An Older Adult Study of Concurrent Visual Feedback in Learning Continuous Balance

Abstract

This investigation manipulated concurrent visual feedback for older adults learning to perform a continuous balance task. We randomly assigned 21 older adults to one of three knowledge of results (KR) groups with varying concurrent but always 100% terminal feedback percentages during acquisition: 100% Concurrent and Terminal (100% C&T), 50% Concurrent and 100% Terminal (50% C & 100% T), and 0% Concurrent and 100% Terminal (0% C & 100% T). The continuous balance task involved learning to maintain balance on a circular platform that moved simultaneously on both anterior/posterior and medial/lateral axes. We tested participants before acquisition phase learning and two days afterwards (retention test). At retention testing, participants in the 50% C & 100% T and those in the 0% C & 100% T conditions demonstrated significantly greater accuracy and stability than did participants in the 100% C&T condition. These findings extend previous research with young adults in supporting the guidance theory that too frequent a provision of concurrent knowledge of results negatively affects learning.

Keywords

ageing balance knowledge of results motor skills retention or recall

Introduction

Motor skills are integral to daily activity across the life span. People of all ages (infants to older adults) perform motor skills, ranging from simple (e.g., reaching for a cup) to complex (e.g., driving a car). Historically, the vast majority of research in motor skill learning has utilized young adult participants in studies of various acquisition variables such as knowledge of results (KR), blocked and random practice, and massed and distributed practice (Adams, 1987). There is a current concensus from prior research that movement response KR is critically significant to motor learning, second only to physical practice for performance enhancement. Questions regarding finer iterations of KR delivery methods—augmented, verbal, terminal, error information—have been the subject of numerous investigations and have focused on all learning phases of acquisition, retention, and transfer (for reviews, see Adams, 1971; Bilodeau, 1966; Magill, 1994, 2001; Newell, 1976; Salmoni, Schmidt, & Walter, 1984; Swinnen, 1996; Wulf & Shea, 2004). Yet, research involving motor skill learning with older adults has been rare (Rabbitt, 1997), with a few exceptions (Swanson & Lee, 1992; Wishart & Lee, 1997; Wishart, Lee, Cunningham, & Murdoch, 2002; for a review of studies of age-related differences, see Voelcker-Rehage, 2008).

The timing issue of when to provide KR has received considerable interest in studies with young adults. With respect to timing, feedback may be categorized as (a) concurrent, delivered during motor skill execution trials, or (b) terminal, delivered after a completed motor execution trial. A potential advantage of concurrent feedback is its allowance for performers to make immediate corrections. Typically, investigations of this timing issue have compared a participant group that receives both 100% concurrent and 100% terminal visual feedback in the acquisition (practice) phase to a group that receives only 100% terminal feedback (Behrman, Vander Linden, & Cauraugh, 1992; Ranganathan & Newell, 2009; Schmidt & Wulf, 1997; Vander Linden, Cauraugh, & Greene, 1993; Verschueren, Swinnen, Dom, & DeWeerdt, 1997; Winstein et al., 1996). These studies have shown that receipt of both 100% concurrent visual feedback and 100% terminal feedback results in superior performance compared with receipt of only 100% terminal feedback. Thus, there is an established positive practice effect of receiving concurrent feedback in the acquisition phase, as evidenced by better error correction even during practice. However, these studies have also shown that on no-KR retention (posttest) testing, participants receiving 100% terminal feedback have outperformed those receiving both the 100% concurrent visual feedback and 100% terminal feedback.

To understand the implications of these findings, it is important to appreciate the distinction between practice performance and learning. Specifically, the performance-learning distinction refers to the well-established finding that performance measures during the acquisition phase may mask the true learning that has occurred when learning is measured by retention or transfer tests following acquisition (Edwards, 2011). Concurrent visual feedback has produced very strong guiding effects in the acquisition phase because the feedback is readily available while learning occurs. This leads participants in the concurrent visual feedback group to outperform those in other terminal feedback groups (e.g., 100% KR, 50% KR) in the acquisition phase. However, when a no-KR retention test is later administered and the concurrent visual feedback is removed, participants who received terminal feedback during acquisition have outperformed those who received concurrent feedback (Behrman et al., 1992; Ranganathan & Newell, 2009; Schmidt & Wulf, 1997; Vander Linden et al., 1993; Verschueren et al., 1997; Winstein et al., 1996). Thus, the effect of receiving concurrent visual feedback in the acquisition phase is temporary and does not result in a permanent learning effect on no-KR retention testing.

As noted earlier, the vast majority of motor learning KR studies have focused on young adults (for exceptions, see Behrman et al., 1992, Piper & Rone, 2001; and Wishart et al., 2002). Studies involving the use of concurrent visual feedback in older adults have shown mixed results. Behrman et al. (1992) compared healthy older adults (M age = 69.00 years) receiving different feedback timing on an isometric elbow extension and force production task. In this study, the concurrent feedback group received 100% concurrent visual feedback during each movement and 100% terminal feedback after each acquisition trial. The 100% terminal group received 100% feedback after each acquisition trial. The 50% terminal group received 50% feedback (after every other acquisition trial) in the acquisition phase. As in studies of younger adults, participants in the concurrent group performed with significantly smaller error during acquisition than those in the two terminal feedback groups; but on no-KR retention testing, participants in the two terminal feedback groups outperformed those in the concurrent visual feedback group. In contrast, however, Wishart et al. (2002) compared healthy older adults (M age = 66.15 years) on a bimanual coordination task, and, in this study, older adults who received concurrent visual feedback performed significantly more consistently even on no-KR retention testing than those in the terminal feedback group. Piper and Rone (2001) compared healthy older adults (M age = 74.90 years) on a limited weight bearing task and found no significant differences on no-KR retention testing between participants receiving concurrent visual feedback and those in the terminal feedback group.

Given relatively scarce and mixed results research on feedback timing in older adults’ motor learning, this study sampled older adults and manipulated concurrent visual feedback to learn a very relevant motor skill for this population, continuous balance. Similar to past studies, we compared a group that received both 100% concurrent visual feedback and 100% terminal feedback (100% C&T) to one that received only 100% terminal feedback group (0% C & 100% T), and, in an attempt to extend Park, Shea, and Wright’s (2000; Experiment 1) findings, we also employed a group that received 50% concurrent visual feedback (every other acquisition trial) and 100% terminal feedback (50% C & 100% T). Park et al. (2000; Experiment 1), in their study of younger adults, found that their 100% C&T group performed with significantly smaller error in the acquisition phase, but, on no-KR retention testing, their 50% C & 100% T and their 0% C & 100% T groups performed with smaller error. We expected that our 100% C&T group would perform with significantly better accuracy and stability in the acquisition phase, but with significantly worse accuracy and stability on no-KR retention testing when compared with the 50% C & 100% T and 0% C & 100% T groups.

Our findings address theoretical implications by examining the guidance hypothesis of manipulating feedback timing in terms of concurrent and terminal KR during practice or acquisiton phase learning. As discussed by Adams (1971) and Schmidt (1975), it was once believed that making KR more frequent, more immediate, and more precise would always enhance motor skill learning. However, Schmidt (1991) proposed and Lai and Shea (1999) refined the guidance hypothesis to explain repeated research findings that concurrent feedback benefited only acquisition performance while harming retention learning. The rationale behind the guidance hypothesis is that a constant presentation of KR may create a learner dependence on external feedback, resulting in the learner’s disregard of his or her own internal feedback. Our findings should also be useful to instructional programs designed to improve older adults’ balance learning in order to prevent falls. Of note, the Centers for Disease Control and Prevention (2012) has estimated a fall prevalence in adults aged 65 years and older of one in every three older adults; and among those 65 years or older, falls are the leading cause of injury-related death. Florence et al. (2018) estimated that the total medical costs for falls was more than $50 billion in 2015. These considerations make continuous balance a critically relevant skill for this population. In 1996, Winstein et al. suggested that physical therapists often used concurrent feedback to train sensorimotor skills, but others have suggested that this rehabilitative approach is suboptimal (Carr, Zachariah, Weir, & McNevin, 2011). Therefore, research is needed with older adults that manipulates different concurrent feedback methods to learn a continuous balance task, both for a general understanding of developmental influences on motor learning and practically, to best inform rehabilitation methods.

Method

Participants

Twenty-one older adults (M age = 82.67, SD = 7.18 years) participated in the experiment. All participants provided written informed consent, were naïve to the purpose of the experiment, and had no prior experience with the balance apparatus. All participants described themselves as healthy with no existing neurological or physical problems. The experiment was conducted in accordance with the institution’s ethical guidelines for research involving human participants.

Apparatus

We used the Biodex Balance System (Biodex Medical Systems, Inc., Model #945-300, Shirley, NY) consisting of a circular platform that moved in the anterior or posterior and medial or lateral axes simultaneously and provided an assessment of dynamic postural stability. The Biodex Balance System has been shown to have intertester reliability of .82 and intratester reliability of .70 (Hinman, 2009; Schmitz & Arnold, 1998). The circular platform allowed movements in a 360° range and up to 20° of tilt at any point in the range. The stability (stiffness) level of the platform was set at eight. A setting of eight is the most stable platform setting, whereas a setting of one is the least stable setting. The display module (Figure 1) showed the percentage of time participants spent in each zone during trials. The Target Zones, A, B, C, and D, were equal to specific ranges of deflection and radiate in concentric circles from the center of the foot platform as follows: Zone A = 0° to 5° foot platform deflection from level, Zone B = 6° to 10° foot platform deflection from level, Zone C = 11° to 15° foot platform deflection from level, and Zone D = 16° to 20° foot platform deflection from level. Participants were instructed that the goal was to maintain the cursor on the center of the display module for as long as possible by balancing in Target Zone A.

Figure 1.

Example of the criterion response display and participant response.

Procedure

Participants were randomly assigned to one of three KR groups. The 100% C&T group received visual feedback during or concurrent with 100% of the acquisition phase trials and after or terminal with 100% of the trials. The 50% C & 100% T group received visual feedback during alternate trials or 50% concurrent feedback and received feedback after all acquisition trials or 100% terminal feedback. The 0% C & 100% T group received no concurrent visual feedback in the acquisition phase but received visual feedback after all acquisition trials or 100% terminal feedback. Concurrent visual feedback was presented as a trace of the movement pattern, and terminal feedback consisted of the percent time in each zone and each quadrant (Figure 1).

For all trials, participants were instructed that the goal was to maintain balance in Target Zone A for as long as possible. Participants were required to stand in a comfortable upright position with feet shoulder width apart and arms at sides. During the no concurrent visual feedback trials, the display module was covered to eliminate concurrent visual feedback while balancing on the circular platform. In contrast, during the concurrent visual feedback trials, participants were allowed to view the display module. Participants were not instructed on how to focus their vision while performing the balance task. All trials were 20 second in duration with a 35-second rest period. The interval of time between the completion of a trial and the presentation of terminal feedback was approximately 30 seconds. The interval of time between the presentation of terminal feedback and the beginning of the next trial was approximately 5 seconds. During the rest period, participants remained standing while the stability platform was locked in a neutral position.

On the first day, participants performed five, 20-second trials without concurrent or terminal visual feedback. The pretest data were used to determine whether participants in the KR groups were statistically similar prior to participating in the investigation and to establish a baseline to show whether improvement occurred. After the completion of the pretest, participants performed twenty, 20-second acquisition trials. On the second day, a 48-hour retention test was administered and consisted of participants performing five, 20-second trials without concurrent or terminal visual feedback.

Data Analysis

Percent of time in Zone A, representing the percentage of time with 0–5° of foot platform deflection, was used as a dependent variable since that measure best reflects the instructions presented to the participants. A high score for percent of time in Zone A indicates better accuracy. Overall stability index was also used as a dependent variable, since it represents the variance of foot platform displacement in degrees by taking into account all movements in the anterior or posterior and medial or lateral axes (Biodex Medical Systems, 2003). A low score for overall stability index indicates better stability. Trials in the pretest phase, acquisition phase, and retention phase were calculated into trial blocks of five trials. The trials in the pretest phase were analyzed in an analysis of variance (ANOVA). The trials in the acquisition phase were analyzed in a 3 × 4 (KR Group × Trial Block) factorial ANOVA with repeated measures on the second factor. The trials in the retention phase were analyzed in an ANOVA.¹ The alpha level of statistical significance was set at p < .05 for all analyses. When the assumption of sphericity was violated, statistical significance was determined using the Greenhouse–Geisser degrees-of-freedom adjustment (Greenhouse & Geisser, 1959). The loci of significant effects were identified using the Tukey HSD test, and partial eta squared (η²_p) was calculated to determine the meaningfulness of significant effects.

Results

Pretest Phase

At pretesting, for percent of time in Zone A, there was no significant main effect for KR Group, F(2, 18) = .82, p = .46. The means for percent of time in Zone A are displayed in Table 1.

Table 1.

KR Group Means and Standard Deviations for Percent of Time in Zone A for Pretest, Acquisition, and Retention.

KR group	Pretest		Acquisition				Retention
KR group	Pretest		1	2	3	4	Actual	Adjusted
100% C&T	Mean	36.23	58.23	62.14	66.80	63.66	31.63	35.93
	SD	25.72	28.09	20.84	18.67	28.99	17.36	–
50% C & 100% T	Mean	54.89	61.31	72.03	72.43	72.60	62.26	59.71
	SD	29.15	27.29	31.80	28.73	27.83	11.98	–
0% C & 100% T	Mean	52.71	59.86	55.57	57.77	56.06	64.37	62.62
	SD	34.11	38.53	40.17	38.30	37.84	24.51	–

C = concurrent; T = terminal.

At pretesting, for the overall stability index, there was no significant main effect for KR Group, F(2, 18) = .70, p = .51. The means for overall stability index are displayed in Table 2.

Table 2.

KR Group Means and Standard Deviations for Overall Stability Index for Pretest, Acquisition, and Retention.

KR group	Pretest		Acquisition				Retention
KR group	Pretest		1	2	3	4	Actual	Adjusted
100% C&T	Mean	7.21	5.64	5.07	4.99	5.11	6.87	6.70
	SD	2.69	1.68	.99	.91	1.64	1.28	–
50% C & 100% T	Mean	5.79	5.27	4.72	4.48	4.39	4.94	5.07
	SD	1.59	1.66	2.22	1.79	1.67	.48	–
0% C & 100% T	Mean	6.22	4.95	5.39	5.53	5.85	5.08	5.12
	SD	2.48	1.92	1.88	2.37	3.28	1.16	–

C = concurrent; T = terminal.

Acquisition Phase

For the percent of time in Zone A during acquisition phase, there were no significant effects for KR Group, F(2, 18) = .29, p = .75, or Trial Block, F(3, 54) = 1.66, p = .19, and there was no signicant KR Group × Trial Block interaction effect, F(6, 54) = 1.28, p = .28.

For the overall stability index during acquisition phase, there were no significant effects for KR Group, F(2, 18) = .30, p = .75, or Trial Block, F(3, 54) = .45, p = .72, and there was no significant KR Group × Trial Block interaction effect, F(6, 54) = 1.67, p = .15.

Retention Test Phase

At retention posttesting, for percent of time in Zone A, there was a significant main effect for KR Group, F(2, 18) = 6.75, p < .01, η²_p = .43, and Tukey post hoc examination revealed that the 0% C & 100% T and 50% C & 100% T groups performed with significantly better accuracy than the 100% C&T group (Table 1).

At retention posttesting, for the overall stability index, there was a significant main effect for KR Group, F(2, 18) = 7.61, p < .01, η²_p = .46, and Tukey post hoc examination revealed the 0% C & 100% T and 50% C & 100% T groups performed with significantly better stability than the 100% C&T group (Table 2).

Discussion

This study purported to manipulate concurrent visual feedback among older adults in their learning of a continuous balance task and to extend Park et al.’s (2000; Experiment 1) young adult findings to older adults. Our research has both practical and theoretical implications for older adults’ fall risks from balance problems (see discussion in the Introduction section to this article) and for testing theoretical viewpoints regarding KR timing. Based on Park et al.’s (2000; Experiment 1) investigation, we expected that the 100% C & 100% T group would perform with significantly better accuracy and stability compared with the 50% C & 100% T and 0% C & 100% T groups. Park et al. found that their experimental groups (100% C & 100% T and 100% C & 0% T) performed with significantly smaller error than their 50% C & 100% T and 0% C & 100% T groups. Our results did not fully support our hypothesis in that our 100% C & 100% T group did not perform significantly better than our other two groups in all respects. Previous research showed that providing concurrent visual feedback in the acquisition phase had a positive, guiding function facilitating practice performance when feedback was available (Behrman et al., 1992; Ranganathan & Newell, 2009; Schmidt & Wulf, 1997; Vander Linden et al., 1993; Verschueren et al., 1997; Winstein et al., 1996), and our results also showed this temporary benefit for the 100% C & 100% T group receiving concurrent visual feedback and terminal feedback during the acquisition phase. Specifically, the percentage of time in Zone A for the 100% C & 100% T participants increased from their pretest to the end of their acquisition phase learning (Table 1). Also, the overall stability index for the 100% C & 100% T group decreased from the pretest to the end of the acquisition phase (Table 2). The findings in this study support the idea that when concurrent visual feedback was available it was a beneficial guide to initial performance.

On retention testing, we expected that the 100% C&T group would perform with significantly worse accuracy and stability than the 50% C & 100% T and 0% C & 100% T groups. Park et al. (2000; Experiment 1) found that the 100% C & 100% T and 100% C & 0% T groups performed with significantly greater error on the no-KR retention test than the 50% C & 100% T and 0% C & 100% T groups. The results of no-KR retention testing in this study supported our research hypothesis in that the 100% C&T group performed with significantly worse accuracy (percentage of time in Zone A) and greater instability (overall stability index) than our other two groups (50% C & 100% T and 0% C & 100% T). This is also consistent with other previous research investigations (Behrman et al., 1992; Ranganathan & Newell, 2009; Schmidt & Wulf, 1997; Vander Linden et al., 1993; Verschueren et al., 1997; Winstein et al., 1996) showing that concurrent visual feedback provides practice guidance when it is available but degrades retention learning when it is removed.

The results of the no-KR retention test in this study support the guidance hypothesis and the negative learning effect associated with concurrent feedback, extending this finding to theory as applicable to older as well as younger adults. Annett (1959, 1969, 1970) stated that augmented feedback should be considered in terms of the informativeness of the task-intrinsic (sensory) feedback and augmented feedback. According to Magill and Anderson (2017), concurrent feedback will most likely lead to a dependency when the benefit (i.e., information value) of task-intrinsic feedback is low, but the benefit of the augmented feedback is high. This negative learning effect has been the most common finding when manipulating concurrent visual feedback (Behrman et al., 1992; Park et al., 2000; Ranganathan & Newell, 2009; Schmidt & Wulf, 1997; Vander Linden et al., 1993; Verschueren et al., 1997; Winstein et al., 1996). Schmidt (1991) proposed three negative learning effects of receiving high relative frequency of KR, or in this case concurrent feedback, in the acquisition phase. First, receiving high frequency of KR (e.g., 100% concurrent visual feedback) may result in performers becoming dependent on the feedback. Second, receiving 100% visual concurrent feedback may block important processing activities which are important for the development of error-detection capabilities. That is, the negative learning effect occurs because concurrent feedback provided during a motor skill results in a strong guiding effect, while blocking important types of task-intrinsic feedback processing (Wulf & Shea, 2004). Similarly, the sensory substitution hypothesis has also been suggested to explain the negative learning effect and involves concurrent feedback directing performers’ attention away from the task-intrinsic feedback, resulting in concurrent feedback serving as a substitute (Schmidt & Wulf, 1997). Third, receiving 100% concurrent visual feedback may result in maladaptive short-term corrections leading to the development of an unstable or inconsistent movement response.

In summary, the results of this study support the guidance hypothesis regarding the timing of KR during motor learning. Providing 100% concurrent visual feedback in the acquisition phase will lead older adults as well as younger adults to become dependent on the feedback and perform poorly on a no-KR retention test. Thus, concurrent feedback (especially 100% concurrent visual feedback during acquisition), most often used by physical therapists for training sensorimotor skills (e.g., balance retraining), should be avoided in rehabilitation training and in interventions designed to prevent falls in older adults. The negative learning effect most often associated with concurrent visual feedback in young adults can be applied to older adults as well. More research involving older adults is needed with regard to the effects of other aspects of concurrent feedback (e.g., type of feedback such as visual or auditory, varying frequencies such as of 25% or 75%, scheduling such as fading, constant, or reverse-fade, and types of motor skills such as simple or complex).

Footnotes

Article Notes

References

Adams

J. A.

(1971) A closed-loop theory of motor learning. Journal of Motor Behavior 3: 111–150.

Adams

J. A.

(1987) Historical review and appraisal of research on the learning, retention, and transfer of human motor skills. Psychological Bulletin 101: 41–74.

Annett

(1959) Learning a pressure under conditions of immediate and delayed knowledge of results. Quarterly Journal of Experimental Psychology 11: 3–15.

Annett

(1969) Feedback and human behavior, Baltimore, MD: Penguin.

Annett

(1970) The role of action feedback in the acquisition of simple motor responses. Journal of Motor Behavior 11: 217–221.

Behrman

A. L.

Vander Linden

D. W.

Cauraugh

J. H.

(1992) Relative frequency knowledge of results: Older adults learning a force-time modulation task. Journal of Human Movement Studies 23: 233–250.

Bilodeau

I. M.

(1966) Information feedback. In: Bilodeau

E. A.

(ed.) The acquisition of skill, New York, NY: Academic Press, pp. 255–296.

Biodex Medical Systems (2003) Balance system operations and service manual, Shirley, NY: Author.

Carr

Zachariah

Weir

McNevin

(2011) An examination of feedback use in rehabilitation settings. Critical Reviews in Physical and Rehabilitation Medicine 23: 147–160.

10.

Centers for Disease Control and Prevention. (2012). Falls among older adults: An overview. Retrievde from http://www.cdc.gov/homeandrecreationalsafety/falls/adultfalls.html.

11.

Edwards

W. H.

(2011) Motor learning and control: From theory to practice, Belmont, CA: Wadsworth.

12.

Florence

C. S.

Bergen

Atherly

Burns

E. R.

Stevens

J. A.

Drake

(2018) Medical costs of fatal and nonfatal falls in older adults. Journal of the American Geriatrics Society 66: 693–698.

13.

Greenhouse

S. W.

Geisser

(1959) On methods in the analysis of profile data. Psychometrika 21: 451–467.

14.

Hinman

(2009) Factors affecting reliability of the biodex balance system: A summary of four studies. Journal of Sport Rehabilitation 9: 240–252.

15.

Lai

Shea

C. H.

(1999) The role of reduced frequency of knowledge of results during constant practice. Research Quarterly for Exercise and Sport 70: 33–40.

16.

Magill

R. A.

(1994) The influence of augmented feedback on skill learning depends on characteristics of the skill and the learner. Quest 46: 314–327.

17.

Magill

R. A.

(2001) Augmented feedback in motor skill acquisition. In: Singer

R. N.

Hausenblas

H. A.

Janelle

C. J.

(eds) Handbook of sport psychology, New York, NY: John Wiley & Sons, Inc, pp. 86–114.

18.

Magill

R. A.

Anderson

D. I.

(2017) Motor learning and control: Concepts and applications, 11th ed. New York, NY: McGraw Hill.

19.

Newell

K. M.

(1976) Knowledge of results and motor learning. In: Keogh

Hutton

R. S.

(eds) Exercise and sport sciences reviews Vol. 4Santa Barbara, CA: Journal Publishing Affiliates, pp. 195–228.

20.

Park

Shea

C. H.

Wright

D. L.

(2000) Reduced-frequency concurrent and terminal feedback: A test of the guidance hypothesis. Journal of Motor Behavior 32: 287–296.

21.

Piper

Rone

S. A.

(2001) Concurrent and summary feedback and the ability of an elderly population to learn a limited weight bearing skill. Journal of Geriatric Physical Therapy 24: 31–34, 40.

22.

Rabbitt

(1997) Ageing and human skill: A 40th anniversary. Ergonomics 40: 962–981.

23.

Ranganathan

Newell

K. M.

(2009) Influence of augmented feedback on coordination strategies. Journal of Motor Behavior 41: 317–330.

24.

Salmoni

A. W.

Schmidt

R. A.

Walter

C. B.

(1984) Knowledge of results and motor learning: A review and critical appraisal. Psychological Bulletin 95: 355–386.

25.

Schmitz

Arnold

(1998) Intertester and intratester reliability of a dynamic balance protocol using the Biodex System. Journal of Sport Rehabilitation 7: 95–101.

26.

Schmidt

R. A.

(1975) A schema theory of discrete motor skill learning. Psychological Review 82: 225–260.

27.

Schmidt

R. A.

(1991) Frequent augmented feedback can degrade learning: Evidence and interpretations. In: Stelmach

G. E.

Requin

(eds) Tutorials in motor neuroscience, Dordrecht, The Netherlands: Kluwer, pp. 59–76.

28.

Schmidt

R. A.

Wulf

(1997) Continuous concurrent feedback degrades skill learning: Implications for training and simulation. Human Factors 39: 509–525.

29.

Swanson

Lee

T. D.

(1992) The effects of aging and scheduling of knowledge of results on motor learning. Journal of Gerontology: Psychological Sciences 47: 406–411.

30.

Swinnen

S. P.

(1996) Information feedback for motor skill learning: A review. In: Zelaznik

H. N.

(ed.) Advances in motor learning and control, Champaign, IL: Human Kinetics, pp. 37–66.

31.

Vander Linden

D. W.

Cauraugh

J. H.

Greene

T. A.

(1993) The effect of frequency of kinetic feedback on learning an isometric force production task in nondisabled subjects. Physical Therapy 73: 79–87.

32.

Verschueren

S. M. P.

Swinnen

S. P.

Dom

DeWeerdt

(1997) Interlimb coordination in patients with Parkinson’s disease: Motor learning deficits and the importance of augmented information feedback. Experimental Brain Research 113: 497–508.

33.

Voelcker-Rehage

(2008) Motor-skill learning in older adults—A review of studies on age-related differences. European Review of Aging and Physical Activity 5: 5–16.

34.

Winstein

C. J.

Pohl

P. S.

Cardinale

Green

Scholtz

Waters

C. S.

(1996) Learning a partial weight bearing skill: Effectiveness of two forms of feedback. Physical Therapy 76: 985–993.

35.

Wishart

L. R.

Lee

T. D.

(1997) Effects of aging and reduced relative frequency of knowledge of results on learning of a motor skill. Perceptual and Motor Skills 84: 1107–1122.

36.

Wishart

L. R.

Lee

T. D.

Cunningham

S. J.

Murdoch

J. E.

(2002) Age-related differences and the role of augmented visual feedback in learning a bimanual coordination pattern. Acta Psychologica 110: 247–263.

37.

Wulf

Shea

C. H.

(2004) Understanding the role of augmented feedback: The good, the bad, and the ugly. In: Williams

A. M.

Hodges

N. J.

(eds) Skill acquisition in sport: Research, theory and practice, London, England: Routledge, pp. 121–144.