Adventures in Cognitive Science 4: Common Coding Theory – How is Perception linked with Action?

Source: pixbay.com

Picking up on my last post about mirror neurons, this time I am going to write about the common coding theory. Just like my last post, I am not going to present one paper in detail, but what I am going to write is based on chapter 7 written by Robrecht van der Wel, Natalie Sebanz, and Günther Knoblich from the book “People Watching: Social, Perceptual, and Neurophysiological Studies of Body” edited by Kerri Johnson and Maggie Shiffrar.
The common coding theory claims that perception and action rely on common codes, this is a quite different view to the classical “sandwich” model, which assumes that information is processed in three stages: perception, cognition and action (this model is very similar to the input-process-output model in computers). These common codes do not represent actions per se, but they do reflect perceptions that are produced by actions. Because actions are coded in terms of their perceptual effects, perceptions and actions are represented similarly. As example think of two people sitting at a table and one is pouring tea from a teapot into a teacup, while the other person watches: both people have a similar representation of the action according to the common coding theory.
First ideas similar to the common coding theory can be found in the works of William James, the “Father of American psychology”, who stated that imaging an action creates the tendency to do that action. Since it was a functional theory no particular prediction where this is happening in the brain were made, but with the discovery of the mirror neuron system a possible neuroscientific explanation of this theory was found. The mirror neuron system comes from neuroscience, the common coding theory comes from cognitive psychology.

Empirical Evidence

Action Observation induces Action Simulation

Common codes are assumed to be used to simulate observed actions, this type of simulation is only related to actions, not to mental states like in the theory of mind, which is about simulations that “put you into someone else's shoes” (when talking about mirror neurons, I mentioned that they are thought to have some connection to empathy, but nothing has been proven so far, this is where the theory of mind comes into play, the common coding theory does not make assumptions about this, it just focuses on the action understanding aspect). The basic idea behind action simulation is, that it allows an individual to make fast predictions about the outcomes of actions of others and therefore allows a faster interaction with the environment.
One question that spawned from the theory that people use internal models to simulate other's action is: “How accurately can an observer predict the consequences of an observed action, depending on the observer's own motor repertoire?” This question was investigated by letting people throw darts and recording it. In the second part of the experiment, the participants were grouped into pairs and also had to watch the videos of another person throwing darts. In both cases the participants had to predict, where the darts would land. At first there was not much difference, whether the participants saw themselves or the other person, in later trails the participants were able to predict the outcomes of their own actions better. The reason why there was no difference in the first trials is probably because the participants had to get used to seeing themselves from an outside perspective, but the experiment suggests that, after they got used to the different perspective, the were able to give better predictions of the outcomes of their own actions than for others, so particular aspects of their own motor repertoire seem to have improved the accuracy of the predictions of the outcomes of their own actions.

Action Expertise influences Action Observation

If action and perception have similar codes, they might influence each other, so action perception might influence action execution and the other way around. One way to see if this is the case is by taking experts in a certain domain and compare their perception to those of novices. The main claim is that if action shapes perception, experts have a different perception for the action domain in which they are experts compared with novices.
One study that studied this phenomenon examined how professional basketball players (experts) have a different perception whether a ball hits a basket compared to novices, coaches and sport journalists (expert watchers). The players were better when it came to predicting whether a throw was successful when showing them only short clips, but when they saw longer clips, which started before the basketball left the hand of the thrower, they made much better predictions if a throw is successful or not, compared to novices and sport journalists. In a second experiment researchers measured the Bereitschaftspotential (“readiness potential”, a measure for the activity in the brain which leads to muscle movement) for expert players and expert watchers when they watched clips of basketball shots: in both groups this Bereitschaftspotential increased when they watched basketball clips, but not when they were watching static images or kicks in football (soccer). Novices did not show any increase in their Bereitschaftspotential. This suggests that people with a visual and motor expertise for an action both run a simulation when observing this action, but for an expert player the nature if this simulation is probably more directly related to the observed action than for expert watchers. Since many expert watchers are also expert players (or were at one point), it is not clear, whether experts made better predictions (compared to novices) because of their expertise in watching, doing the action themselves or because of both. Similar experiments done with dancers came to a similar conclusion.

Common Coding Theory and the Ability to Act

The common coding theory provides a good explanation on how individuals perceive actions of others, but it raises the problem: “how can people know that whether the action is their own, or is from someone else?” The theory of apparent mental causation provides one answer to this question: it says that an individual experiences an action as his own, when a thought about the action arises just before the action happens, when the action is consistent with this thought (when the thought and the action are the same) and when no other cause for the action is present. Another answer to this question would be the sensorimotor account, which suggests that the perceived consequences (which are perceived through the sensory system) have to match the expectations.

Conclusion

The findings regarding the interaction between perception and actions suggest that viewing perception and action as independent processes, which serve higher-level cognitive functions, is outdated. The empirical evidence rather suggests that action production and action perception rely on common codes, which help to predict the actions of others.