| | Biodynamic Feedback Training to Assure Learning Partial Load Bearing on Forearm CrutchesAbstract Krause D, Wünnemann M, Erlmann A, Hölzchen T, Mull M, Olivier N, Jöllenbeck T. Biodynamic feedback training to assure learning partial load bearing on forearm crutches. ObjectiveTo examine how biodynamic feedback training affects the learning of prescribed partial load bearing (200N). DesignThree pre-post experiments. SettingBiomechanics laboratory in a German university. ParticipantsA volunteer sample of 98 uninjured subjects who had not used crutches recently. There were 24 subjects in experiment 1 (mean age, 23.2y); 64 in experiment 2 (mean age, 43.6y); and 10 in experiment 3 (mean age, 40.3y), parallelized by arm force. InterventionsVideo instruction and feedback training: In experiment 1, 2 varied instruction videos and reduced feedback frequency; in experiment 2, varied frequencies of changing tasks (contextual interference); and in experiment 3, feedback training (walking) and transfer (stair tasks). Main Outcome MeasureVertical ground reaction force. ResultsAbsolute error of practiced tasks was significantly reduced for all samples (P<.050). Varied contextual interference conditions did not significantly affect retention (P=.798) or transfer (P=.897). Positive transfer between tasks was significant in experiment 2 (P<.001) and was contrary to findings in experiment 3 (P=.071). ConclusionsBiodynamic feedback training is applicable for learning prescribed partial load bearing. The frequency of changing tasks is irrelevant. Despite some support for transfer effects, additional practice in climbing and descending stairs might be beneficial. DURING REHABILITATION after total hip replacement (THR) or total knee replacement (TKR), patients often must adhere to partial load bearing when using forearm crutches. The purpose is to reduce mechanical irritations during the postoperative healing phase, in which aseptic loosening of the prosthesis could occur and thus endanger the long-term healing process.1 Frequently, patients load the lower extremity of the affected side with the prescribed partial load bearing on a bathroom scale in order to adjust to it.2 The majority of patients, however, vastly exceed the limits set by the surgeon (mean, 225%2; 217%3 of the prescription). Evidence suggests several reasons for overloading, one being that the adjusting to the prescribed partial load bearing using a bathroom scale does not reflect the dynamics of walking. Furthermore, a lack of upper body and arm strength could result in a patient being unable to adhere to the prescribed load.2, 3 Eventually, many patients adopt an unfavorable technique in using the crutches by loading them too late or unloading them too early.2 This causes an overload either shortly after heel strike or shortly before the end of the stance phase.2 Proposed procedures such as extensive training in how to use a bathroom scale for loading, along with concurrent auditory feedback training, are not effective in reducing the produced ground reaction force.4 Video-based instructions5 and terminal augmented feedback training6 can positively affect motor learning. We assume that these findings, particularly those of knowledge of results7 and contextual interference8, 9, 10 research, are transferable to training in partial load bearing. We examined in 3 experiments the effects of biodynamic feedback training. Because it is not known whether subjects vastly overload the affected limb during training, especially in contextual interference conditions, we tested uninjured subjects with prescribed partial loads rather than testing patients. In our first analysis (experiment 1), we wanted to determine whether video-based instructions followed by biodynamic feedback training actually leads to better compliance with partial load bearing prescriptions. Furthermore, we tested the effect of video instructions that differed in the temporal pattern of 3-point gait. Because it is necessary to adhere to the predefined partial load bearing while performing other everyday tasks, we adapted the training to another laboratory experiment (experiment 2) that included going upstairs and downstairs. According to the contextual interference hypothesis, changing tasks more frequently should impair acquisition performance while enhancing retention and transfer. We also examined whether walking practice leads to transfer effects11 on stair tasks (experiment 3). Experiment 1  Augmented feedback is crucial for motor learning.12 A well-investigated, major class of augmented feedback is knowledge of results as verbal (or verbalizable) and terminal information about movement outcome. For example, this could be knowledge of the length of a long jump trial in athletics, or the ground reaction force of the affected lower extremity when training in partial load bearing. In addition to motivational functions, knowledge of results also guides trials. If the target value is known, knowledge of results provides some kind of error information that can be used on the next attempt to change the movement to make it more accurate. This is called the guidance function of knowledge of results. Performance during acquisition of a task will be improved with frequent reporting of results. The learner may, however, become dependent on this augmented information if knowledge of results is given too frequently during acquisition. As a result, it can degrade learning when a subject is tested without augmented information. Thus, reducing the frequency of providing knowledge of results is beneficial in learning motor tasks. This becomes relevant when the critical movement outcome is not available within the target movement. Movement outcome for partial load-bearing gait as the vertical ground reaction force will not be available to patients in everyday life; consequently, they should benefit from being given knowledge of results less frequently. Reduced frequency is supposed to lead to a deeper processing of intrinsic information (eg, proprioceptive) and a more distinctive error detection mechanism so that the learner can make more accurate estimates of his/her movement outcome without receiving knowledge of results after training is complete.7 Winstein and Schmidt13 showed that a systematically lowered knowledge of results frequency across practice (fading) enhances motor learning. The knowledge of results fading schedule offers the advantage of leading to a rapid approximation to the target movement in the beginning because augmented information is given frequently; the learner is thus deterred from becoming dependent on knowledge of results when its relative frequency is gradually reduced at the later acquisition phase. Furthermore, knowledge of results should be given only after a delay of about 5 to 20 seconds, because instantaneous knowledge of results also represses the use of intrinsic information and results in degraded learning, as described earlier.7, 12 We expect these findings of motor learning research and their implications to be useful in learning the skill of adhering to a prescribed partial load while walking on crutches. In general, video-based instructions are appropriate for motor learning. Instructions can be optimized with expedient application of slow-motion pictures, attention-directing elements (eg, circles, arrows), freeze images, and text insertions that highlight crucial movement aspects.5 Our purpose in this study was to determine the effect of biodynamic feedback training with knowledge of results and didactically designed video instruction on adjusting to a prescribed partial load bearing. We hypothesized that this intervention would significantly reduce the deviation from the prescribed load. Previous findings indicate that the temporal pattern of heel strike and crutches’ touch down, as well as toe-off and crutches’ take-off, affect the amount of overloading of the lower extremity with a prescribed partial load.2 Following these findings, the deviation from the prescribed load increases with a later crutches’ touch down (relating to heel strike) and earlier crutches’ take-off (relating to toe-off). In rehabilitation practice, Jöllenbeck and Schönle2 noticed that the lower extremity’s vertical ground reaction force increases more sharply than do the crutches’ forces. We hypothesized, therefore, that an instruction video that emphasizes the crutches’ touch down before heel strike and take-off before toe-off (asynchronous gait pattern) would result in less overloading than would an instruction video that teaches an approximately synchronous gait pattern. Furthermore, later touch down and earlier take-off with the crutches should be correlated with a higher amount of overloading. Experiment 2 Contextual interference refers to settings in which more than 2 tasks are practiced in 1 session (eg, different movement patterns such as level walking and stair climbing with forearm crutches). Contextual interference is defined as high if tasks are alternated from trial to trial in a random order (tasks are unpredictable for the learner) or serial (fixed and predictable order) during a practice session. In contrast, contextual interference is defined as low if tasks are practiced in blocked order so that several trials of 1 task are performed before the next task is practiced. In general, high contextual interference during practice leads to diminished acquisition performance while enhancing learning, as measured in retention and transfer tests.11 After Battig14 formulated the contextual-interference hypothesis within verbal learning, it was first investigated in motor learning by Shea and Morgan.15 In contrast to the acquisition findings, the retention test results of that study demonstrated superior performance for random acquisition conditions (high contextual-interference level) in comparison with blocked acquisition conditions (low contextual-interference level). This reversal of performance from acquisition to retention is the core of the contextual interference effect. Additional transfer was greater for high interference acquisition. The study by Lee and Magill16 supported these findings. Additionally, they demonstrated that random and serial schedules lead to similar advantages in retention, in contrast to blocked-ordered trials. Contextual-interference effects have been explained by numerous hypotheses. The elaboration hypothesis,15, 17 the action plan reconstruction hypothesis,16 and the usefulness hypothesis18 are 3 of the most accepted. According to the elaboration hypothesis, random practice enables intra-item (within several trials of 1 task) and inter-item (between different tasks) elaborative and distinctive processing because multiple tasks are in the memory at the same time. This leads to more distinctive and elaborate processing. The action plan reconstruction hypothesis suggests that random conditions cause the action plan to be forgotten because of intervening trials of other tasks. The learner must reconstruct the action plan for subsequent trials. Under blocked conditions, the action plan is remembered, so there is no need to reconstruct it. Researchers argue that forgetting and regenerating the action plan induces additional processing. According to the usefulness hypothesis of Wulf and Schmidt,18 changing the task after every trial reduces the usefulness of knowledge of results, which is similar to schedules with reduced frequency of knowledge of results (see experiment 1). In a random schedule, knowledge of results is not directly applicable to the following trial of the same task. In general, a high contextual-interference level is deemed to be more demanding for the learner, thus resulting in enhanced cognitive activity and less forgetting than with blocked schedules. Contextual interference has been studied among difference motor tasks. Review articles conclude that the generalizability of this phenomenon is limited.10 Contextual-interference effects appear to occur depending on different tasks and subject characteristics. So a high contextual-interference level is detrimental in certain settings and for some learners. Inexperienced adults or children appear to benefit from moderate contextual interference conditions.9 There is also some evidence for benefits of moderate contextual interference for learners with low task proficiency. Proteau, et al19 combined blocked and random schedules, which resulted in better retention performance without worsening acquisition performance in learning a multi-segmented motor task. Therefore, Landin and Hebert20 postulated the contextual-interference continuum and, hence, the optimal schedule is determined by the learner’s level of proficiency. Our purpose in this experiment was to validate the efficiency of feedback training for learning partial load bearing with a longer retention interval than in experiment 1. Further, we intended to examine whether changing tasks more frequently (increased contextual-interference level) leads to less deviation from a prescribed load in retention and transfer tests than under blocked practice conditions, as postulated by the contextual interference hypothesis. Based on the findings of motor learning research, we hypothesized that a high contextual-interference level should be beneficial for learning partial load bearing if more than 1 movement pattern must be learned, such as level walking of ascending and descending stairs with forearm crutches. Besides contextual-interference effects, we were interested in the influence of arm force and age on performance because others have assumed these variables to be responsible for overloading.2, 3 Experiment 3 Results of experiment 2 showed that practicing partial load bearing in level walking and stepping down a stair with biodynamic feedback reduces both the performance of these practiced tasks and also the performance of the stepping up transfer task. The purpose of this experiment was to examine whether practicing partial load bearing by walking with crutches leads only to improved performance in stepping up and down stairs in the sense of positive transfer effects.11 This appears to be important for rehabilitation after THR and TKR because patients must adhere to the prescribed partial load in different situations. If transfer effects appear, it is questionable whether there is a need to additionally practice other tasks. Methods Experiment 1 Twenty-four uninjured subjects (12 men, 12 women, mean ± standard deviation [SD] body mass, 69.1±12.3kg), with an age range from 19 to 27 years (mean, 23.2±2.3y) who had never used crutches were evenly assigned to 2 random samples, synchronous and asynchronous. Participation was voluntary (in all 3 experiments) and we conducted the study in accordance with the Declaration of Helsinki. All data for all subjects were treated with confidentiality. Subjects in experiment 1 were instructed to load 1 foot with exactly 200N, which is the most common prescription given by surgeons.2 They could choose the side to be offloaded. The instructions included a scale procedure and 1 of 2 different video instructions showing side-view perspective of a subject performing a 3-point partial load-bearing gait pattern. The scale procedure simulated common clinical practice, with the subjects using forearm crutches to load the chosen side with about 200N on the scale 3 times for 3 seconds, controlling the load while watching the scale display. They were instructed to aim for the 20kg marking on the display. As opposed to dynamic gait, the subject performed the partial load bearing in a static posture. Accordingly, the effects of this procedure on learning partial load bearing should be limited.2 The video instructions (≈1min) differed with respect to the advice concerning the time slices between heel strike and crutches’ touch down (d1), as well as toe-off and crutches’ take-off (d2). One video with a synchronous pattern emphasized the synchrony of heel strike and touch down (d1=0) and, accordingly, toe-off and take-off (d2=0). Another video showing an asynchronous gait pattern advised the subjects to touch the ground with the crutches first (d1>0) and to lift the foot before lifting the crutches (d2>0). The subjects performed the observed gait pattern 10 times. The entire instruction block (video plus performing the gait pattern) was accomplished twice. The feedback training consisted of 20 discrete steps at a self-selected walking speed with 50% knowledge of results regarding the maximum vertical ground reaction force (maximum Fz) given according to a fading schedule (fig 1). Only 1 step was performed on each trial. Then the investigator verbally provided the subject with knowledge of results for that single step, with a 5-second delay after the step. The intertrial interval was consistently 25 seconds. The vertical ground reaction force was measured on a force plate (Kistler).a We used a charge amplifier and an analog-to-digital converter to make the data usable for analysis with the software Simi Motion.b Data were recorded at 300Hz. The dependent variable was the absolute error as the absolute deviation of realized vertical ground reaction force and 200N. We considered the initial performance as the first 3 steps’ mean of the absolute error; acquisition was considered as the mean (absolute error) of steps 18, 19, and 20. The retention test consisted of 3 additional steps without knowledge of results after a 10-minute interval. To standardize that interval, subjects had to solve spatial-sense tasks on a personal computer. We used the software package SPSSc for our statistical evaluation. Differences in absolute error between the 2 samples were analyzed with a t test for independent samples. Learning was evaluated with a 1-way analysis of variance (ANOVA) for repeated measures with the factor phase (experiment 3). We analyzed the influence of d1 and d2 on absolute error by calculating Pearson correlations. We set the level of significance at .05 for all statistical tests. Experiment 2 Subjects were 30 uninjured women and 34 uninjured men who had not used crutches in the past 5 years. They ranged in age from 25 to 74 years (mean ± SD, 43.6±13.2y) and weighed 77.3±14.3kg. We assigned subjects to 4 sample groups with differing contextual-interference levels during practice sessions, parallelized by relative arm force (first priority), age, weight, and sex. Tasks to be practiced were walking and stepping down 1 stair on crutches. The transfer task was stepping up 1 stair. Subjects were instructed to load the right lower extremity with 200N. Tasks were performed on a dynamometric gangway (Kistlera) with an optional stair. We considered the tasks to be ruled by different generalized motor programs. Subjects practiced partial load bearing by changing the 2 tasks either in a serial order or after every 7th, 14th, or 28th trial. The experiment consisted of 4 sessions, with 24 to 48 hours between each session (fig 2). We tested maximum isometric arm force during the first session. Relative arm force was designated as being maximum isometric arm force (in newtons) divided by the difference of body weight and 200N. Therefore, subjects were fixed on a custom-made chair, which allowed them to produce force with crutches onto a forceplate under static conditions. Afterward, subjects underwent the scale procedure (see experiment 1). The video instructions emphasized the order of events (see experiment 1; asynchronous). After watching the introduction of 1 task, subjects practiced it 5 times without knowledge of results. They accomplished this sequence twice for each task. The 2 following sessions consisted of 28 acquisition trials. Walking and stepping down was practiced in specific schedules of tasks (see fig 2). Subjects received knowledge of results about produced ground reaction force on 50% of the trials, according to a fading schedule (see experiment 1). We gave knowledge of results as kilograms in rounded full numbers for better interpretation. We measured pretest (M1), acquisition (M2, M4), and retention (M3, M5) performances for all tasks (see fig 2). Subjects performed 2 trials of each task within 1 measure. The dependent variable was the absolute error as the absolute deviation of realized vertical ground reaction force and 200N. We evaluated the general effects of the intervention with 2-way ANOVA (2 [task] × 2 [phase]) for repeated measures. For statistical analysis of contextual-interference effects on acquisition and retention, a 3-way ANOVA for repeated measures was calculated with a 4 (sample) by 2 (practiced task) by 3 (phase) design. For contextual-interference effects on transfer a 2-way ANOVA for repeated measures was calculated with a 4 (sample) by 2 (phase) design. We used Pearson correlation coefficients to analyze the influence of relative arm force and age on learning partial load bearing. The level of significance was set at .05 for all statistical tests. Experiment 3 We instructed 5 men and 5 women (mean ± SD, 40.3±17.6y; weight, 72.9±16.0kg), as described in experiment 2. Subjects practiced walking on crutches for 28 trials. Afterward, the performance in stepping down and stepping up was pretested and measured on a 24-hour delayed retention test. Retention test measures of reduced absolute error in stepping up and stepping down 1 stair would indicate a positive transfer performance. Reduction of absolute error was analyzed using the t test for dependent samples (practice task) and 2-way ANOVA (2 [transfer task] × 2 [phase]). Results Experiment 1 Contrary to our hypothesis, absolute error of the 2 samples did not differ significantly during the first 3 steps (initial performance) after the instruction block (mean ± SD: synchronous, 133.5±90.5N; asynchronous, 107.9±68.7N; t22=.78, P=.444, d=.32). We then consolidated the 2 samples for further evaluation. The time slices d1 and d2 of all steps correlated to the produced vertical ground reaction force. The effect sizes of the correlations, however, did not support the assumption of a strong influence on the produced ground reaction force (d1: r=−.23, P<.001; d2: r=.08, P=.027). Subjects significantly reduced absolute error during the feedback training (effect phase [experiment 3]; initial performance, acquisition, retention; Greenhouse-Geisser, F1.40,32.13=14.65, P<.001, η2=.39). Subjects improved performance because of the significant reduction of absolute error during the feedback training, from 120.7±79.6N to 54.5±35.4N (t23=4.30, P<.001, d=.88). After the 10-minute interval, absolute error did not change significantly (t23=−.94, P=.356, d=−.19) (fig 3). This supports our hypothesis regarding the effectiveness of the feedback training procedure. Experiment 2 Absolute error of the 4 groups for M1 through M5 and the acquisition phases are illustrated in figure 4. From M1 to M5 all groups reduced the absolute error for the tasks of walking (mean range, 258.2–73.3N) and stepping down (mean range, 211.0–77.9N) (main effect, phase: F1,60=92.76, P<.001, η2=.61) and the transfer task (mean range, 197.9–88.8N) (main effect, phase: F1,60=28.05, P<.001, η2=.32). After the first feedback training sequence, the absolute error was significantly reduced from M1 to M3 for the walking and stepping down tasks (main effect, phase: F1,60=90.94, P<.001; η2=.60) and the transfer task (main effect, phase: F1,60=30.40, P<.001, η2=.34) including all groups. The second feedback training sequence (M3 to M5) resulted in a further significant reduction of absolute error for walking and stepping down (main effect, phase: F1,60=12.85, P=.001, η2=.18), but not for the transfer task (main effect, phase: F1,60=2.67, P=.108, η2=.043). The interaction phase by contextual-interference level concerning M1, M4, and M5 does not indicate the expected contextual-interference effects for either the acquisition and retention tests (phase by contextual-interference level: Greenhouse-Geisser, F3.5,67.07=.37, P=.798, η2=.018) or transfer tests (phase by contextual-interference level: Greenhouse-Geisser, F3.97,79.46=.27, P=.897, η2=.013). For M5, relative arm force correlated negatively with the absolute error for the walking (r=−.28, P=.013) and stepping down (r=−.35, P=.002) tasks, but not with stepping up (r=−.19, P=.070). There were higher correlations in the effects of age on absolute error at M5 for walking (r=.45, P<.001), stepping down (r=.40, P=.001), and stepping up (r=.36, P=.002). Experiment 3 Absolute error of walking was reduced significantly from pretest (mean ± SD, 249.1±200.3N) to retention test (t9=3.79, P=.004, d=1.20) (fig 5). Contrary to our hypothesis, reduction of absolute error concerning transfer tasks fell just short of the level of statistical significance (main effect, phase [2]; F1,9=4.19, P=.071, η2=.32). For the transfer tasks stepping down (reduction from 163.1±135.8N to 122.3±134.1N) and stepping up (reduction from 192.0±161.5N to 145.3±146.4N) absolute error tended to decrease. Transfer tasks did not differ significantly (2 [task] × 2 [phase]; F1.9=.28, P=.611, η2=.030). Discussion Training procedures using a bathroom scale or concurrent auditory feedback do not appear to achieve satisfactory results in learning partial load bearing.2, 4 Youdas et al21 found a satisfying acquisition effect of a bathroom scale procedure with healthy subjects but they did not measure learning performance because their subjects were not tested after a retention interval. Furthermore, their chosen target load of 50% of body weight is less critical than lower target loads.2 We were able to show that training procedures containing didactical designed video instructions and terminal augmented verbal feedback concerning the deviation from the prescribed load appears to be helpful when the alignment to a prescribed partial load has to be learned. Experiment 1 Experiment 1 exemplified the effect of biodynamic feedback training with knowledge of results and video instruction. We evaluated a positive effect for controlling the predetermined partial load bearing. Our findings concerning the influence of the time slices between heel strike and crutches’ touch down (d1), as well as toe-off and crutches’ take-off (d2), are contrary to previous measurement data,2 which indicated stronger influence of the time slices on the affected limb’s vertical ground reaction forces. The contradiction may be the result of the different length of time slices in our experiment (mean ± SD: d1, .276±.292s; d2, .069±.104s) compared with former investigations2 (mean: d1, −.052s; d2, −.083s; SD not provided). We assume that the video instructions reduced timing errors sufficiently. Remaining differences in d1 and d2 appear to have no critical influence on the ground reaction force. Experiment 2 The results of experiment 2 do not support our contextual-interference hypothesis.14 Acquisition nor retention findings did not show significant differences between samples with varied acquisition schedules. The level of contextual interference seems to be marginal within the setting of learning partial load bearing for different forms of locomotion. We explain the absence of contextual-interference effects by high task similarity and the low number of practiced tasks, which leads to an insufficient level of contextual interference. In the sense of the challenge-point theory,22 the created contextual-interference level within the serial conditions was not more demanding than the blocked conditions. Therefore, there were no detrimental acquisition or superior retention or transfer scores. In most studies that have shown the contextual-interference effect, 3 or more tasks were practiced. In addition, a high similarity of tasks appears to diminish the contextual-interference effect.23 The findings of experiment 2, however, reconfirm the positive effect of the feedback training for elderly people and the tasks of climbing and descending stairs. Especially considering the effects of the first feedback session, practice in walking and stepping down seems to enable a positive transfer for stepping up. Arm force had less influence on adherence to partial load bearing than we anticipated. We interpret the correlations between age and performance as being a result of age-related decreasing ability of force control that results from degenerative neuronal changes,24 as well as changes in morphology, number, and density of mechanoreceptors, which leads to a degraded tactile sensitivity in the aging.25 Experiment 3 In experiment 3, the reduction of absolute error for transfer tasks fell just short of the level of statistical significance. So we cannot neglect transfer effects, particularly as effect sizes indicate a certain amount of transfer from walking to stepping down and stepping up. Study Limitations Because we investigated uninjured subjects in a laboratory context our findings may have some limitations regarding transferability to rehabilitation practice. It would be of great interest to investigate effects of partial load-bearing training with patients outside of the laboratory and with longer retention intervals. Rehabilitation Implications We state that biodynamic feedback training might be beneficial for patients who must adhere to predetermined partial load bearing. The training should include video instructions for everyday movements on crutches that illustrate permanent unloading of the affected leg during the stance phase. For learning partial load bearing, patients may benefit from biodynamic feedback training that provides knowledge of results. The feedback should be provided terminally, with a delay of about 5 seconds, so that the learner can process his/her inherent sources of information and contrast them with the knowledge of results. Knowledge of results, however, should be provided for only 50% of the steps taken to prevent dependence on augmented feedback. Twenty-eight trials sufficiently reduce deviation from the prescribed load. The second training session leads to an average additional reduction of 25.2N. Therefore, it appears that 2 or more sessions are not necessary. If the prescribed load is changed during rehabilitation, further sessions might be useful. Our results imply that patients in rehabilitation might benefit from biodynamic feedback training for multiple skills such as walking and going upstairs and downstairs. Conclusions Current experiments with patients analyze the effect of the biodynamic feedback training during THR and TKR rehabilitation programs. These experiments will improve the external validity of our findings by comparing laboratory performance with additional mobile measures in everyday situations. Further, we want to review our findings with respect to the influence of arm force and time slices d1 and d2. Patients must unload the affected side for a longer period of time, so exhaustion could lead to greater influence of arm force. Suppliers Acknowledgment All participants received a complimentary Thera-band from Thera-Band GmbH for completing the study. References 1. 1Wirtz D, Niethard F. Ursachen, Diagnostik und Therapie der aseptischen Hüftendoprothesenlockerung—eine Standortbestimmung. Z Orthop. 1997;135:270–280. MEDLINE |
CrossRef
2. 2Jöllenbeck T, Schönle C. Die Teilbelastung nach Knie- oder Hüft-Totalendoprothese—die Unmöglichkeit der Einhaltung, ihre Ursachen und Abhilfen. Z Orthop. 2005;143:124–128. MEDLINE |
CrossRef
3. 3Vasarhelyi A, Baumert T, Fritsch C, Hopfenmuller W, Gradl G, Mittlmeier T. Partial weight bearing after surgery for fractures of the lower extremity—is it achievable?. Gait Posture. 2006;23:99–105. Abstract | Full Text |
Full-Text PDF (240 KB)
|
CrossRef
4. 4Warren CG, Lehmann JF. Training procedures and biofeedback methods to achieve controlled partial weight bearing: an assessment. Arch Phys Med Rehabil. 1975;56:449–455. MEDLINE 5. 5Daugs R, Blischke K, Olivier N, Marschall F. Beiträge zum visuomotorischen Lernen im Sport. Schorndorf: Hofmann; 1989;. 6. 6Schmidt RA, Wrisberg CA. Motor learning and performance: a problem-based learning approach. 3rd ed. Champaign: Human Kinetics; 2004;. 7. 7Salmoni A, Schmidt R, Walter C. Knowledge of results and motor learning: a review and critical reappraisal. Psychol Bull. 1984;95:355–386. MEDLINE |
CrossRef
8. 8Magill RA, Hall KG. A review of the contextual interference effect in motor skill acquisition. Hum Mov Sci. 1990;9:214–289. 9. 9Brady F. A theoretical and empirical review of the contextual interference effect and the learning of motor skills. Quest. 1998;50:266–293. 10. 10Brady F. Contextual interference: a meta-analytic study. Percept Mot Skills. 2004;99:116–126. MEDLINE |
CrossRef
11. 11Magill RA. Motor learning: concepts and applications. 7th ed. New York: McGraw-Hill; 2004;. 12. 12Schmidt RA, Lee TD. Motor control and learning: a behavioral emphasis. 3rd ed. Champaign: Human Kinetics; 1999;. 13. 13Winstein C, Schmidt RA. Reduced frequency of knowledge of results enhances motor skill learning. J Exp Psychol Learn Mem Cogn. 1990;16:677–691.
CrossRef
14. 14Battig WF. Facilitation and interference. In: Bilodeau EA editors. Acquisition of skill. New York: Academic Pr; 1966;p. 215–244. 15. 15Shea JB, Morgan RL. Contextual interference effects on the acquisition, retention and transfer of a motor skill. J Exp Psychol. 1979;5:179–187. 16. 16Lee TD, Magill RA. The locus of contextual interference in motor-skill acquisition. J Exp Psychol Learn Mem Cogn. 1983;9:730–746.
CrossRef
17. 17Shea JB, Zimny ST. Context effects in memory and learning movement information. In: Magill RA editors. Memory and control of action. Amsterdam: North-Holland; 1983;p. 345–366. 18. 18Wulf G, Schmidt RA. Contextual interference effects in motor learning: evaluating a KR-usefulness hypothesis. In: Nitsch JR, Seiler R editor. Bewegung und Sport. Psychologische Grundlagen und Wirkungen. Vol 2: Bewegungsregulation und motorisches Lernen. St. Augustin: Academia; 1994;p. 304–309. 19. 19Proteau L, Blandin Y, Alain C, Dorion A. The effects of the amount and variability of practice on the learning of multi-segmented motor task. Acta Psychol. 1994;85:61–74. 20. 20Landin D, Hebert EP. A comparison of three practice schedules along the contextual interference continuum. Res Q Exerc Sport. 1997;68:357–361. MEDLINE 21. 21Youdas JW, Kotajarvi PT, Padgett DJ, Kaufmann KR. Partial weight-bearing gait using conventional assistive devices. Arch Phys Med Rehabil. 2005;86:394–398. Abstract | Full Text |
Full-Text PDF (164 KB)
|
CrossRef
22. 22Guadagnoli MA, Lee TD. Challenge point: a framework for conceptualizing the effects of various practice conditions in motor learning. J Mot Behav. 2004;36:212–224. MEDLINE |
CrossRef
23. 23Wood CA, Ging CA. The role of interference and task similarity on the acquisition, retention, and transfer of simple motor skills. Res Q Exerc Sport. 1991;62:18–26. MEDLINE 24. 24Cole JC, Rotella DL, Harper JG. Tactile impairments cannot explain the effect of age on a grasp and lift task. Exp Brain Res. 1998;121:263–269. MEDLINE |
CrossRef
25. 25Thornburry JM, Mistretta CM. Tactile sensitivity as a function of age. J Gerontol. 1981;36:34–39. MEDLINE a Department of Sport and Health, Sports Sciences, Faculty of Science, University Paderborn, Paderborn, Germany b Biomechanics Laboratory, Klinik Lindenplatz, Bad Sassendorf, Germany.  Reprint requests to Daniel Krause, MSc, Universität Paderborn, Warburgerstraße 100, 33098 Paderborn, Germany
No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. PII: S0003-9993(07)00231-6 doi:10.1016/j.apmr.2007.03.022 © 2007 American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT
