An Analysis of Libet Experiments in Respect to Free Will

By Ender Tosun

In my book about free will titled “Free Will Under the Light of the Quran”1, I explained about the relationship of free will power with determinism, indeterminism, reductionism, physicalism. The following are from that book and presents and examines some experiments which may have important implications about free will power.

Often times, the opponents of FWP propose the Libet and Libet-type experiments as an argument against free will power.

A summary of the Libet and Libet-type 2 experiments is as follows:

The test subject is asked to do simple acts like moving his finger at times he decides within some time brackets of like 10 seconds. He is also asked to identify at what moment he was conscious of his related wills. A mechanism like a fast running clock is used for this. On the other hand, the brain activity of the agent is measured in order to find out whether there was any electric activity in his brain prior to the conscious will which correlated with the will and the consciousness about it, whether the brain activity would enable the prediction of the will of the test subject before the agent was aware of it. In more sophisticated experiments, the test subject is required to press either the right or the left button on a keyboard, and which parts of his brain becomes active is detected.

Furthermore, in some experiments it is measured whether the agent could veto his will before it was executed or stop during the execution.

The findings are as follows:

There is an electrical activity which is called “readiness potential” which starts to build up approximately 500 milliseconds before the agent consciously wills the related act. In some kinds of experiments, before 10 seconds it is possible to predict what button the agent will press with in 60% of the cases. The test subject is able to veto the will predicted based on his brain activity if he is informed to veto, until 200 miliseconds before the act.

1          Interpretation of The Experiments in Respect to Free Will

When I willed something, did my brain form that will even before I was aware of it, and did I just become conscious of it after it was finalized, without having any additional effect of my own on that will, but thinking that I have produced it?

This fundamental question has been very important regarding free will power.

Many opponents of free will power, strongly argue that Libet-type experiments empirically and convincingly support that the answer to the above question is positive: If before the agent is conscious of any will to move his hand a physical activity builds in the brain, and if only afterwards he feels that he wills to move his hand and then he moves his hand, this shows that the consciousness is a product of that unconscious physical activity. Or, if before ten seconds we can predict based on the physical activity in some parts of brain, with 60% accuracy that the agent will press the red or black button, this means that the consciousness depends on such physical activity. These show that the will is based on and reducible to the physical and deterministic processes. They argue that when there will be more sophisticated means, we will be able to increase the accuracy.

In the following sub-sections, we will see the flaws of such interpretations of Libet Experiments against free will.

2          The Flaws of The Interpretations of Libet Experiments Against Free Will (ILEF)

Note that the following flaws are in the context of using the Libet experiments against free will.

These experiments may prove consistent or useful in studying the functioning of the brain and its modules. This latter point is not analyzed here, since it is beyond the scope of this work.

Also note that the following considerations are based on the free will power as defined in part 3.1.2. So, they may be inapplicable in the context of some other definitions of free will.

2.1      Libet Experiments’ Design Which make These Experiments Irrelevant for the Existence of FWP

It may occur to a lady to will to cook meat-balls for dinner. But suddenly she may remember that there is no meat at home, but there is fish. Such are routine willing mechanisms. Our sub-modules propose draft wills. When our whole, or our essence which is the leader of our whole gets such a draft, it evaluates it by probing other sub-modules which may be relevant. Furthermore, it may get external data, such as checking if there is meat in the fridge if the memory does not return something certain. Or a motherhood module may say that the smallest of the children does not like meat so much, and that he did not eat anything at school.

In this respect, Libet experiments correspond to asking the subject “will to cook fish at non-periodical intervals”. That is, the subject is required to use her sub-modules, not the FWP of the essence of her sovereign whole. And ILEF concludes that the agent does not have FWP. So, Libet experiments as such are totally irrelevant to free will power, they cannot be used for or against the free will of the agent. Therefore, trying to use Libet experiments against free will is trying to disprove the existence of a thing by using an event which cannot contain that thing.

ILEF assumes that the outcome of the sub-module’s free will power is equal to the outcome of the agent’s free will power.

However, the whole of the agent and his essence assesses the draft will by its consciousness and the other powers. The default work of free will power is like this. We do not do whatever comes to our mind. But Libet experiments ask the subject to do whatever comes to his mind; hence, they ask the subject to do a behavior which is not a normal exercise of free will power.

If the accounting department of a company prepares a fraudulent tax declaration, and if it does not become effective before the top authority in the company approves and files it, then the company and that authority cannot be said to have no freedom just because it was prepared prior to the awareness and approval of that authority. They had the freedom, but of course not about something that did not reach them. So, unless Libet experiments prove that the will power at the agent level is executed, even before the agent is aware of it, they are not relevant to the free will power. If they claim that the conscious involvement of the agent is not effective on the will just because the will became predictable to some extent before the agent became conscious of it, this is false since veto tests demonstrate the opposite.

Can the experiment be redesigned so as to test the existence of a real and normal free will power? Of course, it can be. But, when it is redesigned so as to test a normal free will power, then there will not be any need to conduct the test, since the outcome of the test will be clear before it is run:

For example, to test really the free will power, in each round, the test subject must be informed of the prediction of the readiness potential, and he must be required to assess the variable implications of the outcome and finalize his will about moving his hand or not according to those implications. Such a realistic test recognizes that the willing process normally includes (1) an awareness about the draft will, (2) its control and assessment according to other modules and external data, and (3) approval or rejection stages.

The FWP of the sub-module observed in the readiness potential is not a coercive thing upon the FWP of the essence of the agent. It is not even at the level of the overall FWP. It is just one of the inputs. It may be accepted or rejected.

After being conscious of the first will, I may have a second will to change the first will; however, that second will may also have been prepared by my brain before I have been conscious of this second will. So, why would my ability to reject be an argument against ILEF?

The second will will necessarily process the consciousness aspect of the first will and related alternatives. But as the unity of the consciousness and of the whole of the agent is logically and empirically substantiated, and as there cannot be an upward infinite regress, we can say that once the draft will is encompassed by the whole of the agent and presented to other available modules, and assessed by them, then the whole of the agent has encompassed all that is possible to encompass. The second will encompasses the first will’s consciousness aspects and implications. Consequently, we cannot say that the brain produced the will deterministically or indeterministically, without the conscious involvement of the agent.

2.2      The Test Subject Is Aware of The Wills Since the Beginning and Before the Readiness Potentials.

ILEF do not consider the will of the test subject at the start of the test. If the test subject was told to injure each person in the test room at different intervals, probably he would not do it. So, when he agrees to proceed with the test, he is conscious already of what he will do, and he gives the relevant command to his brain at the start. He wills what to do, and he just left to a module the timing of the hand movements. He has also commanded to the relevant modules any restrictions. Therefore, he encompasses almost all things in terms of consciousness at the beginning. If a new thing occurs during the test, this will be incorporated within things to be willed and done. Otherwise, everything is settled in his consciousness in the beginning. If someone shouts during the test “Stop! Do not move your hand anymore!” then the readiness potentials will change according to the conscious will of the test subject.

2.3      Libet Experiments Prevent the Test Subject from Using His Consciousness.

These experiments are structured so that the consciousness of the test subject is excluded from the tests at the beginning as well.

When we plan a will, we say “I will do it next week”, or “I will do it when an event happens”, or “I will not do it”… Then when conditions appear, we develop more specific wills.

The test subject is not allowed in his main will when starting the experiment to form his secondary wills for each movement based upon certain patterns. For example, he cannot will to move his hand precisely at the end of five seconds intervals. Or he cannot decide in his main will, to move his hand upon hearing a noise. The experiment tries to artificially separate the will from the consciousness. Then the subject cannot do anything other than “hiring” a sub-module which maybe sets up a certain timing such as each 20 seconds plus minus 5 seconds.

It is clear that his secondary wills of each movement are tied to his main will. There is no possibility to say that the agent cannot incorporate his consciousness in his main will while determining the passage of time, or the noise as the thing upon which his secondary wills would depend. If instead of moving his hand, he was required to move a table from one side of the room to the other side, he would determine a different timing, and he would obviously have time to change his mind until he reached one side. So, his consciousness is active to some extent in the formation of his readiness potentials.

If the test subject was told to will anything freely, and if he decides to move his hand every 5 seconds, what will be the position of readiness potentials? Was he not conscious of each will at the onset of the experiment? Did he have free will power when he is set free? So, if he is not forced, he can be free and independent of readiness potentials.

Hence, when interpreted against free will, the Libet tests do not allow the test subject to incorporate his consciousness in his wills, so as to conclude that the test subject wills without consciousness. It does not look any different than putting a bird in front of the test subject and telling him to move his hand each time the bird moves, and then concluding that his conscious will was dependent on the movements of the bird, since the movements of the bird were detected to be some milliseconds before his conscious will to move his hand.

Obviously, some wills may be organized so as they depend on an external or random factor. However, this does not mean that all wills are like this.

Therefore, ILEF are not scientific and objective.

Anyway, Libet experiments which measure the vetoing ability of the test subject confirm to a limited extent the obvious effects of FWP observed in our daily lives.

2.4      Insufficient Precision

Let us imagine a test subject who is laying on his right, awake on the bed. We can predict that he will will to change his position. Because if he stays like that for too long, he may feel uncomfortable or even pain. Does this mean that he does not have any free will regarding his position on the bed? As explained in part 3.4.10, a minimal difference in will may make a big difference in the outcomes. And this big amplification power combined with reasoning may produce all that the agent needs in order to be effective on his own.

Likewise, if the test subject had the urge to move his hand at a certain moment in correlation with the readiness potential, does this mean that he could not delay or cancel it? Actually, the tests show that the test subject can cancel or delay these wills.

Readiness potential does not produce high predictability. The predictability is slightly over the random. For example, in the lateral experiments in the form of willing a or b, only 60% of the predictions is true. An unqualified guess would give 50% already.

On the other hand, the more the readiness potential makes the will predictable, the more obvious will the negation of the test subject be. The stronger the prediction on readiness potential, the stronger will be the argument in the negation tests in part 3. The weird implications of 100% predictability explained in that part shows that a sufficient predictability to support ILEF cannot be obtained in Libet type experiments objectively structured in order to identify the reducibility of FWP to unconscious brain processes.

Can a prediction based on readiness potential have 100% accuracy? According to quantum physics and Heisenberg uncertainty, this is impossible. About this impossibility there are two approaches: One of them claims that even if we had full knowledge about a previous state, we cannot predict the future state. The other one claims that we cannot predict, but this is due to the impossibility of having full knowledge about the previous state, because any measurement changes that which is measured. Both ways, the impossibility of 100% predictability of a future state is agreed upon as a well-established principle, by almost all physicists. Anything less than 100% predictability in Libet experiments will be insufficient to support ILEF.

Even if there is no 100% accuracy the prediction, but there is 99% accuracy. Is not the impossibility of full unpredictability trivial in this context? In other words, maybe in very exceptional cases the agent can override the physical, hence it may be possible that often the agent cannot overcome what readiness potential entails.

As long as there is no 100% predictability, free will claim will not be weakened. Because in terms of our tests in part, it is kind of a matter of black and white for the opponent of free will: If the agent is told that he will will the blue candy, and if he can will the red candy once in a million, this means that he can do this negation always, and this one test will show that he has the free will power in principle. Anything less than 100% predictability will not be proof for deterministic processes.

2.5      The Ability to Veto Shows That Consciousness Is Effective in Wills.

And [by] the soul and He who proportioned it

And inspired it [with discernment of] its wickedness and its righteousness,

He has succeeded who purifies it,

And he has failed who buries it [in corruption].

(Quran: 91/7-10)

And We have already created man and know what his soul whispers to him, and We are closer to him than [his] jugular vein

(Quran: 50/16)

The readiness potential is not a coercive factor. It is a part of the agent and it produces an inspiration to the essence of the agent. Yet, essentially the agent may negate it.

Even in the simplest wills, the time bracket of milliseconds allows the tests subject to veto what the readiness potential entailed. So, whether it is a will about timing of a movement, or whether it is a will involving lateral development of multiple readiness potentials, the agent has the capacity to veto the product of the readiness potentials. Hence, if within milliseconds an agent has this ability, then regarding issues where freedom of will can be exercised within minutes or hours or even years, the agent almost always incorporates his consciousness.

On the other hand, it should be noted that even after an act is committed, the agent may revisit it, and change himself or the act.

But whoever repents after his wrongdoing and reforms, indeed, Allah will turn to him in forgiveness. Indeed, Allah is Forgiving and Merciful.

(Quran: 5/39)

1 The book is available for free at: or

2 Benjamin Libet did some of the early forms of these experiments.

3 Note that the negation thought experiments in part are different than veto in Libet experiments in that the negation experiments in that part refute the full determinism as well. Veto experiments would be trivial in that veto power of the agent might be considered within determinism according to the determinist.

Bir Cevap Yazın

Aşağıya bilgilerinizi girin veya oturum açmak için bir simgeye tıklayın: Logosu hesabınızı kullanarak yorum yapıyorsunuz. Çıkış  Yap /  Değiştir )

Google fotoğrafı

Google hesabınızı kullanarak yorum yapıyorsunuz. Çıkış  Yap /  Değiştir )

Twitter resmi

Twitter hesabınızı kullanarak yorum yapıyorsunuz. Çıkış  Yap /  Değiştir )

Facebook fotoğrafı

Facebook hesabınızı kullanarak yorum yapıyorsunuz. Çıkış  Yap /  Değiştir )

Connecting to %s