Skip to content

Cognitive science and threats to free will

It is often asserted that emerging cognitive science – especially work in psychology (e.g., that associated with work on automaticity, along with work on the power of situations to drive behavior) and cognitive neuroscience (e.g., that associated with unconscious influences on decision-making) – threatens free will in some way or other. What is not always clear is how this work threatens free will. As a result, it is a matter of some controversy whether this work actually threatens free will, as opposed to simply appearing to threaten free will. And it is a matter of some controversy how big the purported threat might be. Could work in cognitive science convince us that there is no free will? Or simply that we have less free will? And if it is the latter, how much less, and how important is this for our practices of holding one another morally responsible for our behavior?

(A quick note to those who care, before moving on: here I am assuming with many who work on these issues that the kind of free will that matters is what we might call moral-responsibility-level free will. As Al Mele explicates this kind of free will, it is the kind such that ‘if all the freedom-independent conditions for moral responsibility for a particular action were satisfied without that sufficing for the agent’s being morally responsible for it, the addition of the action’s being free to this set of conditions would entail that he is morally responsible for it’ (2006, 17).)

Philosophers who have thought about how emerging work in cognitive science might threaten free will have identified at least five distinct types of threat. Each one has a common structure: emerging work is purported to suggest, render plausible, or establish claim X, and the truth of claim X is taken to (at least apparently) threaten free will. Here I’ll just summarize each kind of threat, and make a few comments about each one.


1] Determinism

Cognitive science might threaten free will if work in cognitive science is able to show that the brain is deterministic. Of course, whether determinism is a threat depends on what one means by free will. Compatibilists maintain that free will is compatible with the truth of determinism. So if cognitive scientists were to show that the brain – and especially the processes responsible for deliberation and decision-making – operated according to deterministic causal laws, only libertarians about free will would find this threatening.

However, cognitive science is unlikely to show any such thing. Adina Roskies notes that since “we have no objective access to either determinism or indeterminism” the best we can do is to operationalize determinism as predictability (Roskies 2014, 105). Given this limitation, we are unlikely to discover compelling evidence for indeterminism – deterministic systems can display stochastic, and thus seemingly indeterministic, behavior. But does this preclude us finding compelling evidence for determinism? Roskies asks us to imagine that we acquire full information about some area in the brain that subserves decision-making, such that “we could with perfect accuracy predict the subsequent behavior of the agent” (116). This might offer some inductive evidence in favor of determinism. But since any given brain area – and indeed, the brain itself – is an open causal system, it is always possible that an event somewhere else influences the operation of the system in question, rendering the process indeterministic.


2] Naturalism

Cognitive science might threaten free will if work in cognitive science is able to show that naturalism about the mental is true. However, as with determinism, whether naturalism about the mental threatens free will depends on what one means by free will. Let us understand naturalism about the mental as the claim that minds operate entirely according to natural laws, and thus that there is no place for souls or otherwise non-natural substances in our understanding of agency. If neuroscience were to show naturalism about the mental, would this threaten free will? Only if free will requires the operation of souls or otherwise non-natural substances.


3] Eliminativism

The third way cognitive science might threaten free will is related to the second, but distinct in that many naturalists are likely to find this possibility troubling (while they probably will not care about the second type of purported threat). According to this possibility, neuroscience is showing (or will one day show) that the best explanation of human agency need make no reference to intentions, decisions, and the like.

Certainly some cognitive scientists talk as if cognitive science were showing this kind of thing. Here, for example, is Francis Crick endorsing the possibility as fact: “your memories and your ambitions, your sense of personal identity and free will are in fact no more than the behavior of a vast assembly of nerve cells and their associated molecules” (1994, 3).

Now, most philosophers agree that if this kind of ruthlessly reductionist eliminativism were correct, free will would be under dire threat. But it is important to note that many philosophers, as well as many scientists, find nothing in current neuroscience that warrants this kind of eliminativism. It might turn out to be true that the best explanation of human agency is eliminativist, but at present many would argue that we have little reason to think this is so. For those on the lookout for threats to free will, however, the eliminativist threat is something to keep in mind as cognitive science progresses.


4] Modular epiphenomenalism

Modular epiphenomenalism is, according to Eddy Nahmias, the view that “those modules involved in conscious decisions or intention formation do not produce our behavior; rather other modules or processes that involve no conscious states produce our behavior” (2014, 12). The thought is that the discovery of modular epiphenomenalism would threaten free will since, according to Nahmias, “both ordinary intuitions and philosophical theories, compatibilist and incompatibilist alike, suggest that relevant conscious mental processes need to play some causal role in actions that we count as free and responsible” (12).

Nahmias is right about ordinary intuitions, as some of my own work in experimental philosophy confirms (Shepherd 2012). But it seems to me that the inference from these intuitions to an actual threat to free will requires some philosophical work. Why exactly is consciousness important for free will? This is an interesting question, and very few arguments have been offered in connection with it (a recent exception is Neil Levy’s new book: see Levy 2014).


5] Psychological integration

A related option is that even though consciousness itself is not essential for free will, a phenomenon often closely associated with conscious thought and deliberation is essential. It is difficult to pin down this phenomenon with any precision, but we might call it psychological integration. Consider an agent engaged in some bout of deliberation about whether to A or B. She is engaged in a search to find reasons in favor of A-ing and reasons in favor of B-ing. She is also engaged in deliberative processes of prospection, imagination, weighing of reasons, and so on. If she is psychologically integrated, then she is able to compare and contrast these reasons in a coherent way. These reasons are accessible to her deliberative processes, and they influence her deliberation and deciding in part by entering into an integrated cognitive space. If the agent’s decision-making were in fact driven by processes that were not a part of (or partly constitutive of, or whatever) this space – were not weighed in the same way against competing reasons, did not enter into her deliberative activities of prospection and imagination, etc. – then we would worry that the agent herself was not doing the deciding. We would worry that the agent’s decision was not free. After all, none of her deliberative activities interacted with the processes that actually drove the decision. The agent would appear divided to us, her behavior driven by forces she fails to understand or notice. If cognitive science were to show that agents are actually like this – perhaps even if only some of the time – we might thereby find a genuine threat to free will.


Crick, F. 1994. The Astonishing Hypothesis. Scribner’s.

Mele, A. 2006. Free Will and Luck. Oxford University Press.

Nahmias, E. 2014. Is free will an illusion? In Walter Sinnott-Armstrong (ed.), Moral Psychology, Vol. 4: Free Will and Moral Responsibility. MIT Press.

Levy, N. 2014. Consciousness and Moral Responsibility. Oxford University Press.

Roskies, A. 2014. Can neuroscience resolve issues about free will? In Walter Sinnott-Armstrong (ed.), Moral Psychology, Vol. 4: Free Will and Moral Responsibility. MIT Press.

Shepherd, J. 2012. Free will and consciousness: Experimental studies. Consciousness and Cognition 21(2): 915-927.

Share on

1 Comment on this post

  1. Amazing!!! I asked the same question in a google plus community weeks ago!!!

    can psychotherapist/psychoanalyst assume his own freedom and most of all the freedom of the fragile patient that is under a lot of pressure and undergoing strong transfer emotions and feelings?
    the examination of records in ethics cases, malpractice suits, and licensing board hearings. show that countertransference was acted out., it is obvious that sexually intimate behavior between therapists and their clients has emerged as an increasingly serious problem.
    if we look deeper we can see to some people that finished therapy, we can see patients turned to copies by some psychoanalysts.
    “the patient should be educated to liberate and fulfil his own nature and not to resemble ours” Freud

Comments are closed.