What Makes Practice “Perfect?”

Posted by

We all know the famous quote, “Practice makes perfect.” When I was in elementary school, we had a local soccer team come speak to our school. The main speaker brought up the quote by Vince Lombardi, “Practice does not make perfect. Only perfect practice makes perfect.” I remember hearing this quote and leaving the assembly confused as to what “perfect” practice meant… Isn’t practice a time when you are allowed to make mistakes? That’s how we learn, right?

I love the following quote from Noa Kageyama written in her article titled “The Problem with Perfect Practice.”

I’d argue that “perfect” practice is just another name for deliberate practice. Mistakes aren’t the problem. The problem is not taking the time to articulate the specifics of the mistake, the cause of the mistake, and the potential solutions, so you can avoid making that same mistake over and over.

Why do we practice?

The Schema Theory (Schmidt, 1975, 2003; Schmidt & Lee, 2005) is the idea that organized sets of motor commands called motor programs are retrieved from our memory and adapted to a particular situation. It assumes that previously created motor programs are generalized to novel situations with similar conditions. To give the best instructions to your body in varied situations, the motor system collects “data” on the initial conditions, the motor commands, the sensory consequence of those commands, and the outcome of the movement. All of the information together forms a memory representation that encodes and relates the info based on past experiences with similar movements. This is called a schema. When a schema is able to update with new information, learning occurs. A recall schema allows us to retrieve the motor plan and a recognition schema helps us evaluate if the motor plan was executed properly (Maas et al., 2008).

However, if any information is unavailable after a movement, no schema updating (learning) can occur. Thus, if a learner doesn’t know if the action was correct, schemas cannot be updated. Transfer of learning occurs to other similar movements because schema rules have been generated for novel situations. In addition, practicing in a variety of environments will increase the stability of schemas. Schema theory also tells us incorrect movements actually allow for learning to occur. It helps us develop more precise error detection and correction mechanisms (Maas et al., 2008).

Thus “perfect” practice doesn’t mean mistakes cannot be made. Perfect practice means we understand a mistake was made, determined why the mistake was made, and have the ability to adjust the parameters to avoid making a similar mistake.

Principles of Motor Learning

The Principles of Motor Learning are guidelines as to which practice and feedback conditions are best for motor learning. Below is information and tips about the most relevant principles of motor learning:

  • Prepractice: Make sure the learner is prepared for practice with an explanation of the task at hand and the ability to be successful (Maas et al., 2008).
  • Practice amounts: Different tasks require different practice amounts. A large number of practice trials allow more learning opportunities. Why? It enhances recall and recognition schemas. Many instances of retrieval also help automatize motor plans (Maas et al., 2008).
  • Practice Distribution: More time between practice trials or sessions (distributed sessions) allow for greater learning than massed practice (less time between trials or sessions) (Maas et al., 2008).
  • Practice Variability: Constant practice vs. variable practice. We typically start with constant practice to stabilize a motor plan then introduce new parameters to help the plan to generalize (Maas et al., 2008).
  • Practice Schedule: Random practice means a particular motor plan is produced in successive trials, but the target is not predictable. In contrast with blocked practice, where the learner practices a group of the same target movement before moving on to the next target (Maas et al., 2008).
  • Structure of Feedback: Knowledge of results (e.g., try again) is different from knowledge of performance (e.g. your tongue came too far forward for that sound). Typically, when learning a new skill, you provide knowledge of results. Knowledge of performance is given when there is a better understanding of what parameters have to be adjusted to hit the target. How often? Typically higher frequency of feedback when first learning a skill and weaning off to help the learner develop self-awareness. When? When introducing a new skill, immediately after the completion of a movement; however, research has shown delayed feedback helps with generalization and transfer of skills (Maas et al., 2008).
    • Did you know there is a hierarchy of cues/hints we can give as feedback? We start by modeling (e.g. let’s say “cat,” with our /k/ sound, ready? Cat!”). Once a model is no longer needed, we give a verbal cue (e.g., “Try again and let’s see if you can get that sound in the back). Then sometimes a visual cue (e.g. a picture) can be used to remind the learner of their sound. The goal is to gradually reduce cueing to allow for self-awareness and self-correction to occur, ultimately working our way towards the learner independently using their new skill correctly.

How does motor learning theory apply to speech and occupational therapy?

Speech production involves motor plans and schemas that encompass the coordination of all speech production systems. Current conditions (e.g., noise, distance from the listener, jaw position) are evaluated and appropriate parameters are used based on the established recall schema. The consequences of the motor plan are evaluated and compared to the success of the movement. If the listener couldn’t understand the speaker, a parameter may need to be adjusted (e.g., tongue behind teeth to produce /s/) (Maas et al., 2008).

Regarding occupational therapy, acquisition conditions that facilitate retention and transfer of the learned skills are preferred. The principles of motor learning guide occupational therapy similarly to speech therapy with different motor systems (Jarus, 1994).

What is the point of homework/carryover activities?

Homework and/or carryover activities are given to continue to strengthen the motor plan. Unfortunately, many children and adolescents only have a short session once or twice a week with their speech or occupational therapist. That means the majority of their time is spent outside of therapy. This is why we want to stress the benefit of carryover work because it allows the recall schemas to adjust to different parameters and it strengthens recognition schemas. It gives the learner a better chance at automatizing the skill and generalizing/transferring the skill out of the therapy room.

Back to our original question, “What makes practice perfect?”

Perfect practice is individualized based on a learners needs. A general guideline:

  • The more practice the better! A little practice every day is more tolerable and effective than a lot of practice for one or two days.
  • Mistakes are allowed because learning occurs when we are able to identify our mistakes and correct them.
  • Specific feedback on performance (e.g. “Let’s bring our tongue back behind your teeth”) is preferred over knowledge of results (e.g. “Didn’t hear it that time”) after a skill has been established. This goes for positive reinforcement as well (e.g. “I love how your tongue was behind your teeth on that /s/ sound vs. “Good job!”)
  • Gradually reduce cues. A lot of parents tend to automatically model (we know you just want to help!), but it doesn’t always give the learner an opportunity to adjust the parameters without imitating!
practice makes perfect – hand lettering text positive quote


Jarus, T. (1994). Motor Learning and Occupational Therapy: The Organization of Practice. American Journal of Occupational Therapy, 48(9), 810-816. doi:10.5014/ajot.48.9.810

Maas, E., Robin, D. A., Hula, S. N., Freedman, S. E., Wulf, G., Ballard, K. J., & Schmidt, R. A. (2008). Principles of Motor Learning in Treatment of Motor Speech Disorders. American Journal of Speech-Language Pathology, 17(3), 277-298. doi:10.1044/1058-0360(2008/025)

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

Schmidt, R. A., & Lee, T. D. (2005). Motor control and learning: A behavioral emphasis (4th ed.). Champaign, IL: Human Kinetics.