What role does memory robots play in the many worlds interpretation?

Question

Ron Maimon gave a very interesting answer to a question about the many worlds interpretation here. He mentions the essential role of memory robots in MWI, which is something I have never ever heard any other MWI proponent mention at all. Are memory robots — whatever they are — absolutely necessary to make sense of MWI? If so, how come no one else talks about them? Do they have anything to do with Information Gathering and Utilizing Systems (IGUSes) in consistent histories?

Answer 1

The memory robots are introduced in Everett's PhD thesis and published paper, and they are the way in which Everett defines the splitting of the state into separate components. The measured state of a system is relative to the memory state of the robot, which defines the preferred basis. The basic hypothesis is that the world is quantum and superposed, but you don't "feel" superposed, because how the heck would you know what a superposition is supposed to "feel" like? What possible aspect of your experience could lead you to consider the idea that you are in a superposition impossible?

The memory robots are maps from classical memory into quantum mechanical states. The memories are classical information, they can't be superposed, and whenever they end up superposed, that's a split in the robot. The state each copy of the robot sees is the state relative to this memory history.

I don't know why you say nobody talks about this. Everett talks about this. Gell-Mann talks about this. Many people consider many-worlds philosophy of mind rather than physics (and I tend to agree, although the decoherence aspects are physics). In popular accounts, people tend misinterpret Everett as an objective splitting, rather than a subjective collapse (except not through "gaining information", but by identifying your branch of the superposition). Everett was completely clear: he thought of it as a subjective collapse relative to the state of a memory robot. They appear on page 1 or 2 of his paper and thesis.

There are things like "many minds" you read about which simply make the memory robots more explicit, and these are sometimes represented as distinct interpretations. They are really phases of the following cycle: Everett->Misinterpretation of Everett-> rediscovery of Everett-> misinterpretation of the rediscovery. The internet put a stop to that nonsense more than a decade ago, in the mid-1990s.

Everett's robots are the same thing as the IGUSes in consistent histories. Consistent Histories is just an attempt to repackage Everett, given what a big flop the original presentation turned out to be. The repackaging extends the ideas nontrivially and rejiggers the philosophy in ways a positivist wouldn't care about.

It seems clear to me that Pauli and Heisenberg had some of Everett in mind, perhaps there is an off chance that Bohr did too (although this is much less likely). The positivism tends to disguise this, since they just formulated the theory in the most palatable positivist form, as a prescription for calculating the results of experiments. This practice continues to this day, and it is laudable because it keeps positivism alive, and ultimately, this is the correct way to formulate theories. Privately, you see Heisenberg and Pauli convinced that the collapse is a property of mind, not a property of matter. With Everett, you can see where the "mind" business is coming in--- it's just that the robot can't internally feel superposed, so when it is superposed, it makes a choice of which branch to follow.

This idea has popped up in philosophy, but nobody likes to cite Everett. In philosophy, it is sometimes attributed to Daniel Dennett. It is compatible with viewing the mind as a computer program and the brain as a computer, an obvious point of view persuasively and irrefutably argued by Turing and one which Everett held (although in the 1950s it wasn't controversial). This idea, to humanity's eternal shame, is considered once again controversial today.

Answer 2

The point of the memory robots was to provide a quantum mechanical model of observers rather than treating observations as an exception to quantum mechanical equations of motion. A quote from Section 5 of Everett's original “Relative State” Formulation of Quantum Mechanics:

We have the task of making deductions about the appearance of phenomena to observers which are considered as purely physical systems and are treated within the theory. To accomplish this it is necessary to identify some present properties of such an observer with features of the past experience of the observer. Thus, in order to say that an observer 0 has observed the event α, it is necessary that the state of 0 has become changed from its former state to a new state which is dependent upon α.

It will suffice for our purposes to consider the observers to possess memo- ries (i.e., parts of a relatively permanent nature whose states are in correspondence with past experience of the observers). In order to make deductions about the past experience of an observer it is sufficient to deduce the present contents of the memory as it appears within the mathematical model.

As models for observers we can, if we wish, consider automatically func- tioning machines, possessing sensory apparatus and coupled to recording devices capable of registering past sensory data and machine configurations. We can further suppose that the machine is so constructed that its present actions shall be determined not only by its present sensory data, but by the contents of its memory as well. Such a machine will then be capable of performing a sequence of observations (measurements), and furthermore of deciding upon its future experiments on the basis of past results. If we consider that current sensory data, as well as machine configuration, is immediately recorded in the memory, then the actions of the machine at a given instant can be regarded as a function of the memory contents only, and all relavant experience of the machine is contained in the memory.

The motivation for IGUSes is similar, see Section VII of Quantum Pasts and the Utility of History by Hartle.