Strength and conditioning coaches as well as athletic team coaches constantly search for the most efficient ways to improve the efficiency and effectiveness of their training and practice programming. Coaches, regardless of the sport, age of participants, and level of talent, face the task of teaching a novel motor skill or refining an already learned skill. In the past, before the advent of video technology, coaches relied on providing verbal information and feedback as a means of instruction. During this time, research focused on the content of the information and feedback that was presented to the individual. The temporal location of the feedback presented was also of concern to researchers (7,13-16).
The research conducted on the content of the feedback provided to an individual has focused predominantly on knowledge of results and in more recent years has evaluated knowledge of performance. The distinction between the 2 types of augmented feedback is an important one and plays a large role in the acquisition and refinement of a motor skill. Knowledge of results is postmovement feedback that pertains to the outcome of a movement in terms of an environmental goal. Knowledge of performance is feedback directed toward the movement pattern of the learner rather than the outcome of the movement. Knowledge of results research has revealed that the precision of feedback plays a substantial role for reducing variability and that quantitative knowledge of results was more beneficial than qualitative knowledge of results (7). Increases in technology in recent years has afforded researchers a means of evaluating kinematic aspects of movement with much more ease, which has consequently increased the breadth of the research on knowledge of performance.
Knowledge of performance is often presented in the form of verbal feedback or video feedback regarding the kinematics of the movement. Kinematic information pertains to the limb position and velocity, movement time, and patterns of coordination. When a desired movement outcome depends on interactions among many segments, kinematic feedback has been demonstrated to be more effective than knowledge of results (10,12). Indications are that the real world movements observed in the athletic arena are multiple degree of freedom movements, which may benefit more from kinematic knowledge of performance than knowledge of results (8,11).
With the advent and improvement of video analysis software, strength coaches as well as athletic coaches have been increasingly using video as method of feedback administration. This new trend in feedback administration opens up new questions for motor behavior research. Early reviews of video feedback research suggest that it may be an ineffective means of presenting knowledge of performance to promote skill learning and refinement (1,9). Explanations for the lack of effectiveness of video feedback on motor skill acquisition and refinement in these early studies include the introduction of overly complex information, failing to provide information about critical aspects of the skill, and not providing enough detail for error detection. Video feedback administered in conjunction with verbal attentional cueing as well as information including error correction strategies have been shown to benefit skill acquisition and refinement (5,6).
Research conducted previously by Kernodle and Carlton (5) and Landin and MacDonald (6) are among the few studies conducted on knowledge of performance using multiple degrees of freedom movements, which require a high level of coordination and temporal sequencing. Studies conducted on these types of movements are beneficial to feedback research in that they represent movements that are present in everyday life as well as sporting events. Coaches, athletic trainers, physical therapists, and alike who are involved with increasing an individual's proficiency at a given skill can benefit from the findings of such research. Strength coaches in particular administer feedback to individuals about their performance on a given skill, typically the execution of a particular lifting technique. A lift that is commonly used by strength coaches is the hang power clean. The hang power clean is a lifting technique used to develop power in an athlete, and the movement has many degrees of freedom as well as temporal sequencing. The importance of the hang power clean to strength coaches and the characteristics of the exercise make this movement beneficial to study.
The hang power clean exercise has been found to produce high bar velocities, high ground reaction forces, and high power outputs (2). This exercise involves all of the major muscle groups of the body and is a technical lift that requires balance, coordination, timing, and power (3). For these reasons, the hang power clean has been used in athletic training for many years because of its ability to train athlete's muscles to produce power. Also, the mechanics of this movement closely mimic and are specific to the demands placed on athletes in a variety of sports (i.e., hockey, football, basketball) in which being able to produce power is beneficial to performance. The hang power clean is a complex behavior that provides a model to evaluate the effects of various forms of augmented knowledge of performance feedback on performance. Three forms of knowledge of performance feedback were assessed in the present experiment: video + cues, verbal-only, and video-only. Previous research conducted by Kernodle and Carlton (5) and Janelle et al. (4) found that feedback provided by video alone did not yield changes in performance but did show positive changes in performance when video feedback coupled with attention cueing and corrective information was administered. It was expected that (1) video + cues would enhance performance of the hang power clean to a greater extent than that of verbal or video feedback alone; (2) verbal-only feedback will enhance performance of the hang power clean to a greater extent than video feedback alone; (3) video-only feedback will show no improvements in performance. The purpose of the study was 2-fold: (1) to evaluate the effects of 3 types of augmented feedback on young adults' weight-lifting movement form and (2) to assess the effects of augmented feedback on measures of strength and power.
Experimental Approach to the Problem
A mixed-model research design was used in which the performance of women assigned to 3 conditions was assessed before training, during 6 training sessions, and after training. Participants were stratified by height (greater or less than 67 inches) and by experience with the hang power clean (greater or less than 24 mo) and then randomly assigned using a random number table to 1 of the 3 experimental groups: video-only, verbal-only, or video + cues. The protocol used to assess the effects of feedback on women's performance was based on methods developed by Janelle et al. (4). In addition, quantitative assessments of participants' movement patterns during the hang power clean were conducted.
The rationale for these hypotheses is the earlier findings of previous research conducted by Janelle et al. (4). Limited research has been conducted on the effects of altering the methods of providing feedback to individuals performing a dynamic multiple degree of freedom movement. This knowledge is important for athletes because it may decrease the likelihood of injury during the execution of dynamic forceful movements and for instructors who seek efficient methods of modifying athletic performance.
A volunteer sample of 26 female NCAA Division-I athletes ranging in age between 18 to 22 (mean = 20, SD = 1) years participated in the study. Women were recruited from the university NCAA Division-I soccer and volleyball teams. All women completed a questionnaire that provided information concerning past strength training experience. For women who completed the entire study, experience with the hang power clean ranged from 2 to 48 (mean = 17.8, SD = 16.2) months, academic year ranged from first to third (mean = 1.8, SD = 0.7) year, years of participation in sport ranged from 4 to 17 (mean = 11, SD = 4.6) years, and years of weightlifting experience ranged from 1 to 7 (mean = 3.0, SD = 1.9) years. All data collection occurred during both teams off-season training. A participant was excluded from participation if she had less than 2 months experience with the execution of the hang power clean, no prior completion of a 1 repetition maximal (1RM) test, or any medical contraindications to the performance of the hang power clean. One woman was excluded from participation on the basis of the exclusion criteria. During the course of the study, 8 women withdrew from the study: 3 because of sport related injury and 5 because of missed training sessions. All participants gave their informed consent consistent with the policies of the institutional review board for use of human subjects in research. Participant characteristics are presented in Table 1.
Participants attended 9 sessions, 2 per week (Monday and Friday), which was the normally scheduled weightlifting and conditioning time for each sport team. Participants also performed an additional strength training workout each Wednesday; the same format as Monday and Friday's workout was followed with the exception that the hang power clean task was not performed. All trials of the hang power clean were performed on a 2.44 m by 2.44 m platform. During the first and last sessions of the study, 2 platforms were used to conduct the 1RM test, and 2 platforms were used to conduct the power test. For the power test, platforms were outfitted with a TENDO Weightlifting Analyzer (model V-104, TENDO Sports Machines, Slovak Republic), which was used to calculate peak muscular power and average power output during the execution of the movement. The TENDO Weightlifting Analyzer was set-up according to the manufacture's instruction manual to ensure that the cable output, filter value, and weight was set properly on the microcomputer.
During each of the 9 trainings sessions, 1 platform was assigned to participants in each of the 3 treatment conditions. Platforms were separated by approximately 15 m, which ensured the relative isolation of participants in each feedback conditions. Each platform was equipped with a Panasonic digital video camcorder (model PV-GS300, USA) mounted on a tripod set 11 feet away on the right side of the participant and positioned at a height 4 feet 8 inches. Platforms designated to the video-only and video + cues groups were also equipped with a Dell laptop (model Precision M-65, USA) as a means of administering feedback. Feedback for the video-only and video + cues groups was administered using Dartfish Advanced Video Analysis Software (version 188.8.131.52).
Before beginning the prescribed exercises, participants assigned to the video-only group and those assigned to the video + cues group were asked to watch a video presentation of an expert lifter who modeled the hang power clean. Participants in the verbal-only group were asked to listen to a narration of the movement provided by an audio transcription by way of the Dartfish software. After the presentation of information, participants performed their normal warm-up activities particular to their team. Upon completion of the warm-up, participants were asked to perform a 1RM test of the hang power clean. After a 2 to 3 minute rest period, participants were then asked to perform the muscular power test, which involved one set of 4 repetitions of the hang power clean using 70% of each individual's 1RM.
After warm-up activities, participants completed 4 sets of the hang power clean at 75% of their 1RM (1 set = 4 successive repetitions of the lifting movement). Participants were assigned to 1 of 4 platforms, each equipped with a Panasonic digital video camcorder (model PV-GS300). The video camera was turned on before the beginning of the experimental protocol and turned off once each participant's performance on each set had been captured. Participants were instructed to turn and face the camera before each set for identification purposes, prepare to execute the lift, and then execute the set. After completion of each set, the participants placed the weighted barbell back in the weightlifting rack, helped adjust the weight on the barbell for the next participant, and then rested until their next turn (approximately 2-3 min). Digital images were recorded on a cassette, stored, and secured for later computer analyses.
Sessions 3 to 8
After warm-up exercises, participants completed 4 sets (4 repetitions each) at 75% of their 1RM of the hang power clean. Participants were assigned to 1 of 3 platforms, appropriate to their experimental group, each of which was equipped with a video camera. Lift performance was video taped throughout each experimental protocol, and images were stored on a cassette, stored, and secured for later computer analyses. After completion of each set, the participant placed the weighted barbell back in the weightlifting rack, immediately proceeded to an assigned feedback station, received the appropriate feedback, and then rested until their next turn (approximately 2-3 min).
Video-only Feedback Protocol
After each set, each participant assigned to the video-only condition was immediately shown video images of her lifting performance. The participant was allowed to view the video of her performance throughout the duration of 1 set (4 repetitions, approximately 45 s).
Video + Cues Feedback Protocol
After each set, each participant assigned to the video + cues condition was immediately shown video images of her lift performance from the prior session with visual task relevant cues added to the video and was also given verbal feedback from the researcher concerning performance during the previous set. The task-relevant cues not only drew attention to the movement aspects that were in most need of correction but also included information about how to correct the errors present in the movement (transitional information). Feedback was provided for approximately 45 seconds.
Verbal-only Feedback Protocol
After each set, each participant assigned to the verbal-only condition received 45 seconds of verbal feedback typical of that provided by strength training specialists. The feedback administered at the verbal-only or the video + cues stations was directed toward aspects of the movement most pertinent for subject's training. Once the subject demonstrated competence in a particular movement aspect, feedback regarding that aspect was withdrawn and the researcher evaluated another movement aspect.
The feedback administered to the verbal-only group and the verbal attention cues for the video + cues group can be found in Table 2. The visual cues involved the use of arrows and pointers to direct the participant's attention as well as statements written on the screen. The cues corresponded to 9 movement aspects that were task analyzed (See Table 3). The cues were chosen on the basis of the recommendation of certified strength and conditioning coaches (NSCA CSCS, USWF Club Coach). Recommendations were based on the coaches' education and certifications as well as years (approximately 3-16) of experience teaching weightlifting techniques.
During session 9, participants completed warm-up exercises and then performed the 1RM test of the hang power clean and the muscular power test. Both tests were performed in the same manner as during the first session of the experiment.
Individuals' performances were collected from session 2 through 8 and were evaluated using the Dartfish software. The 9 movement indices of the hang power clean are described in Table 3. The methods used to measure each of the 9 indices are described in Table 3. An average score for each of the 9 indices was calculated from averaging participants' performance measured during the second and third repetitions of each of the 4 sets.
Outcome measures also included the 1RM and the muscular power test. Scores on the 1RM test indicated the maximum amount of weight that the individual could successfully lift for 1 repetition and was measured in pounds. Scores for the 1RM test were manually recorded and then were entered into a computer database. For the muscular power test, each individual's peak power output (Watts) was recorded for each repetition. Power output scores were recorded manually after each of the 4 repetitions and averaged.
Each of the 9 form measure scores were analyzed separately by way of a mixed-model 2-way analysis of variance (ANOVA) in which the 3 training conditions constituted the between-groups factor and participants' performance measures during each of the training session constituted the repeated-measures factor. Violations of sphericity were corrected by adjusting degrees of freedom according to the Huyhn-Feldt test. An alpha of 0.05 was used for all ANOVAs. Estimates of effect size (ωp2) were reported for significant main effects and interactions.
The form measures were further analyzed to test the reliability of the scoring procedure. A rater, who was unaware of participants' assignments to training conditions, was taught the methods of data collection and then conducted the analysis process for 3 subjects in each of the 3 feedback conditions. The scores obtained from the blind rater and the scores obtained by the primary researcher were correlated. The stability of scores measured during successive training sessions was determined by assessing the performance of participants assigned to the video-only treatment group. An intraclass correlation (ICC) using a 2-way random consistency model was computed on participants' scores across the 6 training sessions. Intraclass correlations were computed separately for each of the 9 movements. The 1RM and muscular-power scores were analyzed separately by way of a mixed-model 2-way ANOVA in which the 3 training conditions constituted the between-groups factor and participants' performance measures on the first and last session constituted the repeated-measures factor.
The analyses of each of the 9 movement indices are presented in Table 4. Statistically significant interactions between the Group and Time factors were found for movement indices that assessed torso inclination relative to the ground, F(8.2,57.3) = 4.17, p < 0.05, ω2p = 0.60, distance ears are in front of the bar, F(7.2,50.1) = 2.84, p < 0.05, ω2p = 0.43, position of the bar relative to the toe, F(9.5,66.8) = 1.85, p = 0.071, ω2p = 0.32, angle of the body, F(10,70) = 4.27, p < 0.05, ω2p = 0.53, and bar relative to toe in full extension, F(9.2,64.1) = 2.57, p < 0.05, ω2p = 0.46. Movement indices that did not yield significant Group × Time interactions included time heels are out of contact with the floor, F(10,70) = 0.40, time of elbow rotation under the bar, F(10,70) = 1.11, knee angle, F(10,70) = 0.66, and center of hip relative to toe, F(10,70) = 0.28. The analyses of each of the muscular outcome measures are presented in Table 5. There were no differences present among the 3 feedback groups for both 1RM and muscular power tests. There was, however, a main effect observed over time for each of the outcome measures, F(1,14) = 22.28, p < 0.05 and F(1,14) = 14.69, p < 0.05, respectively, indicating that the performance of participants in all 3 groups improved similarly as a function of training. Analyses of the stability of the participants' performance yielded ICCs that ranged between 0.95 and 0.99, indicating that scores obtained during successive training sessions were stable.
Reliability estimates of independent rater's scoring for each of the 9 movement indices ranged between 0.94 and 1.0, indicating that the data collection protocol from the Dartfish software was reliable and independent of the rater. The discriminate validity of Dartfish is not well established, and limited research has been conducted to assess the reliability and validity of the video-analysis software. The limited amount of research that has been conducted on Dartfish reliability and validity did, however, provide the basis for camera positioning, resolution, and distance of participants from the camera used in the present study. Overall findings and the effects of the feedback conditions on the movement form are presented in Table 6.
The impacts of 3 different types of augmented knowledge of performance feedback were compared: video + cues, verbal-only, and video-only. The level of measurement used in the study provided information concerning 3 phases of the hang power clean. Analyses of movement components that underlie the start phase of the hang power clean provide partial support of the hypotheses that video feedback provided with attentional cueing would lead to increased performance when compared with the verbal-only and video-only feedback conditions. Of particular interest was the finding that the video + cues condition did not yield greater performance changes over the verbal-only condition as predicted. Also of interest was the finding that the video-only condition failed to improve performance.
Movement aspects present in the start phase of the hang power clean include (1) torso inclination relative to the ground, (2) distance ears are in front of the bar, and (3) position of bar relative to toe. Participants in the video + cues feedback group did not perform movements better than participants in the verbal-only group. It was expected that the addition of visual cueing and transitional information would lead to an increase in performance over the other 2 conditions as evidenced in previous research conducted by Kernodle and Carlton (5). The participants in the video + cues feedback group and participants in the verbal-only feedback group did, however, perform better than those in the video-only group. Thus, it appears that both verbal knowledge of performance and video + cues knowledge of performance are sufficient to improve select movements that constitute the hang power clean. Importantly, participants in the video-only group evidenced no gains in hang power clean performance.
Movement aspects present in the pull phase of the hang power clean include (1) angle of body relative to ground in full extension, (2) position of bar relative to toe in full extension, and (3) time heels are out of contact with the floor. Participants' performance in both the video + cues group and the video-only group failed to show improvement. It was expected that the video + cues group would yield greater performance changes over the verbal-only and video-only conditions. However, participants in the verbal-only group performed 2 movements better than subjects who received video + cues or video-only feedback. The lack of improvement in the video + cues condition over the verbal-only and the increased performance in the verbal-only condition may indicate that these particular movement aspects of the hang power clean are better modulated by way of verbal feedback.
Movement aspects present in the catch phase of the hang power clean include (1) time of elbow rotation under the bar, (2) knee angle, and (3) center of hip joint relative to toe. Analyses performed on aspects of the catch phase revealed no changes as a function of training. It is plausible the lack of improvement on this aspect of the hang power clean is caused by the rapid and ballistic characteristics of the movement components. The speed in which catch phase movements are performed may limit intrinsic feedback and, thus, may not allow for a precise cognitive representation to be formulated.
Prior research has shown that kinematic video feedback with attention cueing (e.g., transitional information) is more beneficial for performance than verbal or video feedback alone (5). This was not the case in the present study because the participants in the video + cues group did not increase performance more than the verbal-only group. It may be the case that the verbal cues used in both feedback conditions were sufficient to direct participants' focus to critical aspects of the movement and to provide the participant with information concerning the appropriate corrections. Kernodle and Carlton (5) observed that kinematic feedback given verbally with cues that contained transitional information yielded greater improvements in performance over video replay coupled with cues that did not contain transitional information and simply focused the performer's attention to parts of the movement that contained errors. Both feedback conditions (i.e., verbal cues with transitional information and video replay with attentional focusing cues) enhanced the learning of complex movements more than verbal knowledge of results-only feedback or video knowledge of performance-only feedback. In the present study, both verbal and visual feedback conditions contained transitional information, and the results of the present study suggest that the form of the feedback, (i.e., video vs. verbal) may be of less importance on performance than transitional information presented.
The results of the present study corroborate those obtained in several other studies showing that simply providing a learner with video feedback without additional cues has little effect on skill acquisition (1,9). Improvements in skilled performance necessitates that some form of cued feedback be provided. However, the effects of cued feedback on dynamic multiple segment actions appear to be complex. Knowledge of performance may benefit some, but not all, movement components. It is unlikely that any one form of knowledge of performance, in and of itself, will lead to improvement in all aspects of movements that underlie the complex behaviors seen in sporting events. Nevertheless, more research is needed that examines the acquisition of complex multiple degree of freedom movements and how types of feedback influence performance and learning. Additional information will provide the basis for developing a component-specific approach to athletic skill training. It will be important for both researchers and strength trainers to analyze the components of sport-related movements and to identify components that may be differentially influenced by different types of feedback. In the present study, video replay plus verbal feedback was not found to differ from verbal-only feedback conditions. However, the participants in the study were young athletes with prior experience in weightlifting. It remains to be determined whether video feedback may aid novices who may be able to use video information to develop cognitive representations involved in planning movements.
There are limitations to this research. Several participants withdrew from the study, which resulted in small sample sizes that provided less statistical power than desired. However, the measures obtained from the multiple components yielded reliable scores with low variability, suggesting that the final sample size was sufficiently large to interpret the results obtained. An additional limitation was the lack of stringent experimental control typical of laboratory-based studies. The present study was conducted under training conditions typical for competitive athletes, and it was designed for practical application in the strength and conditioning arena. In addition, the study was limited to athletes with prior weight training experience. Although all participants had previously participated in strength training using the hang power clean exercise, the amount of previous experience with the hang power clean varied among athletes and could have possibly affected outcomes. The young athletes in the present study may have been in different stages of learning the hang power clean, and these differences may have affected how the different types of feedback influenced performance improvements.
Despite these limitations, the results obtained reinforce the view that strength and conditioning professionals should be aware of the role of knowledge of performance feedback on skill development. Regardless of whether knowledge of performance feedback is administered verbally, visually, or in combination, it is important that the feedback does not simply direct an individual's attention to the aspect of the movement that is in error but also that the feedback provides information concerning how movement error can be reduced. This conclusion is of particular importance as it relates to female athletes. There is a paucity of feedback research in general for female athletes and more importantly a lack of research on the effects of altering feedback for dynamic movements of female athletes.
The results of the present study corroborate those obtained in several other studies showing that simply providing a learner with video feedback without additional cues has little effect on skill acquisition. Improvements in skilled performance necessitate some form of cued feedback. However, the effects of cued feedback on dynamic multiple segment actions appear to be complex. Knowledge of performance may benefit some, but not all, movement components. It is unlikely that any one form of knowledge of performance, in and of itself, will lead to improvement in all aspects of movements that underlie the complex behaviors seen in sporting events. Nevertheless, more research is needed that examines the acquisition of complex multiple degree of freedom movements and how types of feedback influence performance and learning. Additional information will provide the basis for developing a component specific approach to athletic skill training. It will be important for both researchers and strength trainers to analyze the components of sport-related movements and to identify components that may be differentially influenced by different types of feedback. Strength and conditioning professionals should be aware that regardless of whether knowledge of performance feedback is administered verbally, visually, or in combination, it is important that the feedback does not simply direct an individual's attention to the aspect of the movement that is in error but also that the feedback provides information concerning how movement error can be reduced. Results of the present study also address an underserved population of female athletes, and the findings have practical application for athletic coaches and strength coaches alike. Female athletes may respond differently to different types and schedules of feedback as opposed to their male counterparts, in which much of the research has focused on. This research is aimed at helping athletic coaches and strength professionals improve the way in which feedback is administered to maximize the potential of their female athletes.
1. Arnold, RK and Rothstein, AL. Bridging the gap: application of research on videotape feedback and bowling. Motor Skills
1: 35-62, 1976.
2. Garhammer, J. Power production by Olympic weightlifters. Med Sci Sports Exerc
12: 54-60, 1980.
3. Garhammer, J. Power: kinesiological evaluation. NSCA J
6: 60-63, 1984.
4. Janelle, CM, Barba, DA, Frehlich, SG, Tennant, LK, and Cauraugh, JH. Maximizing performance feedback effectiveness through videotape replay and a self-controlled learning environment. Res Q Exerc Sport
4: 269-279, 1997.
5. Kernodle, MW and Carlton, LG. Information feedback and the learning of multiple-degree-of-freedom activities. J Motor Behav
24: 187-196, 1992.
6. Landin, DK and Macdonald, G. Improving the overheads of collegiate tennis players. J Appl Res Coach Athletics
5, 85-90. 1990.
7. Magill, RA and Wood CA. Knowledge of results precision as a learning variable in motor skill acquisition. Res Q Exerc Sport
57: 170-173, 1986.
8. Mononen, K, Vitasalo, J, Konttinen, N, and Era, P. The effects of augmented kinematic feedback on motor skill learning in rifle shooting. J Sport Sci
21: 867-876, 2003.
9. Newell, KM. Skill learning. In: Human Skills
. Holding, DH, ed. New York: Wiley, 1981.
10. Newell, KM and Carlton, MJ. Augmented information and the acquisition of isometric tasks. J Motor Behav
19: 4-12, 1987.
11. Newell, KM and Walter, CB. Kinematic and kinetic parameters as information feedback in motor skill acquisition. J Hum Move Stud
7: 235-254, 1981.
12. Schmidt, RA and Lee, TD. Motor Control and Learning: A Behavioral Emphasis
. Champaign, IL: Human Kinetics, 2005.
13. Sparrow, WA and Summers, JJ. Performance on trials without knowledge of results (KR) in reduced relative frequency presentations of KR. J Motor Behav
24: 197-209, 1992.
14. Weeks, DL and Kordus, RN. Relative frequency of knowledge of performance and motor skill learning. Res Q Exerc Sport
69: 224-230, 1988.
15. Weeks, DJ, Zelaznik, H, and Beyak, B. An empirical note on reduced frequency of knowledge of results. J Hum Move Stud
25: 193-201, 1993.
16. Young, DE and Schmidt, RA. Augmented kinematic feedback for motor learning. J Motor Behav
24: 261-273, 1992.
Keywords:© 2010 National Strength and Conditioning Association
weight lifting; dynamic; strength training; coaching; cues; transitional