a meaningful way to diagnose, evaluate and rehabilitate (Maes et al., 2017). By virtue of it already being acknowledged that a reduced BMC performance is correlated with certain forms of mild cognitive impairment (MCI) and Alzheimer s disease (AD). In the case of a detection of an age-related decline in finger coordination tasks, it could have an important contribution to the treatment and progression of dementia (Kiyama et al., 2014) Future Research - Further research is necessary to reflect on the cognitive load of the different ratios. We agree that ratio 1:1 is easier because the cognitive load is lower than in ratio 3:1. - The EEG has to be interpreted, because in this way future research can optimise the area of stimulation in a specific task. - The sample size of this study is very small so future research is necessary on a bigger population so the results are generalisable. 42

45 Conclusion The main findings of the present study are that older adults performed the BMC task generally significantly worse than younger adults, which is an indication to perform physiotherapy with older adults. Since the addition of visual augmented feedback has a major positive influence on the performance of a BMC task, and the complexity of the task also has an impact on the BMC task performance, this should be taken into consideration during the physiotherapy sessions. For example, more complex exercises should be included together with gradually declined feedback. At last the influence of stimulation is still unclear, so further research is necessary. 43

46 44

47 Appendices Appendix 1: Recruitment flyer Bent u ook nieuwsgierig naar nieuwe ontwikkelingen in de wetenschap? Doe dan nu mee met het onderzoek naar bimanuele coördinatie Het$ labo$ motorische$ controle$ en$ neuroplasticiteit$ van$ de$ KU$ Leuven$ voert$ in$ samenwerking$ met$ de$ Universiteit$ Hasselt$ een$ onderzoek$ naar$ het$ functioneren$van$het$brein$en$de$relatie$met$motorisch$ functioneren.$hiervoor$zijn$we$opzoek$naar$$gezonde' (niet+rokers),'rechtshandige'jongeren'tussen'18'en' 30' jaar$ die$ gedurende$ een$ eenmalige$ sessie$ willen$ deelnemen$aan$ons$onderzoek.$ $ Het$experiment$(duur:$max.$2' uur)$zal$plaatsvinden$in$ Leuven$(Sportkot)$en$zal$als$volgt$verlopen:$ Eerst$ zal$ u$ enkele$ vragenlijsten$ mogen$ invullen.$ Vervolgens$zal$u$een$Electro'Encephalogram'(EEG)$ muts$ aangemeten$ krijgen,$ waarmee$ we$ uw$ hersenactiviteit$ kunnen$ meten.$ Deze$ hersenactiviteit$ zal$ gemeten$ worden$ tijdens$ het$ uitvoeren$ van$ een$ uitdagende$bimanuele$coordinatie$taak. Geïnteresseerd? Neem dan contact op met dr. Koen Cuypers (bij voorkeur): koen.cuypers@faber.kuleuven.be Telefoonnummer: Vergoeding U krijgt na afloop van de studie 1 filmticket of boekenbon. 45

48 Bent u ook nieuwsgierig naar nieuwe ontwikkelingen in de wetenschap? Doe dan nu mee met het onderzoek bimanuele coördinatie en veroudering Het$ labo$ motorische$ controle$ en$ neuroplasticiteit$ van$ de$ KU$ Leuven$ voert$ in$ samenwerking$ met$ de$ Universiteit$ Hasselt$ een$ onderzoek$ naar$ het$ functioneren$van$het$verouderende$brein$en$de$relatie$ met$ motorisch$ functioneren.$ Hiervoor$ zijn$ we$ opzoek$ naar$ $ gezonde' (niet+rokers),' rechtshandige' 65+ plussers' (max.' 77' jaar)$ die$ gedurende$ een$ eenmalige$ sessie$ willen$ deelnemen$ aan$ ons$ onderzoek.$ $ Het$experiment$(duur:$max.$2' uur)$zal$plaatsvinden$in$ Leuven$(Sportkot)$en$zal$als$volgt$verlopen:$ Eerst$ zal$ u$ enkele$ vragenlijsten$ mogen$ invullen.$ Vervolgens$zal$u$een$Electro'Encephalogram'(EEG)$ muts$ aangemeten$ krijgen,$ waarmee$ we$ uw$ hersenactiviteit$ kunnen$ meten.$ Deze$ hersenactiviteit$ zal$ gemeten$ worden$ tijdens$ het$ uitvoeren$ van$ een$ uitdagende$bimanuele$coordinatie$taak. Geïnteresseerd?? Neem dan contact op met dr. Koen Cuypers (bij voorkeur): koen.cuypers@faber.kuleuven.be Telefoonnummer: Vergoeding U krijgt na afloop van de studie 1 filmticket of boekenbon. 46

49 Appendix 2: Informed written consent Titel van de studie: Modulatie van leeftijdsgerelateerde veranderingen van functionele netwerken in bimanueel visueel-motorisch leren Opdrachtgever van de studie: KULeuven Onderzoeksinstelling: Faber Ethisch comité: Centraal CME : Comité voor Medische Ethiek UZ Leuven UZ Leuven Herestraat Leuven Plaatselijke onderzoeker: Raf Meesen raf.meesen@faber.kuleuven.be I Noodzakelijke informatie voor uw beslissing om deel te nemen (4 pagina s) Inleiding U wordt uitgenodigd om deel te nemen aan een academische studie. Voordat u akkoord gaat om aan deze studie deel te nemen, vragen wij u om kennis te nemen van wat deze studie zal inhouden op het gebied van organisatie, zodat u een welbewuste beslissing kunt nemen. Dit wordt een geïnformeerde toestemming genoemd. Wij vragen u de volgende pagina s met informatie aandachtig te lezen. Hebt u vragen, dan kan u terecht bij de onderzoeker of zijn of haar vertegenwoordiger. Dit document bestaat uit 3 delen: essentiële informatie die u nodig heeft voor het nemen van uw beslissing, uw schriftelijke toestemming en bijlagen waarin u meer details terugvindt over bepaalde onderdelen van de basisinformatie. Als u aan deze studie deelneemt, moet u weten dat: Ø Deze studie opgesteld is na evaluatie door een ethisch comitée. Ø Uw deelname is vrijwillig; er kan op geen enkele manier sprake zijn van dwang. Voor deelname is uw ondertekende toestemming nodig. Ook nadat u hebt getekend, kan u de onderzoeker laten weten dat u uw deelname wilt stopzetten. Ø De gegevens die in het kader van uw deelname worden verzameld, zijn vertrouwelijk. Bij de publicatie van de resultaten is uw anonimiteit verzekerd. Ø Er is een verzekering afgesloten voor het geval dat u schade zou oplopen in het kader van uw deelname aan deze studie. Ø Indien u extra informatie wenst, kan u altijd contact opnemen met de onderzoeker of een medewerker van zijn of haar team. Doelstellingen en verloop van de studie Deze studie is georganiseerd om inzicht te krijgen in leeftijdsafhankelijke veranderingen in het verouderende brein. In een eerder onderzoek hebben we kunnen aantonen dat er tijdens de uitvoering van tweehandige taken (zoals bijvoorbeeld bij het dichtknopen van een hemd) verschillen zijn tussen jongeren en ouderen in de werking van het brein. De studie van de onderliggende redenen waarom er verschillen zijn en hoe de verschillende gebieden in het brein onderling communiceren is noodzakelijk om tot een dieper inzicht te komen in het functioneren van het verouderende brein. In de huidige studie willen we met de informatie bekomen uit onze eerdere studie nieuwe revalidatiestrategiën uitwerken door gericht de ouderdomsgerelateerde veranderingen van de functionele netwerken tijdens de uitvoering van tweehandige taken te moduleren. Aan deze studie zullen 40 personen deelnemen. Om na te gaan of u in aanmerking komt voor de studie moet u de volgende lijst grondig doornemen. Indien GEEN van deze exclusie-criteria op u van toepassing is, kan u deelnemen aan de studie. Gezien zwangerschap een exclusie-criterium is zal er een zwangerschapstest uitgevoerd worden. Metaal in het hoofd Pacemaker / Draden / Implanteerbare defibrillator Biostimulator of TENS-apparaat Aneurysmaclips (hersenbloedvaten, aorta, etc. ) Intracraniële clips Geïmplanteerde medicatiepomp Granaatscherven (eender waar in het lichaam) Ooit metaalarbeider geweest in het verleden 47

50 Eén van volgende aandoeningen gehad in het verleden: ü Hersentrombose ü Hersenbloeding ü Hoofdtrauma ü Meningitis (hersenvliesontsteking) ü Lange periode van bewusteloosheid (>1u) ü Migraine ü Epilepsie Voorgeschiedenis van hersenchirurgie Familiale voorgeschiedenis van therapieresistente epilepsie Voorgeschiedenis van actief misbruik van een bepaalde stof gedurende het laatste jaar Getatoeëerde eyeliner Zwangerschap Huidletsels op de plaats van stimulatie Deze lijst van exclusiecriteria is ruimer dan strikt noodzakelijk voor de in dit document beschreven techniek, dit is gedaan omdat er mogelijks nog vervolgstudies komen op dit experiment waarbij er andere technieken gebruikt zullen worden. Hierbij willen we expliciet vermelden dat elk bijkomend onderzoek wat buiten de in dit document beschreven context valt alleen kan en zal plaatsvinden na een bijkomende aanvraag aan het ethisch comité en waarvoor goedkeuring verkregen werd. Het experiment zal worden uitgevoerd in twee sessies met 10 dagen tussen. Een sessie zal ongeveer 2 uur in beslag nemen (30 minuten voor instructies / voorbereiding, 60 minuten voor de metingen en de interventie, met inbegrip van kleine pauzes en 30 minuten om op te ruimen). Tijdens de eerste sessie zal de onderzoeker u vragen om alle voor de studie noodzakelijke gegevens en informatie te verzamelen - zoals uw demografische gegevens (leeftijd, gewicht, lengte, geslacht) evenals gegevens over uw medische voorgeschiedenis, uw geneesmiddelengebruik, uw afhankelijkheid van bepaalde producten (tabaksgebruik, alcoholverbruik) enz. Daarna starten de metingen van de hersenactiviteit (EEG) en de interventie, transcraniële microstroom stimulatie (TCS) Een elektro-encefalogram (EEG) is een toestel dat de activiteit van de hersengebieden meet. Het toestel bestaat uit een muts met meetelectroden die op het hoofd wordt geplaatst. De interventie bestaat uit Transcraniële Micro Current Stimulation (TCS) dit is een zeer lage stroom (microstroom) die aangebracht wordt op het hoofd. Dit gebeurt door middel van 2 spons elektroden die op het hoofd bevestigd worden. In één sessie bestaat dus uit vier delen 1. Verzamelen van gegevens zoals leeftijd, geslacht, geneesmiddelengebruik, Voorbereiding EEG: U zal gevraagd worden om een muts op te zetten. 3. Bepalen van de activiteit van de betrokken hersengebieden tijdens de uitvoering van een tweehandige taak 4. Stimulatie ( echte of schijn stimulatie) van de hersengebieden met microstroom De tweede sessie verloopt gelijkaardig aan de eerste sessie enkel bij het verzamelen van de gegevens zal er nagegaan worden of er verschillen zijn in bv medicatiegebruik ten opzichte van de eerste sessie. Beschrijving van de risico s en van de voordelen Mogelijke risico's die kunnen optreden tijdens TCS en EEG zijn huidirritaties onder de elektroden en lichte hoofdpijn in het geval van TCS. Personen met huidletsels op de stimulatieplaats worden niet opgenomen in het onderzoek. Personen die enig ongemak ondervinden mogen op elk moment vrijwillig ontslag uit de studie nemen. Ook moet u niet verwachten dat uw deelname aan deze studie u persoonlijke voordelen zal opleveren. U moet begrijpen dat uw deelname aan deze studie ervoor zal zorgen dat wij leeftijdsafhankelijke veranderingen in de hersenen beter begrijpen en bijgevolg in de toekomst betere behandelingen kunnen ontwikkelen. Na elke sessie krijgt u een filmticket of boekenbon. Intrekking van uw toestemming U neemt vrijwillig deel aan deze studie en u hebt het recht om uw toestemming voor gelijk welke reden in te trekken. U hoeft hiervoor geen reden op te geven. Als u uw toestemming intrekt, zullen de gegevens bewaard blijven die tot op het ogenblik van uw stopzetting werden verzameld. Dit om de geldigheid van de studie te garanderen. Er zal geen enkel nieuw gegeven aan de opdrachtgever worden gegeven. De opdrachtgever/verantwoordelijke van de studie zou ook kunnen beslissen om de studie te stoppen indien de verzamelde gegevens sneller dan voorzien een antwoord brengen. 48

51 Als u aan deze studie deelneemt, vragen wij om: Ø Tenvolle mee te werken voor een correct verloop van de studie. Ø Geen informatie over uw gezondheidstoestand, de geneesmiddelen die u gebruikt of de symptomen die u ervaart te verzwijgen. Ø Uw onderzoeker op de hoogte te brengen als men u voorstelt om aan een andere studie deel te nemen zodat u met hem/haar kan bespreken of u aan deze studie kunt deelnemen en of uw deelname aan de huidige studie moet worden stopgezet. Contact Als u bijkomende informatie wenst, maar ook ingeval van problemen of als u zich zorgen maakt, kan u contact opnemen met de hoofdonderzoeker (Meesen Raf) op het telefoonnummer ( ) of raf.meesen@faber.kuleuven.be Titel van de studie: Modulatie van leeftijdsgerelateerde veranderingen van functionele netwerken in bimanueel visueel-motorisch leren II Geïnformeerde toestemming Deelnemer Ik verklaar dat ik geïnformeerd ben over de aard, het doel, de duur, de eventuele voordelen en risico s van de studie en dat ik weet wat van mij wordt verwacht. Ik heb kennis genomen van het informatiedocument en de bijlagen ervan. Ik heb voldoende tijd gehad om na te denken en met een door mij gekozen persoon, zoals mijn huisarts of een familielid, te praten. Ik heb alle vragen kunnen stellen die bij me opkwamen en ik heb een duidelijk antwoord gekregen op mijn vragen. Ik begrijp dat mijn deelname aan deze studie vrijwillig is en dat ik vrij ben mijn deelname aan deze studie stop te zetten zonder dat dit mijn relatie schaadt met het therapeutisch team dat instaat voor mijn gezondheid. Ik begrijp dat er tijdens mijn deelname aan deze studie gegevens over mij zullen worden verzameld en dat de onderzoeker en de opdrachtgever de vertrouwelijkheid van deze gegevens verzekeren overeenkomstig de Belgische wetgeving ter zake. Ik stem in met de verwerking van mijn persoonlijke gegevens volgens de modaliteiten die zijn beschreven in de rubriek over het verzekeren van de vertrouwelijkheid. Ik ga ermee akkoord / Ik ga er niet mee akkoord (doorhalen wat niet van toepassing is) dat de studiegegevens die voor de hier vermelde studie worden verzameld, later zullen worden verwerkt, op voorwaarde dat deze verwerking beperkt blijft tot de context van de hier vermelde studie. Ik heb een exemplaar ontvangen van de informatie aan de deelnemer en de geïnformeerde toestemming. Naam, voornaam, datum en handtekening van de deelnemer Onderzoeker Ik ondergetekende Raf Meesen onderzoeker, verklaar de benodigde informatie inzake deze studie mondeling te hebben verstrekt evenals een exemplaar van het informatiedocument aan de deelnemer te hebben verstrekt. Ik bevestig dat geen enkele druk op de deelnemer is uitgeoefend om hem/haar te doen toestemmen met deelname aan de studie en ik ben bereid om op alle eventuele bijkomende vragen te antwoorden. Ik bevestig dat ik werk in overeenstemming met de ethische beginselen zoals vermeld in de "Verklaring van Helsinki", de "Goede praktijk" en de Belgische wet van 7 mei 2004 inzake experimenten op de menselijke persoon. Naam, Voornaam, Datum en handtekening Naan, Voornaam, Datum en handtekening van de vertegenwoordiger van de onderzoeker van de onderzoeker Titel van de studie: Modulatie van leeftijdsgerelateerde veranderingen van functionele netwerken in bimanueel visueel-motorisch leren Aanpassen / optimaliseren van leeftijdsgerelateerde veranderingen van de communicatie tussen de hersengebieden die betrokken zijn bij tweehandige oog-hand coördinatie. 49

52 III Aanvullende informatie 1: Aanvullende informatie over de organisatie van de studie Achtergrondinformatie Met de huidige studie willen we de activatie de hersengebieden tijdens het uitvoeren van een tweehandige taak aan de hand van eerder verricht onderzoek op een gerichte wijze bijsturen, optimaliseren. Doel van het onderzoek Met de bekomen informatie willen we ons een duidelijk beeld vormen van de mogelijkheden om de hersenactivatie van ouderen bij te sturen, optimaliseren om in de toekomst, gerichte revalidatie strategieën te ontwikkelen Alvorens van start te gaan met het onderzoek wordt er steeds een korte demonstratie van de gebruikte technieken voorzien, waarna u vragen kan stellen en kan beslissen om deel te nemen. Gegevensverzameling Vooraleer het onderzoek gestart wordt zal u gescreend worden om te zien of u in aanmerking komt voor het onderzoek. Metingen / interventie Een elektro-encefalogram (EEG) zal gebruikt worden om de activiteit van de betrokken hersengebieden te meten. Het toestel bestaat uit een muts met meetelectroden die op het hoofd wordt geplaatst. Hiermee wordt de activiteit gemeten van de verschillende hersengebieden. Met Transcraniële Micro Current Stimulation (TCS) wordt een microstroom aangebracht op het hoofd. Dit gebeurt door middel van 2 spons elektroden die op het hoofd bevestigd worden. Duur van het onderzoek Het experiment zal worden uitgevoerd in twee sessies met 10 dagen tussen. We verwachten dat een sessie ongeveer 2 uur in beslag zal nemen. Uiteraard voorzien we extra tijd om al uw vragen te beantwoorden, en de vragenlijst te overlopen Elke sessie duurt ongeveer 2 uur (30 minuten voor instructies / voorbereiding, 60 minuten voor de metingen en interventie (met inbegrip van kleine pauzes) en 30 minuten voor opruimen. Risico s verbonden aan deze studie Er zijn weinig of geen risico s verbonden aan deze studie. De meting van de hersenactiviteit gebeurd met een elektro-encefalogram (EEG) Dit is een niet invasieve meting (niet invasief wil zeggen dat we niet binnendringen in het lichaam, andere voorbeelden van niet invasieve metingen zijn een bloeddrukmeting en een echografie). Naast de meting van de hersenactiviteit wordt er ook een zeer lage stroom aangebracht op het hoofd (Transcranial micro Current Stimulation of afgekort TCS). Dit gebeurt door middel van 2 bevochtigde sponsjes die op het hoofd aangebracht worden. Deze niet-invasieve en nauwelijks voelbare stimulatie zal maximaal 30 minuten duren. De veiligheid van deze stroom is getest in eerder onderzoek, in deze studie blijven we onder alle voorgeschreven veiligheidslimieten (Bikson et al., 2009)* en (McCreery et al., 1990)*. Om de werking van de behandeling na te gaan zullen sommige proefpersonen echte stimulatie krijgen, terwijl anderen een schijn stimulatie krijgen aan de schijn stimulatie is geen enkel risico verbonden. Samenvattend : mogelijke risico's die kunnen optreden tijdens deze studie zijn huidirritaties onder de elektroden en lichte hoofdpijn in het geval van TCS. *(Bikson et al., 2009) : maximale stroomdichtheid: ma/cm 2 (limiet voor weefselschade: 25 ma/cm 2 (McCreery et al., 1990)) in deze studie is de gebruikte stroomdichtheid ma/ cm 2 Voordeel voor uzelf als deelnemer Er zijn geen directe voordelen verbonden aan deelname aan deze studie. Door deel te nemen aan deze studie draagt u uw steentje bij aan de wetenschap. De resultaten van deze studie kunnen reeds bekomen bevindingen bevestigen en wetenschappelijk verder ondersteunen, wat eventueel tot nieuwe behandelingsstrategieën leiden kan. Nadat de resultaten van deze studie geanalyseerd en verwerkt zijn, zal u indien u dit wenst uitgebreid geïnformeerd worden over onze bevindingen. Na elke sessie krijgt u een filmticket of boekenbon. 3: Aanvullende informatie over de bescherming en de rechten van de deelnemer aan een studie Ethisch comité Deze studie werd geëvalueerd door een onafhankelijk ethisch comité Medisch Ethisch Commitee Leuven dat een gunstig advies heeft uitgebracht. De ethische comités hebben als taak de personen die aan studies deelnemen te beschermen. Ze controleren of uw rechten als patiënt en als deelnemer aan een studie gerespecteerd worden, of de studie wetenschappelijk relevant en ethisch verantwoord is. Hierover brengen de ethische comités een advies uit in overeenstemming met de Belgische wet van 7 mei U dient het positief advies van de Ethische Comités in geen geval te beschouwen als een aansporing om deel te nemen aan deze studie.

53 Vrijwillige deelname Aarzel niet om alle vragen te stellen die u nuttig vindt voordat u tekent. Neem de tijd om er met een vertrouwenspersoon over te praten, als u dit wenst. U heeft het recht om niet deel te nemen aan deze studie of met deze studie te stoppen zonder dat u hiervoor een reden hoeft te geven, zelfs al hebt u eerder toegestemd om aan deze studie deel te nemen. Uw beslissing zal in geen geval uw relatie met de onderzoeker en de voortzetting van uw therapeutische behandeling veranderen. Als u aanvaardt om aan deze studie deel te nemen, ondertekent u het toestemmingsformulier. De onderzoeker zal dit formulier ook ondertekenen en zal zo bevestigen dat hij u de noodzakelijke informatie voor deze studie heeft gegeven. U zult het voor u bestemde exemplaar ontvangen. Kosten in verband met uw deelname U zult een vergoeding krijgen in de vorm van een filmticket of boekenbon voor uw deelname aan deze studie. Uw deelname zal echter voor u geen bijkomende kosten met zich meebrengen. Vertrouwelijkheidgarantie Uw deelname aan de studie betekent dat u ermee akkoord gaat dat de onderzoeker gegevens over u verzamelt en dat de opdrachtgever van de studie die gebruikt voor onderzoek en in het kader van wetenschappelijke en medische publicaties. De onderzoeker is verplicht om deze verzamelde gegevens vertrouwelijk te behandelen. Dit betekent dat hij zich ertoe verbindt om uw naam nooit bekend te maken bijvoorbeeld in het kader van een publicatie of een conferentie en dat hij uw gegevens zal coderen (uw identiteit zal worden vervangen door een identificatiecode in de studie). De persoonlijke gegevens omvatten geen combinatie van elementen waarmee het mogelijk is u te identificeren in overeenstemming met de Belgische wet betreffende de bescherming van de persoonlijke levenssfeer. De (gecodeerde) onderzoeksgegevens kunnen doorgegeven worden aan Belgische of andere regelgevende instanties, aan de ethische comitée. Uw toestemming om aan deze studie deel te nemen betekent dus ook dat u akkoord gaat dat uw gecodeerde gegevens gebruikt worden voor doeleinden die in dit informatieformulier staan beschreven en dat ze worden overgedragen aan bovenvermelde personen en/of instellingen. De opdrachtgever verbindt zich ertoe om de verzamelde gegevens enkel in het kader van deze studie te gebruiken. Indien u uw toestemming tot deelname aan de studie intrekt, zullen de gecodeerde gegevens die al verzameld waren vóór uw terugtrekking, bewaard worden. Hierdoor wordt de geldigheid van de studie gegarandeerd. Er zal geen enkel nieuw gegeven aan de opdrachtgever worden doorgegeven. Verzekering Conform de Belgische wet van 7 mei 2004 inzake experimenten op de menselijke persoon, is de opdrachtgever zelfs foutloos, aansprakelijk voor alle schade die de deelnemer of zijn rechthebbenden opliepen en die rechtstreeks dan wel onrechtstreeks verband vertoont met het experiment. De opdrachtgever van deze studie heeft een verzekering afgesloten die deze aansprakelijkheid dekt. Indien U schade zou oplopen ten gevolge van uw deelname aan deze studie zal die schade bijgevolg worden vergoed conform de Belgische wet van 7 mei

54 Appendix 3: Questionnaires First Name: Last Name: Subject Data and Identity Gender: M / F Date of birth and age: Dominant Hand: Right / Left Contact Information Address: Phone: Medical History Epilepsy: Migraine (w/wo aura): Metal Implants: Heart Complications: Other: Medication: 52

55 Questionnaire Session Number: 1 / 2 Before beginning: 1) How would you rate your level of attention? Place an X à Distracted Fully attentive ) How would you rate your level of fatigue? Place an X à Extremely Tired Extremely Energetic End of Block 1: 3) How would you rate your perception of the stimulation? Place an X à No feeling at all Extremely intense feeling ) How would you rate your level of attention? Place an X à Distracted Fully attentive ) How would you rate your level of fatigue? Place an X à Extremely Tired Extremely Energetic

56 End of Block 2: 6) How would you rate your perception of the stimulation? Place an X à No feeling at all Extremely intense feeling ) How would you rate your level of attention? Place an X à Distracted Fully attentive ) How would you rate your level of fatigue? Place an X à Extremely Tired Extremely Energetic End of Block 3: 9) How would you rate your level of attention? Place an X à Distracted Fully attentive ) How would you rate your level of fatigue? Place an X à Extremely Tired Extremely Energetic Do you believe you received real stimulation today? Yes No 54

57 Appendix 4: Point Angle Difference A. Whole Model (Mixed Model Log PA) 55

58 B. Interaction-effect: Group*Ratio (Tukey s) MULTIPLE COMPARISONS FOR GROUP*RATIO à LEAST SQUARES MEANS ESTIMATES: Group Ratio Estimate Std Error DF Lower 95% Upper 95% Young 1-1 1, , ,802 1, , Young 3-1 2, , ,802 1, , Old 1-1 2, , ,802 2, , Old 3-1 2, , ,802 2, , à TUKEY HSD ALL PAIRWISE COMPARISONS (Adjusted DF = 591,0) Group Ratio -Group -Ratio Difference Std Error t Ratio Prob> t Lower 95% Upper 95% Young 1-1 Young , , ,94 <,0001* -1, ,47702 Young 1-1 Old , , ,58 <,0001* -1, ,63048 Young 1-1 Old , , ,74 <,0001* -2, ,08408 Young 3-1 Old , , ,96 0,2027-0, ,12843 Young 3-1 Old , , ,13 0,0002* -1, ,32517 Old 1-1 Old , , ,15 0,0002* -0, ,17172 C. Main-effect: Block (Tukey s) MULTIPLE COMPARISONS FOR BLOCK à LEAST SQUARES MEANS ESTIMATES: Block Estimate Std Error DF Lower 95% Upper 95% 1 2, , ,119 2, , , , ,119 1, , , , ,119 1, , , , ,119 2, , à TUKEY HSD ALL PAIRWISE COMPARISONS (Adjusted DF = 591,0) Block -Block Difference Std Error t Ratio Prob> t Lower 95% Upper 95% 1 2 0, , ,08 0,0115* 0, , , , ,76 <,0001* 0, , , , ,07 0,7097-0, , , , ,67 0,0387* 0, , , , ,15 0,0002* -0, , , , ,82 <,0001* -1, ,

59 D. Interaction-effect: Ratio*Block (Tukey s) MULTIPLE COMPARISONS FOR RATIO*BLOCK à LEAST SQUARES MEANS ESTIMATES: Ratio Block Estimate Std Error DF Lower 95% Upper 95% , , ,49 2, , , , ,49 1, , , , ,49 1, , , , ,49 1, , , , ,49 2, , , , ,49 1, , , , ,49 1, , , , ,49 3, , à TUKEY HSD ALL PAIRWISE COMPARISONS (Adjusted DF = 591,0) Ratio Block -Ratio -Block Difference Std Error t Ratio Prob> t Lower 95% Upper 95% , , ,66 0,1358-0, , , , ,61 0,0001* 0, , , , ,72 <,0001* 0, , , , ,67 0,9977-0, , , , ,02 0,9707-0, , , , ,86 0,0827-0, , , , ,90 <,0001* -1, , , , ,94 0,5217-0, , , , ,05 0,4475-0, , , , ,34 0,0202* -0, , , , ,64 0,7258-0, , , , ,20 1,0000-0, , , , ,56 <,0001* -1, , , , ,11 1,0000-0, , , , ,28 <,0001* -1, , , , ,58 0,0088* -1, , , , ,75 0,6564-0, , , , ,50 <,0001* -2, , , , ,39 <,0001* -1, , , , ,69 0,0059* -1, , , , ,86 0,5818-0, , , , ,61 <,0001* -2, , , , ,70 0,6898-0, , , , ,53 0,0105* 0, , , , ,23 <,0001* -1, , , , ,84 0,5956-0, , , , ,92 <,0001* -1, , , , ,76 <,0001* -1, ,

60 Appendix 5: Track Target Difference A. Whole Model (Mixed Model Log TT) 58