Friday, February 03, 2012

The Role of Learning and Memory in "Free Will"

One common definition of “free will” is that a person can decide or choose among multiple alternatives without being forced by physical laws, luck, fate, or divine will. Most of us feel we are in charge of our choices when no outside force requires us to make a particular choice. But it is fashionable these days for scholars to insist that free will is an illusion, a trick the brain plays on us. I will spare you the philosophical knots of specious assumptions and convoluted logic that that scholars tie themselves into.

Why do I bring this up? What has the “free will” issue have to do with learning and memory? Everything.

Human brains make choices consciously and unconsciously by real-time evaluation of alternatives in terms of what has been learned previously from other situations and of their anticipated usefulness. This learning occurs in the context of the learned sense of self, which begins unconsciously in the womb. The conscious brain is aware that it is aware of choice processing and makes decisions in light of such understanding. When a given alternative choice is not forced, the conscious mind is aware that it is not obliged to accept any one choice but is "free" to select any one of the available options. Such realization might even guide many decisions at the subconscious level. In either case, the probable value of each alternative is weighed in neural networks, which collectively reach a "decision" by inhibiting networks that lead to less-favored alternatives. Thus, network activity underlying the preferred choice prevails and leads to a selective willed action. What governs the  network activity causing the final choice is the activity in other networks, which in turn is governed by stored memories and real-time processing of the current environmental choice contingencies. 

What usually gets left out of free-will discussions is the question of how a brain establishes stored-memory preferences and how it evaluates current contingencies. These functions surely cause things to happen, but what is the cause of the cause? Any given brain can choose within certain limits its learning experiences and what it will store as lasting memory. Those choices in turn are often governed by what a brain has learned about the self-interest value associated with given contingencies. So, it is learned values that underlie much of choice behavior.

Brain circuitry assigns value, and values chosen are largely optional choices. The conscious brain directs the choices that govern value formation, reinforcement, and preservation in memory.

Now we are confronted with explaining how neural circuit impulse patterns (CIP) representing the sense of self can have a free will. First, I reason that each person has a conscious Avatar that is created by brain as an active agent to act in the world on embodied brain’s behalf, as explained more completely in my recent book.[1] This is reminiscent of the 3rd Century idea of a homunculus, a “little person” inside the brain. The modern view is that this homunculus exists in the form of mapped circuitry.

Certain maps are created under genetic control. These include the topographic map of the body in the sensory and motor cortices. Then there is the capacity for real-time construction of maps of the body in space that resides in circuitry of the hippocampus and entorhinal cortex. Other maps are created from learning experience from the near-infinite circuit capacity of association cortex. What is learned in these maps is stored in memory as facilitated circuit synapses and deployed “on-line” in the form of CIP representations of what was originally learned. New learning likewise exists as CIP representations in sub-network populations.

The Avatar itself is a constellation of CIPs representing the conscious sense of self. Certainly, by definition, the Avatar can make choices and decisions. The Avatar is released in wakefulness from its stored representation to make its own choices and decisions. Avatar processing is certainly not random, and presumably can occur with more degrees of freedom than found in unconscious mind.

If the Avatar exists as CIPs, how can something as “impersonal” and physiological as that have any kind of “will,” much less free will. Let us recall that “will” is little more than an intent that is often coupled with bodily actions to achieve the intent. This kind of thing occurs even in the circuitry that controls unconscious minds. These circuits automatically generate actions in response to conditions that call for a response. Such actions are stereotyped and inflexible whenever they are controlled without conscious oversight (Fig. 1).

Fig. 1 llustration of how alternative ideas or choices may be processed in the brain. Each alternative is represented as circuit impulse patterns (CIPs) within a subpopulation of brain, which be considered as constituting part of the sub- or non-conscious mind. Each population’s activity interacts with the others — and with the CIP representation of the Conscious Avatar. When activity level in any one subpopulation reaches a threshold, it suppresses activity in the alternative representation populations, leading to selection of that population’s activity as the choice result. The Avatar CIP is poised to influence activity in the alternative sub-populations and thus can help direct the final processing result.


The Avatar must have some criteria which its circuits use to make a given decision. Those criteria have been learned and remembered. When CIP processes operate in Avatar circuitry, the Avatar population activity can modulate the alternative-choice representations in the context of self-awareness according to the informational representations of past learning and value assessments of current contingencies. You might say that when the brain generated the CIPs to represent the sense of self, those CIPs were endowed with a certain autonomy and freedom of action not available to the other CIPs in the brain that constituted unconscious mind.

People who believe that humans have no free will are hard-pressed to explain why no one is responsible for their choices and actions. What is it that compels foolish or deviant behavior? Is our Avatar compelled to believe in God or to be an atheist? Is our Avatar compelled to accept one moral code over any other? Is it compelled to become a certain kind of person, with no option to “improve” itself in any self-determined way? Are we compelled in our choices of learning experiences? If so, what or who does the compelling? Are we inevitable victims of genetics and experience?

It seems to this Avatar that current debates about determinism and free will tend to obscure the important matters of our humanness. The door to understanding what is really going on is slammed shut by assertions that value choices and the decisions that flow from them cannot be free because they are caused by neural circuit impulse patterns. Free will debates distract us from a proper framing of the issues about human choices and personal responsibility.

While it is true that Avatar circuitry is programmed by genetics and experience, the Avatar makes choices about who to interact with and what experiences to value, promote, and allow. The Avatar can insist that some lessons of experience need to be remembered. In short, the Avatar gets to help shape what it becomes.


[1] Klemm, W. R. 2011. Atoms of Mind. New York: Springer. http://www.springer.com/biomed/neuroscience/book/978-94-007-1096-2