I suggest that you study fetal development in the womb. However, it would be difficult to define what you mean by "experience". The entire nervous system--peripheral and central--forms a network. When do our first memories begin to take shape? I can't answer such questions, of course, since I don't have any expertise in neuropsychology.
I can't answer it either, I don't even know if it is even known. But let's take a simple example and say that a baby is introduced to a ball, then it is exposed to its functionality it. A ball can jump, if it hits an object it bounces off it. If the baby accidentally makes the ball hit an object so it breaks, it might get told that it is a bad thing.
All these are what I refer to as experiences, these will influence the behavior of the baby later on. It will know that smashing things with a ball is not good. Obviously, at this point, these are extremely simple. But as the baby grows, it can understand the connection that breaking things is bad, even if it is not done with a ball.
So if these are so strong influences on the baby that it shapes how it does things later on, then it isn't really using free will in that sense, it is behaving based on these.
That is a key issue with compatibilism how do you explain the first influence as not having a cause?
I'm not suggesting that robots have free will in the same sense that humans do, and we do have a long way to go before we can build machines with central nervous systems anything like that found in animals. Right now, we are just at the stage of simulating certain types of behaviors and rudimentary learning strategies that allow limited self-programming. For example, walking robots have to have some awareness of their own bodies, because they have to watch where they step and learn to avoid obstacles. When they trip, they have strategies for getting up and back on their feet. What they can't do very well is learn cumulatively from experiences. They don't form and retrieve memories in the same way that animals do--via associative episodic memory.
Agree, that it is a huge issue they are working on, the world we operate in is extremely complex, yet all life learns how to navigate in it fast. Humans are pretty slow at it, but most animals can walk or at least crawl around exceptionally fast.
And clearly, robots lack this, because we are trying to "recreate it" for them, probably because we don't really know how life forms do it. Think about the insane amount of energy that goes into a robot and still it can't even outperform a "simple" life form, that might only be minutes old.
But surely at some point we will crack the code for how to do it, it is already much better than it used to be.
It's interesting that you should use the Mars rover as an example, because they do have a fair bit of autonomy built into them, and I have had a small encounter with one in the past. I was once invited to watch a demonstration of a voice interface in which an astronaut gave commands to examine and retrieve rock samples. At one point, the robot was issued a command to approach a rock, and it just sat there, not responding to repeated commands. It took the team about half an hour to discover what the problem was--a sensor had become blocked with dirt. It wasn't able to see the rock, but it had no way to diagnose its problem and report it to the astronaut. IOW, the team need to build greater self-awareness into the rover to make it function properly in such cases. People, of course, have a much greater sense of self-awareness, self-diagnosis, and strategies for overcoming problems. More importantly, their lengthy childhoods give them ample opportunity to develop their own "programs". Biological brains are analog computers, so they don't actually have anything analogous to programs in digital computers. They continuously rewire themselves physically as they learn.
Agree, but I was thinking that it takes about 5-20 minutes, to send a signal to Mars, which is far too long it the rover is about to run into a hole, so it needs to be able to navigate on its own.
In regards to biological brains, we take a whole lot of shortcuts, we don't really need the exact details, and we can run over a rocky area mostly unharmed, despite not knowing the exact positions of each rock. But we know enough about the rocks to do it effectively most of the time.
We already know that humans have it, because we can never know the future. A choice is free or unobstructed if we have control over our actions.
Even if our future is unknown, if you are convinced by former influences that the possible future that awaits you is the best, even if it is wrong, then you didn't make a free will choice, you could even adjust your choice as you get more and more information, but again these would be influences. And if you are not in control of these influences then it is a valid argument that you don't have free will.
But control is a scalar concept--a matter of degree. We more or less control our movements, so it is not a question of whether we do or don't have free will, but how much control is necessary before we judge it to be free will.
Again it depends on how you look at it. If I throw a rock at you and you dodge it, then the rock is the reason for your movement. If you go get water it's probably because you are thirsty. You don't get thirsty because you go to drink water.
The part where I kicked the ball intentionally, yes. The part where it hit someone in the head, not necessarily. It could have been aimed deliberately or just an accident.
This is where the issue of tracking it all the way back becomes an issue in the deterministic world view, there wasn't an option for you to not hit the person in the head.
Again I don't say I agree, simply that this is their argument.
They do make choices, make guesses about future outcomes, and choose the optimal path to achieve a desired outcome, but they are not yet capable of altering their future behavior to the same extent that biological machines with brains are.
Not yet at least, it depends on whether we get to a point where AI (AGI) can reason just like humans. That is what they are trying to achieve.