An Introduction to Applying Feedback to Increase Motor Learning.

Updated: Aug 12


Hi everyone. This is Steve from Neurovlog. Glad to have you aboard again. Thought I'd do some animation stuff tonight instead of the typical, live action sequences that we've been doing. And, tonight I'm gonna actually be talking about motor learning, and, specifically, about augmented feedback. Now, feedback and motor learning have actually been out there in the literature for decades and decades. They've been an area of study in all kinds of areas of performance. It's just that in the last 20 years or so people have grabbed on to it for the purposes of clinical care. So, here are some of the resources i want to start off with. My doctoral degree is actually in motor learning from the 90s, and there are a number of great textbooks on motor learning that give you a general overview of some of the concepts that I'm going to talk about. So, here are some of them. I wanted to put up this chart for starters just to give you an overview of what feedback looks like. Now, on the left there you can see that we have feedback - internal feedback - and it simply refers to feedback derived through our own devices. So, things like seeing the outcome of our of our performance, feeling the outcome of our performance, adjusting our grip or adjusting our stance or something like that based on how successful we are. Those are pretty obvious but it's also important to note that when I have someone who has their systems compromised, (i.e., they have an impairment) that becomes more important for me to rely on that external feedback - that cueing during a performance; knowledge of results; knowledge of performance. Things like that that I as the therapist am going to provide in lieu of the fact that the patient doesn't necessarily have some of those internal feedback mechanisms in place. Now there are a few caveats to providing feedback. First of all whether it's desirable to provide one or both types of augmented feedback; that is either knowledge results, knowledge of performance, or both...is mostly determined by the characteristics of the learner, such as their skill level, their past experiences, and things like that, and also the nature of the task being learned. If someone is a newer learner, for example, high precision feedback is important but you need to also provide time for error correction and for error detection. And, I see this a lot in an experienced therapist where they're just very excited to provide instruction and feedback and modeling and all those types of things, but it's also important to lay off a little bit, right? To allow them to learn the skill without you necessarily puppeteering the skill because we want the skill to transfer to the real world environment when you're no longer with them. Now we also know that the use of sensory information, or biofeedback or video feedback, is really valuable in clinical things because it conveys information to a patient about things they might not otherwise observe. It could be muscle activity through biofeedback, but it can also be things like video feedback if you're providing for example video from your phone. Now, importantly the recent research on the optimal amount of feedback suggests that more is not better. And, again, I think about especially inexperienced therapists (whoop there goes my alarm). I think about inexperienced therapists who are so excited to share information, and so excited for their patients to learn, and they tend to provide a lot of feedback, even though the initial performance of a skill is facilitated when feedback is provided after every practice attempt. It's subsequent removal and withdrawal, so in other words, not providing as much as the patient gets used to the activity leads actually to poor retention. Now, you might think to yourself, well, I need to provide lots and lots of knowledge of results, lots and lots of feedback to the patient. And, there's even suggestions in the literature about this...but the thing is is that we know that more is not necessarily better. There's something called the guidance hypothesis: that says that even though high knowledge of results frequency guides the learner towards correct performance very quickly, that over time it promotes over dependence on this feedback. So, people are dependent on it, and then when you're no longer with the patient, their performance declines because you're not with them anymore. On the other hand, and interestingly, low no knowledge of results frequencies (so less knowledge of results) forces the patient to go into problem solving and independently explore the skills in her inherent dynamics and movements and they actually have better retention of the skills. So, at first it looks more messy, but then over time the skill is actually retained more effectively. Finally, there's something called summary knowledge of results and average knowledge of results, or summary kr and average kr. And, what that basically means is that you're waiting for a period of time; waiting for the patient to perform the skills - say five times - before you provide feedback. So, they get that chance to explore the movement; figure out how the movement works; figure out where they should position different elements of the movement or different body parts, i should say, and what's going to lead to the most success. And, then after those five attempts, you provide knowledge of results, so you're providing it on an augmented feedback schedule rather than just providing it every single time. There's also something called knowledge of results delay or knowledge of results withholding. And, what that simply means is that you either delay the feedback for the patient to provide you with feedback, or you withdraw feedback entirely over time. So, why might I want to do that? Well, maybe, I want the patient to tell me how the movement is going to build insight so that they can become an independent learner when i'm no longer with them. That would be a great reason to do it. That's something that my team's done for 15 or 20 years with constraint induced therapy and with other types of therapy. So, play around with those different types of feedback and specifically those types of knowledge of performance and as always you can always find us at www.neurovlog.com. Click on the join button to join us and have access to more of these vlogs as well as infographics and other resources. Thank you.