The question of free choice is one of the “hard” philosophical questions that has been debated for thousands of years. Most people, if asked if they have free choice – or free will as some call it – will say with strong conviction that they do. We really don’t like the idea that we are totally predictable or that life is predetermined.

- Some argue that everything is determined by the initial conditions at the time the universe came into existence (if it did). Those conditions and their evolution to the present instant are the result of a set of natural laws.
- Others think we are choosing freely when we choose or act
*in accordance with our own desires and values*. - Finally, a free choice is one where
*we are able*to choose something different than what we, in fact, do choose.

These three different views are by no means an exhaustive list of all the variations of opinions, but they briefly cover a broad range of ideas proposed.

- With respect to determinism, it would seem the theory of quantum mechanics – at least as we currently understand it – would strongly suggest that type of “hard” determinism is unlikely, or even impossible. Quantum theory is filled with truly random outcomes that don’t actually exist until they are measured or observed.
- Are we exercising free choice when we do what our habits and belief systems would almost certainly cause us to choose?
- We may choose to act differently than our habitual pattern only if a foreign, or one might say, random influence enters our decision-making process. Are we then the ultimate source of our choice?

I like to think of choice in terms of predictability in the same way that random numbers can have predictability from 0.5 (perfectly random: entropy = 1.0) up to 1.0 (totally predictable: entropy = 0). These measures can also be represented in terms of bits of information received. This provides a way to quantify free choice when it is usually entirely subjective.

Let’s say we are driving and we come to a fork going left or right. To the left is town and our workplace – we always go left, though we don’t *have* to. The predictability is near, but not exactly, 1.0. There is always a tiny possibility of going right, in spite of everything. 999 times out of 1000 we go left and 1 time, for perhaps inexplicable reasons, we turn right. The predictability is therefore, 0.999 and with a corresponding entropy of 0.011408 bits of information. In contrast, we drive to an unfamiliar area for an outing. 6 out of 10 turns we go left for no particular reason, though there would be many qualifications to the assumption of “no particular reason”. The average predictability is 0.6 giving entropy of 0.970951 bits of information. For any choice, the bits of entropy can be normalized by dividing by the maximum possible bits of entropy in the choice. This will give a ranking from 0 to 1 representing the degree of “freeness” of the choice.

The bits of entropy can be considered a measure of free choice. These example choices have a maximum of 1 bit because there are only two possible outcomes. Other choices may be more complex involving multiple bits of entropy. Shannon information theory can be used to calculate the number of bits of entropy, and even the rate at which the information is obtained or *received*. The information rate, in bits per second, is (1-*H*)(*bits*/*t*), where *bits* it the total number of possible bits in the choice and *t* is the time in seconds to make the choice.

Clearly from this probabilistic perspective, habitual behavior greatly reduces the degree of free choice. Even though the habitual behavior is in our own interest, there really is no choice being made at all. It is almost entirely deterministic, like a computer algorithm.

This method of measurement can be applied to other ideas of free choice, as long as a proper interpretation is found for representing bits of information.