The existence of free will has been debated in philosophy circles for almost as long as there have been philosophers to fill them. Many have tried to reason whether we, as humans, control our own actions and fate, or if our actions are just really at the whim of destiny, god, or some other pre-determining force. The problem of free will has been argued by many people over different scopes. Numerous professionals, from philosophers to lawyers, have attempted to find the inner motives that govern our actions and the over-arching theme of these miscellaneous inquests is that free will is an incredibly abstract topic. The definition of personal freedom has been argued and changed as new discoveries and perspectives have emerged.

For the purpose of this article, we will use a more classic definition of free will, the definition described by Benidictus Spinoza in the 15^th century. In Spinoza’s work, “Ethics,” he discusses free will as “(t)hat thing is called ‘free’ which exists solely by the necessity of its own nature, and in which its own nature alone is the cause of its actions. On the other hand, that thing is necessary, or rather constrained, which is determined to a fixed and definite method of existence or action by something external to itself." Even though Spinoza argued that humans don’t have free will, I will use his definition because it is philosophically cut and dry and untainted by the new views that are being shaped by advances in cognitive science. I will use this argument to analyze in which of the three main frameworks – dynamicism, connectionism and computationalism – free will could exist.

Of the three framework, the one that has the hardest time explaining the concept of free will is dynamicism. Dynamicism states that cognition is a dynamic process, best understood in terms of differential equations than it is in computations or connections. The main idea behind dynamicism is that cognition is almost expressly controlled by the surrounding environment; sensory inputs are the independent variables in a differential equation whose dependent variables are all of a person’s possible thoughts and actions. In this view, the opportunity to think and act freely is never presented, and free will can't exist.

Even though this framework is the most ideal one for creating models that can work in and adapt to different environments, it falls short of explaining a number of factors, other than free will, that make human cognition exceptional. A biggest problem, in the eyes of Andy Clark, is that the dynamicism framework does not explain the “space for reasons.” Clark argues that our minds are knowledge-based and not physical-causal systems. This means our actions and our intelligence comes from knowledge we have acquired through living and adapting to our world and not just the environmental factors that are affecting us at any given moment. By using what we have learned through our actions and their consequences, we are able to make our own choices, using our acquired knowledge and the present situation as a guide. The space for reasons argument touches on all the factors that make free will impossible in a dynamic system since something controlled only by its environment can't exercise true freedom. It is important to note the dynamic framework is not proven wrong just because it does not have space in it for free will. The dynamic philosophy of cognitive science could be true, and our assumption of free will could be false.

The other two frameworks, computationalism and connectionism, both seem to have the ability to account for free will in human cognition. These frameworks look at how internal representations of environmental inputs are used to create human cognition, instead of the environment itself. This allows for more complex knowledge gain and control over oneself and one’s body. Computaionalism assumes human intelligence stems from the mind’s use of symbols to represent the physical world and uses these representations as a basis for knowledge and action. The computational theory of intelligence stems from human’s experience creating intelligent machines, computers and the strategies we use to program them. The computational view looks at how a system can create outputs by discerning inputs and following a set of given instructions on how to convert said inputs into said outputs.

This works very well in describing the intelligence of computers and related systems, but so far, it has been difficult to figure out how a computational system could ever be created without some higher power instilling the instructions of creating outputs from inputs upon them. It has been argued that a computational system can never have a truly original action since it will only be able to preform what its programmers tell it to. This is a critical flaw in the possibility of free will existing in such a system since the action of a person is governed by limiting instructions endowed from a higher power and not by the choices that the system chooses to make freely.

The last of the three frame works might be the best for figuring out how free will factors into human cognition. In the connectionist framework, cognition is a product of the activation/relaxation/back propagation of connections between different nodes that link inputs to hidden processors (neurons in the brain’s case) to outputs. Like the other two frameworks, this one is not perfect. Connectionist models are based (loosely) on the configuration of neuron networks in the brain, but their complicated nature makes creating a model of structured intelligence very difficult. Lucky for us, the concept of free will is extremely abstract and intrinsically unstructured, so free will may exist in a connectionist network, after all! If the ability for free action is held together by a connection of neurons, then the ability to decide one’s own actions should be destroyed when the connections are destroyed.

The destruction of neural connections is exactly what happens in people diagnosed with the behavioral variant of frontotemporal lobar degeneration, which is a mental disorder characterized by the atrophy of neurons in the frontal and temporal lobes in the brain. This condition does not hinder free will by taking away your ability to produce action. Instead, you have no ability to control what deeds you choose to do. The afflicted are totally at the whim of their “automatic response,” which is the first response your subconscious initiates upon the discovery of any possible stimulus. Sufferers do not find themselves constrained by the inability to act but the inability to choose what environmental input they decide to react to. Researchers studying the link between this disease and free will noted, “Patients with this disorder cannot suppress the urge to utilise an object in the immediate environment… it is the stimulus, rather than the circumstances, which guides the response. Such an individual is at “the mercy” of the environment.”

A person affected by this disorder can't consciously choose their own actions and can't exercise, what we would consider, free will. There may be a problem here if damage to the connectionist network between neurons can strip a person of free will. Maybe there is truly no free will at all, and we are controlled by these neural networks beyond our consciousness. After all, if something is free only because another, higher power allows it to be so, it is never truly free. We are, in a way, controlled by this higher power of neuron connections. However, there are some instances where a person can be stripped of their freedom by a power that had never had any prior control of their independence. A (purely hypothetical) example: If I were to throw the next random person I see into an inescapable cell, they would lose their personal freedom since I am constraining their ability to act as they choose. Even though they are not free now, it would be preposterous to claim that the now imprisoned person was never truly free just because someday they may have the misfortune of running into me when I’m trying to prove a philosophical point. The same could be true with free will and these neural connections. A person has the ability to exercise their free will until the connections brake down to a point of being unable to control the unconscious “automatic response.” Since the connectionist model is the only one that can explain the stripping of free will, it is also the only model that has free will as a possible implementation of cognition.

By breaking down the three models of cognition, I was able to find that connectionism has the best chance of including free will in human intelligence. This is not to say connectionism is the best model; it simply says if free will exists in humans (which I hope it does), this is the best model to represent how it may mesh with the rest of the mind’s function. If free will is not intrinsic to human cognition, then perhaps, the other two models are better representations for human perception. Since these models are created to find the best way to analyze and predict human behavior, free will might be an over-arching problem in all three of them. If something has free will, then it should be impossible to truly predict its actions since there is no set basis for its actions. Of course, based on the environment, a researcher may be able to predict what a human might do. If it starts to rain, it could be assumed that a rational person would try to get inside. Free will destroys the predictive function of all these models since they will never be able to predict who will go inside and who will stay out to dance.