Research interest in the application of behavioral interventions to improve sports performance is increasing (Martin, 2003). The utility of behavioral strategies such as modeling, goal setting and public posting, prompting, shaping, contingent feedback, self-monitoring, and reinforcement, has been demonstrated with a multitude of sports including ballet, basketball, football, gymnastics, tennis, track and triathalon, and volleyball (Martin, Thompson, & Regehr, 2004).

One promising strategy for improving sporting skills that has heretofore received limited research evaluation is teaching with acoustical guidance (TAGteach). TAGteach is a behavioral strategy, derived from clicker training widely used with animals, that involves the delivery of an acoustic stimulus, a previously conditioned reinforcer, contingent upon the correct performance of a target behavior (TAGteach International, 2012). TAGteach utilises behavioral chaining procedures and positive reinforcement in order to change behavior (TAGteach International, 2012).

TAGteach procedures have been widely used in a variety of mainstream settings to train many different skills but have received comparatively little attention in the research literature (Fogel, Weil, & Burris, 2010). Research studies to date, have evaluated the utility of TAGteach procedures for establishing, or improving the performance of, various sport skills. For example, Stokes, Luiselli, Reed, and Fleming (2010) implemented TAGteach procedures to improve the offensive pass-blocking of four high school football players. The impact of descriptive feedback, or descriptive feedback and video feedback, was evaluated prior to the implementation of TAGteach. The TAGteach procedure was associated with high levels of accuracy for each participant. However, maintenance of behavior change following the initial training phase was noted to be poor. Fogel and colleagues (2010) examined the effectiveness of TAGteach procedures for teaching a golf swing to a novice using a multiple baseline design across skills. Results indicated that the learner acquired four of the five targeted golf swing skills and that these skills generalized to a novel golf club. Recently, Harrison and Pyles (2013) evaluated the utility of TAGteach to improve the tackling skills of three high school football players. The intervention, evaluated within a multiple baseline across participants design, led to notable improvements in the tackling skills of each participant, even when participants' speed of movement was increased. In addition, two unpublished dissertations have also provided further demonstration of the utility of TAGteach for improving yoga performance (Andrews, 2014) and dance performance (Quinn, 2013).

The need for further research on the efficacy of TAGteach for improving sporting performance has been emphasized (e.g., Luiselli, Woods, & Reeds, 2011; Fogel et al., 2010). The current research study sought to contribute to the existing body of research by examining the effect of TAGteach on the passing skills of college rugby athletes.

Method

Participants and Setting

Three male university rugby athletes participated in this research study. Participants' ages ranged from 19-24 years (M=20.66, SD=2.88). Participants were identified and recruited by the university rugby team's head coach as they were underperforming during training sessions and games, defined as falling below an 85% mastery criterion for correct performance of the targeted passing skills. Each of the participants had at least 12 months of university-level rugby experience.

The intervention was conducted in an indoor training room at the participants' university. The training room was approximately 405sqm with a hardwood surface. Sessions were conducted at least once per week, over a four month period, during the team's regularly scheduled training sessions. The intervention was delivered by a trainer (first author) who was a second year Masters student in Applied Behavior Analysis and who had completed the required training in TAGteach.

Materials

This study required a number of materials. A tagger, a hand held device that makes a clicking sound when pressed, was used to provide acoustic feedback to participants. A task analysis for each of the four targeted passing skills was developed by the researcher in collaboration with the rugby team's head coach. This was achieved by reviewing videos of a high performing, veteran, rugby athlete. A rugby ball was required for training sessions. Finally, a camera with a video function was also used for data collection.

Response Measurement and Interobserver Agreement

The dependent variable was the percentage of correctly performed "tagged" steps of each of the four passing task analyses (available from the authors upon request). The passing skills targeted were: (1) a ruck pass (all participants; included 20 steps); (2) a trap pass (Participant 1 only; included 24 steps); (3) a hold pass (Participants 1 and 2; included 7 steps), and (4) a direct pass (Participants 2 and 3; included 32 steps). Each of these passing skills comprises an important prerequisite skill for complex rugby movements. Further, the skills were topographically different and therefore intervention for one skill was unlikely to produce concurrent improvements in the other. The target passing skills for each participant were selected on the basis of their current level of performance by the rugby team's head coach.

Performance was assessed by the researcher using video footage recorded at each training session. An independent observer also reviewed and scored 49% of training sessions for Participant 1, 51% of training sessions for Participant 2 and 57% of training sessions for Participant 3 in order to verify the accuracy of all data recorded. An agreement was defined as both observers scoring a target behavior as either occurring or non-occurring. For each task analysis, percentage of interobserver agreement was calculated by dividing the number of agreements by the total number of agreements and disagreements multiplied by 100. Mean interobserver agreement was found to be 94% with a range of 86%--100% for Participant 1,71%-100% for Participant 2 and 85%-97% for Participant 3.

Design

A multiple-baseline design across behaviors was used to evaluate the effects of TAGteach on the rugby passing skills of all three participants. This is a single-subject research design which employs a staggered introduction of an independent variable on one or more target behaviors (Kazdin, 2011). Researchers utilize the multiple-baseline design to evaluate interventions by examining the changes following each baseline. If changes are shown across the staggered baseline and intervention phases, the effects can be attributed to the intervention rather than to extraneous variables. The multiple baseline design is employed when it is not possible for participants to return to original baseline (Kazdin, 2011).

Procedure

Baseline. During the baseline phase, participants were required to perform each of their targeted passing skills a minimum of four times during their regularly scheduled training session. The head coach instructed the participant to set up for a ruck, direct, trap or hold pass as appropriate. The coach delivered verbal feedback, typical of a regular training session. Baseline data were used to identify appropriate tag points for each of the passing task analyses; steps within each task analysis that were performed correctly in 60% of opportunities or below during baseline were assigned a tag point.

Teaching with Acoustical Guidance. The researcher (first author) was trained in TAGteach using the TAGteach International Online Certification course (TAGteach International, 2012). TAGteach involves the use of an acoustic marker, unique phrasing and positive reinforcement to enhance teaching and learning (TAGteach International, 2012). TAGteach terminology (Fogel et al., 2010) includes: (1) Tagger--a device that produces a distinct audible stimulus used to mark the correct execution of a target behavior, or task analysis step; (2) Tag point--a single targeted step or behavior for which correct performance is marked with an acoustic stimulus; (3) Tag--the acoustic stimulus used to mark the correct execution of a specific behavior; and (4) Debrief--a post-training verbal discussion between the trainer and learner in which the performance is evaluated and progression to the next tag point is considered.

Intervention began by familiarizing participants with the principles of TAGteach and with the tag that would be used to indicate correct performance. Participants' questions or uncertainties were also addressed. To habituate participants to the tag, and to reduce the chance of its negatively impacting upon performance, the acoustic stimulus was sounded for the correct demonstration of mastered movements (e.g., a catch using both hands) during a brief period of free play. The trainer paired the tag with verbal praise such as "well done!" or "nice job!" to condition the tag as a reinforcer and to ensure that participants associated the tag with the successful execution of a behavior. Participants were also taught that the absence of the tag (acoustic stimulus) meant they should self-assess the performance of the skill.

The trainer began each TAGteach session by informing the participant of the targeted steps (tag points) that would be tagged if performed correctly. Participants subsequently received a tag if the tag point was performed correctly. If it was not performed correctly, there were no consequences and participants completed the remaining steps of the task analysis. Participants were required to perform the skill at least six times during each TAGteach training session. Criteria for the conclusion of a TAGteach session were the completion of at least six trials of the target passing skills (e.g., ruck pass), and the completion of the tag point correctly in the trial immediately prior to session termination. Participants were debriefed at the beginning of the subsequent training session, and a mutual decision was made to either continue working on the same tag point or to introduce the next tag point if they had demonstrated mastery (i.e., over 95% of attempts performed correctly in three out of five previous consecutive trials). The identified tag points were targeted in succession until mastery was achieved on all. At the end of each session, all the trials performed during the session were converted to a percentage to determine the levels of performance for a target behavior. The order of the passing skills targeted was randomized for each participant. Progression to the next targeted skill occurred when the participant demonstrated mastery of all tag points in the current skill.

The experimenter also reviewed videos of Baseline and TAGteach conditions to determine the timed speed for each pass performed. This was analyzed by starting a stopwatch when the participants touched the rugby ball and stopping the stopwatch upon release of the pass. This analysis was only conducted on the ruck and direct/trap passes because the hold pass "catch and release" timings occurred too rapidly to record accurately. All speed timings for each participant were recorded and compared across conditions.

Results

The implementation of the TAG intervention resulted in improvements for each of the participants on each of the targeted passing skills. The percentage of task analysis steps completed correctly by Participant 1 during baseline and intervention for each of the three targeted passing skills is shown in Figure 1. For the rack pass, Participant l's performance improved from a mean of 59% (range 55-70%) of correct steps in the task analysis during baseline to a mean of 85% of correct steps during intervention (range 65-95%). For the trap pass, performance increased from a mean of 72% (range 67-75%) during baseline to a mean of 79% (range 75-96%) during intervention. For the hold pass, the mean percentage of correct steps in the task analysis completed during baseline was 89% (range 71-100%). Although the mean performance score at baseline was higher than the inclusion criterion of 85%, this skill was targeted during intervention. Performance increased marginally during intervention (M=92%; range=86-100%). The number of trials required to reach criterion ranged from 5-22 trials across the passing skills.

Participant 2's performance during baseline and intervention is shown in Figure 2. For Participant 2, the number of trials required to reach the criterion ranged from 3-5 trials. For the direct pass, Participant 2's mean performance during baseline was 84% (range 8190%) of task analysis skills performed correctly. This increased to a mean of 97% during the TAGteach phase (range 94-100%). For the ruck pass, Participant 2's mean number of steps completed correctly during baseline was 69% (range 55-85%) and increased to 94% during the TAGteach phase (range 85-100%). For the hold pass, Participant 2's percentage of correct steps completed increased from a mean of 87% during baseline (range 71-100%). Although the mean score at baseline was above the inclusion criterion of 85%, the skill was targeted due to inconsistent performances, with one of the steps achieving a score of 25% accuracy. Performance of this skill increased to a mean of 97% (range 86-100%) during the intervention phase.

Participant 3 was forced to withdraw from the study prematurely due to an injury that required surgery. Figure 3 represents the percentage of steps of each task analysis completed correctly for the two passing skills targeted for this participant prior to his withdrawal. Participant 3 required five TAG trials to reach criterion for the rack pass and 13 TAG trials to reach criterion for the direct pass. The mean percentage of rack pass task analysis steps completed correctly during baseline was 79% (range 65-90%) and increased to 92% during intervention (range 85-100%). For the direct pass, mean performance was 84% (range 72-94%) during baseline and increased to 93% (range 99-97%) during the intervention phase.

The differences in timed speed from baseline to intervention conditions were compared using a series of dependent t-tests. Significant improvements in speed of passing were observed for Participant 1 (ruck pass, t(13) = 2.39,p = .04; trap pass, t(13) = 3.91, p = .001) and for Participant 3 (ruck pass, t(9) = 2.416, p = .04; direct pass, t(18) = 2.66,p = .02). However, no significant differences were observed for Participant 2 (ruck pass, t(11) = .28, p = .78; direct pass, t(12) = .29,p = .78).

Discussion

To our knowledge, the current study is the first to assess the impact of TAGteach for skills training among high-performance athletes. Our results indicate that TAG is an effective method of increasing performance of athletic skills even among experienced athletes. The average increase in performance across participants and passing skills was found to be 11.7% (range 3-26%). This improvement might be considered large given participants' high levels of baseline performance. Also, anecdotal reports indicated that the intervention and its effects were well received by the rugby team's head coach. These results are in line with previous studies that reported a positive impact of TAGteach on a variety of athletic skills and the positive perception of TAGteach interventions by stakeholders (Andrews, 2014; Fogel et al., 2010; Harrison & Pyles, 2013; Quinn, 2013; Stokes et al., 2010).

In the current study, a stringent terminal criterion of 95% of task analysis steps performed correctly in three out of five consecutive trials was implemented for each targeted passing skill. However, in spite of this, the accuracy of performance of the skills was variable across the intervention period. Increases in the speed of passing were somewhat inconsistent as Participants 1 and 3 demonstrated significantly faster passing times during intervention relative to baseline for the two passes that were timed (ruck/trap and ruck direct respectively) but there was no difference observed for Participant 2. Thus, the outcomes of the current study indicate that the TAGteach intervention may have efficacy in increasing fluency in performance of rugby passing skills though further research is required to investigate such effects.

Fluent performance is both accurate and speedy (Binder, 1996) and is imperative in high-level athletics; Binder (2003, p. 14) has remarked upon this importance, noting "merely making the right move or being able to execute a play correctly is not sufficient for success. There is always a need for quickness, smoothness, and a lack of hesitation" in sporting performances. Fluent performance has been demonstrated to result in better retention or maintenance of intervention effects, endurance or the ability to apply the skills over extended periods of time, and better application or generalization of the skills to other contexts or settings (Binder, 1996), all key outcomes for athletic performance.

Harrison and Pyles (2013) noted initial decrements in performance when participants were required to increase the speed of skill execution that were remedied with further intervention, but speed of performance has not been otherwise considered during previous evaluations of TAGteach. Previous studies have suggested poor maintenance of skills learned via TAGteach (Stokes et al., 2010) and a failure to acquire skills during intervention (Fogel et al., 2010), and have suggested the need for procedural refinement (Fogel et al., 2010). Future research should examine specifically the contribution of fluency measurement and fluency building strategies to the efficacy of TAGteach interventions, including the impact on the maintenance and acquisition of targeted skills, performance variability and application of targeted skills in competitive settings. Given the paucity of professional rugby leagues in the USA, the opportunity to measure performance over an extended period to determine the maintenance effects of Tagteach interventions may be limited. However, outside of the USA other countries including Ireland where this study was undertaken, university rugby athletes often continue their sport following completion of their degree. Therefore, the potential to measure maintenance effects in future studies is likely.

There were a number of limitations to the current study. Neither the generalization of treatment effects to game play nor the maintenance of treatment effects following the conclusion of intervention was assessed. Further, it was not possible to collect treatment integrity data. Finally, the inclusion of participants for whom high levels of correct performance were observed during baseline may also be criticized. Previous research investigating the efficacy of TAGteach procedures has employed novices or those evincing lower levels of 2013; Quinn, 2013; Stokes et al., 2010) allowing for a clearer demonstration of the effects of TAGteach on performance. However, the current study included participants who in spite of their high levels of baseline performance were identified as underperforming relative to their teammates. Thus, it was considered of interest to examine the efficacy of TAGteach for improving the performance of skills already occurring at expert levels, given the focus upon TAGteach's efficacy for high-performance athletes.

Tiffany Elmore

Trinity College Dublin

Olive Healy and Sinead Lydon

National University of Ireland, Galway

and

Clodagh Murray

London South Bank University

Address Correspondence to: Olive Healy Ph.D., School of Psychology, Trinity College Dublin, College Green, Dublin 2, Ireland. Tel: 00353 1 896 1175, Email: [email protected].

References

Andrews, J. S. (2014). Evaluating the effectiveness of TAGteach for teaching yoga postures to novice yoga practitioners (unpublished master's dissertation). Retrieved from: http://scholarcommons.usf.edu/etd/5171/

Binder, C. (1996). Behavioral fluency: Evolution of a new paradigm. The Behavior Analyst, 19, 163-197.

Binder, C. (2003). Doesn't everybody need fluency? Performance Improvement, 42, 14-20.

Fogel, V. A., Weil, T. M., & Burris, H. (2010). Evaluating the efficacy of TAGteach as a training strategy for teaching a golf swing. Journal of Behavioral Health and Medicine, 1, 25-41.

Harrison, A. M., & Pyles, D. A. (2013). The effects of verbal instruction and shaping to improve tackling by high school football players. Journal of Applied Behavior Analysis, 46, 518-522.

Kazdin, A. E. (2011). Single-case research designs (2nd ed.). New York, NY: Oxford University Press.

Luiselli, J. K, Woods, K. E., & Reed, D. D. (2011). Review of sports performance research with youth, collegiate, and elite athletes. Journal of Applied Behavior Analysis, 44, 999-1002.

Martin, G. L. (2003). Sport psychology: Practical guidelines from behavior analysis (2nd ed.). Winnipeg, Manitoba: Sport Science Press.

Martin, G. L., Thompson, K., & Regehr, K. (2004). Studies using single-subject designs in sport psychology: 30 years of research. The Behavior Analyst, 27, 263-280.

Quinn, M. J. (2013). Utilizing TAGteach to enhance proficiency in dance movements (unpublished master's dissertation). Retrieved from: http://scholarcommons.usf.edu/etd/4751

Stokes, J. V., Luiselli, J. K., Reed, D. D., & Fleming, R. K. (2010). Behavioral coaching to improve offensive line pass-blocking skills of high school football athletes. Journal of Applied Behavior Analysis, 43, 463-472.

TAGteach International (2012). What is TAGteach? Retrieved from: http://www.tagteach.com/What_is_TAGteach

Caption: Figure 1. Percentage of correctly completed task analysis seps performed by Participant 1 during baseling and TAGteach for the three target passing behaviors.

Caption: Figure 2. Percentage of correctly completed task analysis seps performed by Participant 2 during baseling and TAGteach for the three target passing behaviors.

Caption: Figure 3. Percentage of correctly completed task analysis seps performed by Participant 3 during baseling and TAGteach for the three target passing behaviors.