“Get set”...“use your legs”...“keep your wrist locked”...“step in!”...“move forward”...“hit higher”...“follow-through”...
If you walk by a tennis court during a lesson, these are only some of the phrases you’ll hear being shouted from across the net. The problem (beyond their vagueness) - as we alluded to in last week’s post - is the constant bombardment of these phrases. At times, players don’t just hear one cue or piece of feedback, but 5 or more. And further to that, it’s on every ball! Is this feedback facilitating learning? Or hindering it?
In this follow-up post, we’ll not only look at when we should provide feedback, we’ll also look at how often we should be providing pieces of feedback to players and why this is important (hint - it has to do with how our brain’s ability to process information).
In terms of feedback timing, there are 2 types - concurrent and terminal.
When providing feedback during the execution of a task, we would consider that to be concurrent. For example, when an athlete is moving the barbell up during a squat, you might say “push, push, push, push, push”. Or when a rally ball is short, you might say “up, up, up, up”.
Concurrent feedback is something we all use. I often find that after I’ve initially set-up a drill, providing feedback concurrently accelerates the drills effectiveness. Once the drill or activity is running, I try to avoid this type of feedback as it facilitates implicit, rather than explicit, learning to take over.
Overall, coaches and researchers caution against the over-reliance of concurrent (and constant) feedback as it often turns into ‘white noise’, not benefitting the athlete in any meaningful way. From a player and parents perspective, this may sometimes be a challenge - “the coach doesn’t say much”. Truth is, the coach is likely purposely holding back their feedback, allowing the player to ‘figure it out’ before attempting to interve.
Now of course there’s a difference between and absent-minded coach, one that just isn’t in tune with what’s going on, and one that’s withholding feedback for the player’s sake. But those of you reading this post are likely part of the latter, rather than the former.
Then there’s terminal feedback…
When we provide feedback after the execution of a task, that’s referred to as terminal. During a cross-court exchange, if you provide feedback after a player hits a shot, but the rally is still going, that’s not terminal, that’s still concurrent. The rally must end, then feedback given, for it to be considered terminal - otherwise, the information won’t be processed effectively.
But that’s not the only time terminal feedback exists. If you’re going through serve practice, feedback could be given after one attempt - i.e. immediate. Or feedback could be delayed - after a sequence of 10 serves, after the drill, after the practice and so on.
Providing feedback after a practice or a match is typically considered a debrief - and it’s probably more important than what is actually said during practice (and especially more important than the 10 cues that were said during a forehand drill).
Here’s what elite coaches from other sports (Altis - Elite Athletics Training) have to say about terminal feedback:
“Discrete skills such as a clean or squat are generally addressed with terminal cues as there is opportunity post skill to debrief. However, key point – wait until you see a consistent error before jumping in and correcting! Don’t create athletes that can’t operate without your feedback. Allow them space to develop their problem solving ability – this actually builds confidence.”
I believe the same can be said in tennis as the serve, groundstrokes etc are considered to be classified as either discrete skills (i.e. there’s a definite beginning and end to them) or serial skills (a series of discrete skills).
Interestingly enough, research has indicated that when athletes are given the choice, they prefer to receive feedback about 30% of the time. In many tennis settings, however, the reverse (or worse) is true - coaches are giving feedback 70% of the time or more!
It does depend on the athlete you’re working with and where your skills as a coach lie. Younger coaches tend to say a lot to prove that ‘they know what they’re talking about’. Experienced coaches have ‘been there and done that’ and don’t need that validation.
For instance, ever seen a practice session with 2 pro players and an experienced coach? Very little is said during the actual session. When I work alongside elite coaches, we purposely set objectives before practice, provide minor cues during breaks/changeovers and a more in-depth debrief at the end of a session. This is even more true as we approach competition periods - I should reiterate, players need space to problem solve on their terms.
But beyond that, there are different feedback frequencies that coaches can explore with their players. They are as follows:
With faded feedback, a coach will initially provide feedback on every (or almost every) attempt. Research suggests that this helps accelerate the learner’s path towards the movement goal. As the movement becomes more proficient, feedback is provided less frequently - in effect, fading out. The ultimate goal being that the learner can achieve the intended movement without a dependence on the coach and/or the feedback.
The beauty here is that feedback can also be faded back in - in case the movement has regressed in some way. Once it’s back on track, the feedback again is withdrawn.
With this form of feedback, a preset ‘degree of acceptability’ is established - with no feedback given when performance falls within the bandwidth and feedback given when it falls outside of the band of acceptability. The key here is that the learner is aware, before the fact, that if nothing is said, the movement is basically ‘correct’.
Research (Sherwood 1988) found that when the band was larger (~10%) compared to the target goal, it was more effective than smaller bands (1-5%). The theory being that less feedback (because of a larger bandwidth) will produce stable and consistent actions over time.
Here, feedback is given after a series of attempts - like 5 or 10, for instance. Interestingly enough, this feedback type has been shown to be more effective than trial-to-trial feedback - even though mistakes can be higher during practice, with this approach.
Interestingly enough, researchers found that getting feedback after every attempt promoted too much dependence. At the same time, feedback that was too infrequent (say, after 100 trials), didn’t guide the learner efficiently enough. Based on several experiments, feedback post about 5 trials seems to be most optimal when it comes to longer-term learning.
Tip: If you’re basket-feeding, try for multiple series of about 5 attempts, before intervening with feedback (even if you detect an error beforehand). This approach might have several benefits - first, the player has freedom to self-correct. Second, 5 attempts is still relatively low, so they won’t ingrain a bad habit. And lastly, from a physical standpoint it’s more specific to the demands of actual tennis-play (i.e. work:rest ratios).
This is pretty self-explanatory - feedback is only given when the learner requests it. I’ve encountered this on many occasions when working with elite players. But is it effective?
According to a throwing task study, participants that self-directed feedback, had better throwing accuracies compared to a group that was given faded feedback (feedback frequencies and types were matched). How can this be? Wulf and others in this field of study found that when feedback is learner-determined, it tends to be requested more following successful/correct attempts, compared to poor ones. Isn’t that interesting?
It seems that these performers crave positive reinforcement as this drives further motivation to ‘do it well’; thus, having a beneficial effect on learning.
Overall, the key takeaways (as I see them) are as follows:
Give students more chances to use intrinsic feedback (as we spoke about in last week’s post). If they are aware of the desired goal/task, let them feel the ball, hear it come off their strings, see the outcome and if necessary, make self-adjustments.
Step in with feedback when a clear error pattern is observed (hint, a pattern means it happens many times, not just once or twice).
More often than not, provide terminal feedback - determine when is the most appropriate moment in your setting, based on your athlete and their level of expertise.
Reinforce good movement/attempts more often than you’re accustomed to (and watch what happens).
Scale feedback to develop self-sufficient players, rather than over-dependant ones.
I’d like to finish this topic (for now) with the following quote from Schmidt and Lee (Motor Learning and Performance 2014):
“An instructor could give feedback about countless features of the action after every performance attempt. But, overloading the learner with too much information is a potential problem. Information-processing and memory capabilities of the learner—particularly a child—are limited, so it is doubtful that the youthful learner can take in and retain very much information during multiple feedback presentations. It is also doubtful that the learner can be very effective in correcting the next action in more than one way, particularly with feedback about motor patterning…. In general, too much information is generally not useful.... A good rule of thumb is to decide what error is most fundamental and focus the feedback on that.”
WANT TO BE UP-TO-DATE ON PROMOS AND THE LATEST SPORT SCIENCE & TENNIS RESEARCH?
No fluff or spam - just a weekly post and the latest promos.