Philip Jose Farmer’s Riverworld series (which I’ve mentioned before, upon rereading the first volume) begins with every human who ever died waking up on an artificial planet, resurrected as young adults (unless they died younger). Fashions in scientifictional resurrection have changed since 1971, so here’s how I might do it, given a boundless supply of handwavium.

I’d start by using my time-viewer to make a scan of your brain from each time you slept; because, even if you never had traumatic brain damage, I’m not sure that the last version of you is your favorite. I then build a composite from these scans, by attaching each neuron not to other neurons in the same snapshot but (wherever possible) to the corresponding neurons in the two adjacent versions.

At first you literally can’t hear yourself think, because the internal monologue consists of many overlapping whispers. I’ll probably have to suppress fear.

The real difficulty is output, when thousands of versions of you are trying to control the same (simulated) body. One possible answer is thousands of bodies, initially occupying the same space, intangible to each other (but tangible to everything else), each with a tiny density and strength. Bodies close together in the grand sequence will tend to behave in unison; but “you” will probably split into several sub-sequences separately converging, though the flow of thoughts along the chain is never completely broken.

As you get settled in your new brain, you weed out many redundancies but also make some new long-range links. And that raises another interesting question: how important is it that our brains are divided into lobes? Learning consists of making and breaking links between neurons; would it be good or bad to take away the physical constraints on such links? (Maybe that’s why the robot version of JW, in Ken MacLeod’s The Stone Canal, is told that his mind has changed too much to be put back into an organic brain.)