The way people talk about it makes it sound indistinguishable from "random will". If you believe in the existence of a "self" in any form, be it the chemical signals and electrical impulses in your material brain, or a ghost existing outside of space and time controlling your body like a puppeteer, you must believe in one of you believe in that self having free will.
Say you were to run a scenario many times on the same person, perfectly resetting every single measurable thing including that person's memory. If you observe them doing the same thing each time then they don't have this quality of free will? But if you do different things each time are you really "yourself"? How are your choices changed in a way that preserves an idea of a "self" and isn't just a dice roll? Doesn't that put an idea of free will in contradiction with itself?
Edit: I found this article that says what I was trying to say in much gooder words
I like this question because it asks how is 'free will' defined exactly. Depending on how that is done, one might argue that sociopaths have more 'free will' than others due to social context. I think this is pretty useful to force philosophy to have a more social context with this stuff.
It's similar to the contrast between "liberty to X" vs "liberty from X". Defining 'free will' only in the context of individualism doesn't seem to be very helpful, but I've not seen much of this discussed around this topic. It's mostly tug-of-war BS between philosophy, religion, neuroscience. We need more social sciences in there.
Sartre defines free will as being our ability (or our curse) to choose and to be responsible for those choices. Furthermore, our freedom is dependant upon the freedom of others, so in this context sociopaths who encroach upon the freedom of others in turn reduce their own freedom. At least in existential, Levinasian, or Derridean philosophies, freedom is always defined in relation to others and the responsibility we have for them and ourselves.