If we are to compare the human mind with a computing device, it would soon become clear that it is peculiar in its methods. In Clark’s (2014) lecture and in Pfeiffer’s (2012) lecture, the concept of embodiment is used to show that many of the actions assumed to involve complicated mental processing can occur by virtue of the characteristics of the system. Recent work with robots has brought this facet to bare with passive walking devices, which use the swinging of the attached limbs to create smooth walking motions. All that is needed is a gradual decline and gravity will cause the device to move. Both Clark and Pfeiffer demonstrate clearly in their talks that instead of building a robot that controls every facet of the action, using the body of the system to take on most of the processing creates a much more efficient system. In such as system the body itself will do most of the walking and the computing system can work through gentle nudges and corrections to maintain its action.
According to Clark, the human brain works as a Hierarchical Predictive Processing device, working tirelessly to predict future sensory inputs. Learning occurs from the discrepancies between the actual inputs and the predicted ones through subtle corrections and alterations. The system is proactive and constantly interacts with the environment to test its predictions and assert its effectiveness in making them. The predictions are a template that it attempts to apply to the world. If it does not match, it can use the differences to create a more accurate prediction. This prediction system is extremely efficient, using the least parameters needed for accuracy as possible. Just like the previously described locomotion systems, it will outsource any functions it can to the outside world so that it can work with minimum effort. However, this minimal approach to prediction can also lead to failures. Clear examples of these failures can be seen in visual or auditory illusions such as the hollow face illusion, the sine-wave speech illusion, and the McGurk Effect.
Information is entering the system in a constant flow, and the organism interacts with its environment to choose the next stream of sensory inputs. This creates a shifting dynamic entanglement between the brain, the body and the world. Using this view, it is clear how other processing functions can be outsourced to the world, such as keeping reminders in your mobile phone. Clark and Chalmers (1998) describe an individual named Otto, who suffers from alzheimer’s disease and is incapable of relying on his own memory, and so employs a journal to take over this task for him. A similar example is described in Clark’s (2014) lecture, where he describes Patrick Jones, a catholic deacon who overcomes his long-term memory deficits through the use of evernote on his smartphone. In both these examples, we can see how instead of utilising mental resources to attempt to store information mentally, external sources are used for storage instead. But memory is not the only example of externalising cognitive resources. The use of gesticulating while speaking has been shown to be more connected to mental processing than communication. People will gesture when speaking even in the dark or when alone and preventing individuals from gesturing will reduce their performance. Showing an important integration with the entire system.
In this new era of smartphones, the potential to outsource cognitive processing has increased dramatically. The answer to nearly every question is but a Google query away, and multiple sensors can be used to accurately log past locations and actions. Technology is working to further reduce the mental processing required for a variety of tasks, which can fundamentally change how we think. It then becomes clear that sources outside the mind not only aid in certain cognitive functions but are integral to it. Certain utilisations of external resources mirror internal mental processing so thoroughly that it would be easy to connect the two. Is having a smartphone remind you when you get home to take out the bins any different than the memory surfacing in your mind independently? This train of thought is the basis of the Clark and Chalmers (1998) argument for the extended mind. They conclude that because of these similarities, it can be said that the external resources involved in these processes are a part of the system, in that the resource becomes an extension of the mind itself. In the example of Otto, his journal is an extension of his mind just as someone’s memory is an extension of theirs.
A problem subsequently arises from the vague definition that both the concept of “mind” and “cognition” hold. As Adams and Aizawa (2004) observe, there is “no well-established theory of just exactly what constitutes the cognitive”. This broadness in definition means that each individual will carry their own notions of what does and does not constitute as a cognitive process. Adams and Aizawa, for example, define cognition to be “constituted by certain sorts of causal processes that involve non-derived content”. Using this definition, the Clark and Chalmers conclusion of an extended mind is completely incompatible. As you may note however, if you ask any individual in the general public, they will not return this same definition. What you will find instead is a variety of different constraints and possibilities attached to each personal account. The result is then not a discussion on the extended mind but on the extended definition. A satirical parody of the topic, titled the Extended Breath, works to show how depending on the scope of your definition, breathing can involve just the lungs or the entire environment around the person, including respirators, the oxygen, and every other facet that can be attached to the system. How far into the system you go to explain breathing is an arbitrary line that depends on how deep an explanation you need. To try and argue over where that line should be becomes a futile pursuit in explaining to the world your unique definition of breathing. I’m not interested in an extensive account from every person on where that line should be, and so fail to understand why Adams and Aizawa work so vehemently to draw such a line with cognition. There seems to be no winners in discussing an extended definition.