Learning Through Observation

Recently, I sat in a lecture on observational learning, which was supplemented by a another lecture on self-regulated learning. Surprisingly, there hasn’t been a handful of research regarding observational learning as it pertains to motor skill learning and performance. I wanted to dig in a little deeper and share with you a quick overview.

It all started with Giacomo Rizzolatti (Discovery of mirror neurons) when his work with canonical neurons and the neural representation of motor movements in monkeys led to the discovery of mirror neurons. The basic idea was that canonical neurons are transforming affordances of objects. Meaning, it’s enough to observe a door knob for you know how to grasp it (in this case, it was a peanut). But it was soon realized that in some cases, just showing, wasn’t enough. In order for the monkey’s neurons to fire, the object had to be grasped. This led to the idea of congruence. The monkey needed to be able to also grasp the object in order for its mirror neurons to fire when a person grasped the same object.

Mirror neurons are neurons with motor processes that fire not only during action execution, but also when observing someone else performing the same or similar action. Likewise, observation of movements also activates the same areas that are used to preform those movements. In a review by Iacoboni (2009), it was concluded that;


  • Mirror neurons fire during observation, but not during observation of pantomime. Meaning, they are goal oriented.
  • Mirror neurons fire when say graspable objects are occluded by a screen.
  • Mirror neurons discharge in response to the intention or goal rather than action itself.
  • Mirror neurons are Multimodal; they can discharge to sound.
  • Mirror neurons are related to motor experience with a given action.

As a function of experience, Orgs et al., (2008) found that event related desynchronization in alpha and beta frequency of experts was modulated by an individual’s expertise with a certain movement style. So yes, the more you see it, the better. Further, the phenomena of observational learning has played in imperative role in motor learning albeit the research evidence is not immense. Coaches and clinicians are constantly using demonstration as a tool to enhance the rate of learning or optimizing current performance.

Observational Learning

Horn et al., (2007) found that novices in a model group, where they were observed a video model, learned a maximum velocity back-handed reverse baseball pitch to a greater extent than those who practiced based on just verbal instruction/guidance. What was really interesting was that the model group individuals showed immediate change in their intra-limb coordination, mimicking closely to the model’s relative motion pattern. In fact, ball speed was also improved. As a result, in early acquisition (i.e., rehabilitation) a model may represent an efficient and stable behavioral change that can enhance the rate of learning. In other words, a model can also be used as a constraint that allows salient information to be perceived. Watching an expert is typically the approach coaches adhere to when teaching a motor skill. Sometimes, its even the coach itself, whom they think is the expert. Clark & Ste-Marie (2007) wanted to see what would happen if the self was used as a model intervention. They used children who were learning how to swim. One group either saw a videotape of their own performance (self observation) where as the other group saw an edited video of their own best performance (self modeling). The self modelling group performed better. They concluded that implementing self-modelling interventions is a useful strategy to optimize learning.

Practicing in Groups

 Dyad training is considered effective and efficient. In fact, most teams practice in dyads. In slight contrast, rehab settings most often times don’t. Shea and colleagues (1999) compared three groups (individual, dyad alternate, and dyad control) using a balance task. In the dyad alternate group the order looked something like this; partner 1 went while partner 2 observed and on the next trial, partner 2 went and partner 1 observed. In the dyad control group, partner 1 went, performed all trials while partner 2 observed, and then they switched. They found that the dyad alternate group did best in retention whereas the dyad control group came in second, still doing better than the individual group. It was concluded that dyad training is beneficial due to observational learning, increased motivation (support/competition) and sharing or receiving feedback. Thus, practice should include observation and dialog between learners. Applying this type of framework is something coaches should strive to do. Lastly, Granados & Wulf (2007) found that observation and dialogue are also beneficial to motor learning in dyad practice, though these results should be extended to larger groups.


 The aforementioned literature goes in line with the soft versus hard assembled mechanisms debate. A hard assembled mechanism is independent of the immediate context, but is revealed across multiple contexts. For example, my rule about a “swish” in basketball remains whether on a basketball court, shooting a paper ball into a garbage can, etc. In contrast, a soft assembled mechanism is constrained within context. As mentioned by Kloos & Van Orden (2009), this can be the kinematics of a limb in a particular action. Like Bernstein (1967) alludes to, repetition of this soft assembly will reveal assemblies that have unique kinematics, albeit not context free. I bring this up because motor learning is typically associated with prescriptive, direct learning approaches that enable such “soft assemblies” to be formed, providing only temporary solutions. Here, it is demonstrated that observational learning is a powerful tool that enables coaches and clinicians to constrain information without using a prescriptive approach. As a result, you no longer have the formation of “soft assemblies,” but more efficient and stable behavioral changes.



  • Coaches should integrate observational learning during practice.
  • A self-modeled approach, where good performances are seen and referenced improve motor performance and learning.
  • When practicing in dyads, allow for dialogue to occur and strategies to be verbalized.
  • Particularly in rehabilitation settings, observational learning is powerful for the efficient and effective behavioral changes that allow for retention and transfer.



 Clark, S. E., & Ste-Marie, D. M. (2007). The impact of self-as-a-model interventions on children’s self-regulation of learning and swimming performance. Journal of sports sciences25(5), 577-586

Granados, C., & Wulf, G. (2007). Enhancing motor learning through dyad practice: contributions of observation and dialogue. Research quarterly for exercise and sport78(3), 197-203.

Horn, R. R., Williams, A. M., Hayes, S. J., Hodges, N. J., & Scott, M. A. (2007). Demonstration as a rate enhancer to changes in coordination during early skill acquisition. Journal of Sports Sciences25(5), 599-614.

Iacoboni, M. (2009). Imitation, empathy, and mirror neurons. Annual review of psychology60, 653-670.

Kloos, H., & Van Orden, G. C. (2009). Soft-assembled mechanisms for the unified theory. Toward a unified theory of development: Connectionism and dynamic systems theory re-considered, 253-267.

 Orgs, G., Dombrowski, J. H., Heil, M., & Jansen‐Osmann, P. (2008). Expertise in dance modulates alpha/beta event‐related desynchronization during action observation. European Journal of Neuroscience27(12), 3380-3384.

 Shea, C. H., Wulf, G., & Whltacre, C. (1999). Enhancing training efficiency and effectiveness through the use of dyad training. Journal of motor behavior31(2), 119-125.



Leave a Reply