Here's another thought experiment that philosophers sometimes appeal to in this context. Suppose someone invents a teleportation machine (like in Star Trek). The machine scans your body, vaporizes it, and then recreates a molecule for molecule duplicate somewhere on Mars. Would you survive this process? That is, would the person on Mars be the same person who stepped into the machine on earth? Or would you cease to exist, only to be replaced on Mars by someone who is exactly like you? If personal identity is just a matter of physical continuity, then you probably don't want mess around with teleportation.

It isn't entirely obvious that I, say, could have been female from conception, and the assumption that the fertilizing sperm was different certainly leaves that possibility open. But if I could have been female, then your great-great-great grandfather (let's say) could have been female, and one of your great-great grandwhatever's parents would have had a hard time conceiving a child together.

Maybe that isn't the sort of possibility you had in mind. But it's not obvious how to restrict it and still get plausible results. Let's suppose you could have been the result of fertilization by a different sperm. What's so special about the sperm? Why not a different ovum, too? But now consider that other ovum and sperm. The latter could have fertilized the former even if the ovum and sperm from which you were actually formed still got together. But then are you your own twin? I don't think so. So it doesn't look as if you could have been the product of a different ovum and a different sperm. But if not, then why should only one of them be required? It's not obvious how to answer this question.

Our atomic structures don't remain the same over time. Some changes in structure sustain personal identity. Others don't, such as death. And some would argue that certain types of changes in neural structure would result in person a being replaced by a different person, b, as owner of the same body (say if b's neural structure were copied in a's brain).

This is one of the more contentious and continuing questions that I've encountered in philosophy...and the way you put it makes perfect sense. So you are likely to get a great variety of answers.

I like J. David Velleman's account of triadic, reflexive selfhood: he argues that the "self" has different meanings depending on the question it is used to answer: questions about metaphysical persistence (how is the entity I call my "self" the same thing now as it was in the past and will be in the future, which I take to be the focal point of your question); about psychological self-regard (when I think about my "self," what is it I'm thinking about?); and about the generation of autonomous action (how does a particular action have its source in my "self" as opposed to some outside force or influence?). You can find this account in Contours of Agency: Essays on Themes from Harry Frankfurt, ed. Sarah Buss (Cambridge, MA: MIT Press, 2002), pp. 91-123, and criticism that it is overly inclusive, granting selfhood to things like robots, in Diana Tietjen Meyers' “Who’s there? Selfhood, self-regard, and social relations,” Hypatia 20 (2005), pp. 200-215.

My own answer is that all three of Velleman's meanings can be brought together by thinking of the self as an autobiographical narrator whose story extends backward and forward in time, who creates a self-image as protagonist of the story and who authors the protagonist's actions. It might be said that the "self" is just what a given self, or autobiographical narrator, perceives itself to be from its own first-person perspective. Certainly a self is capable of transformation without ceasing to be the same self...but how much? I'd say the changes wrought by your various dramas and experiences are part of the autobiographical narrative of a single self. But there are many philosophers who would disagree, and have very good reasons for doing so.

I think most philosophers nowadays would say, no: If your mother had never gotten pregnant, then you simply would not have existed. It's not a pleasant thought, at least not for you, but there you have it.

I can well imagine that there could be such evidence. But for the evidence to be truly trustworthy, it would have to be collected by people who were neutral, more or less, on what it was supposed to demonstrate, and the evidence would have to be in some sense replicable, and to stand up to critical scrutiny by reasonably neutral parties. So far as I'm aware, there is no evidence for reincarnation that meets anything like this sort of standard, however.

Pinning down precisely wherein personal identity consists is certainly a thorny problem in philosophy, which has been debated for centuries and still seems quite far from a definitive solution. I can't promise to solve it for you, but here are a few considerations that you might find it interesting to ponder:

1. Let's look at the way you set up your question: "As you gain knowledge and change your opinion, your personality changes and you seem to be totally different then you were before." You're using the same pronoun throughout, which is already enough to imply that there is just one enduring thing here to which this pronoun continues to refer. If you really believed that the things which existed in these different times really were distinct, wouldn't you refer to them in different ways? Moreover, you're attributing to change to something, but that too seems to imply an enduring identity. The very notion of change, the notion that a thing is now different from how it used to be, suggests that the same thing existed then as exists now, but merely happens to have different properties in these two different times. This, in turn, implies that the 'self' cannot simply be equated with the bundle of properties that it possesses, and hence that these can come and go without any change in the identity of the self. But perhaps you might feel that it's a mistake to put too much weight on the accidents and imperfections of language, and don't feel that we can draw any valid metaphysical conclusions from the way we happen to use pronouns and so forth. In which case, then consider this:

2. How do you propose to 'live completely in the present'? It's easy enough (though imprudent) to disregard the long-term consequences of one's behaviour: but, no matter how self-indulgent and dissolute one's behaviour might be, one does still have to wait some time to reap the benefits. I'm glad that you added "... or even 5 minutes ago" to the way you set up the question, because this is a very important point. If we push your argument to its logical conclusion, you won't have any reason to care about what experiences 'you' (or, to avoid begging the question, someone very much like you) are going to have in five minutes from now, or even in just a few seconds. Suppose you were to decide: "I don't care about the future; I don't care whether I'm going to be living on the streets in a year's time, or hungover tomorrow; I'm just going to get blind drunk tonight, because tonight is all I care about." But your argument would suggest that the person (or people) who exists through the remainder of this evening is not the same person as you: so why would you care about the pleasures they are going to have in the near future, any more than you care about the pains they're going to incur in the further future? All that you can do right now is lift the glass to your lips. It takes time for the alcohol to get into your stomach, and then into your bloodstream, and finally into your brain: but, by then, and by your logic, it will no longer be your brain. So the argument in favour of disregarding negative consequences to yourself seems to speak equally against seeking positive benefits for yourself.

3. But let's suppose that the person who exists in a year, or in five minutes from now, really is a different person from you, as different from you as the person you see on the other side of the room right now. Don't we still have responsibilities towards other people? If you were to do something purely self-serving, without giving any regard to the negative impact that your action was having on the people currently around you, most people would say that you were morally at fault. Why, then, shouldn't we also say the same thing about future individuals? Even if we allow that they are not literally identical with you, shouldn't you still give them the same regard as you would to any other distinct person?

The Law of Identity states that each object is identical to itself -- hard to deny. "Daniel is identical to Daniel" is a particular instance of that Law.

Your clone is not identical to you: if you and your clone we're alone in a room and we counted the number of objects in the room, we'd get two, not one.

"Daniel is identical to Daniel" does not express that the word to the left of "is identical to" is the same word as the word to the right of it: it expresses that the object the first word refers to is the same as the object the second one refers to. This can be made plainer by considering, for instance, this claim: "Daniel Defoe is identical to the author of Robinson Crusoe." This is a true identity claim, even though the words "Daniel Defoe" and "the author of Robinson Crusoe" are not themselves identical.

It would be difficult to say why the Law of Identity is true. Any defense of it would either involve using words like "equals," "same as," etc. all over again, or would use words that are themselves less clear than the word "identical." For that reason, I wouldn't understand anyone who questioned the Law. If someone said that Bob Dylan wasn't identical to Robert Zimmerman, I'd simply assume he wasn't using "identical" to express the same relation mentioned in the Law of identity.

Suppose there were a homunculus. Would it be like me? That is, would it have conflicting motives? Foggy beliefs? Occasional weakness of will? And while we're at it, would it make any difference if the homunculus were located in one compact region of the brain? Or woud it do just as well if it were distributed over different parts of the brain, and perhaps not even clearly confined to the brain alone? What would the homunculus have to be like to do the intellectual job that's at issue? And do we really need a lot of science to know that whatever we are, we aren't simple unities?

An utter disunity isn't an agent. But think about the difference between my academic department and a random collection of professors. My department is made up of diverse individuals who don't always agree. But the department has a plan of organization, it deliberates and it acts. The members of the department co-operate to get things done, and the dissenters accept decisions of the department, once they're made, even if they aren't pleased about them. We can't say anything like this about a random collection of professors. For many purposes, my department is an agent, and it can be held responsible for what it does.

Christine Korsgaard (See her "Personal Identity and the Unity of Agency" in Philosophy and Public Affairs 18:2, 1989) argues that the unity of an agent isn't metaphysical; it's practical. We count as unified agents because we actually do manage to get past the conflicts among our motives and act one way rather than another, and because we can look at our actions from a unified standpoint. That includes things like acting on the basis of reasons and principles. This seems right quite apart from whether there's a homunculus hidden somewhere within us.

I'd add that part of what you say already gestures toward this picture. You talk about "complex living systems." We are, indeed, such things. A system isn't just a haphazard collection of processes. Systems are self-sustaining complexes that, as you point out, are more than the sums of their parts. But why isn't that good enough? Why can't the right sort of system be an agent? The homunculus wouldn't really add to the story. Indeed, as my comments above suggest, even if there were a homuculus, we might well have to tell a systems story about it anyway.

Let's call the being that results from all this replacement Sarah2. We can ask a pair of questions that seem different. One is whether Sarah2's conscious states will be like Sarah's. I agree with Mark that the answer to that question is yes; at least, it's hard to see why it would be no. But we can ask another question that seems to a different one: is Sarah2 the same person as Sarah? That's a lot more controversial.

A comparison, based on an example by Peter van Inwagen: Suppose little Johnny builds a house from a small number of blocks and leaves it in the middle of the floor. And suppose that I come in and clumsily kick the house over. If I re-arrange the blocks in exactly the same way, then the house I assemble will be indistinguishable from the one Johnny built, but it's not so clear that it's literally the same house. And if I actually replace the blocks with new ones that are just like the old ones, then it's even less clear.

So if we cryo-freeze Sarah, interrupting her normal biological and psychological processes, and then perform this massive replacement, there's at least room to wonder whether it's literally the same person. Sarah2 will no doubt think she's Sarah, but she could be wrong for all that.

This is part of a big debate, of course. One good collection that provides a wide range of background readings with a nice historical introduction is Raymond Martin and John Barresi's anthology Personal Identity, published by Blackwell.