Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Your memories are almost certainly false

  1. Apr 25, 2010 #1
    It’s overwhelmingly probable that all of your memories are fake.


    Entropy is a measure of the disorder of a system. The higher the entropy, the higher the disorder.

    If a deck of cards is ordered by suit and then within each suit by ascending rank, then that’s a low entropy state. This is because out of the 8.06 * 10 to the 67th (52!) possible unique arrangements of the cards in a standard 52 card deck, there’s only 24 that fit that particular description.

    A “random looking” arrangement of the deck is a high entropy state, because there are trillions of unique arrangements of a standard 52 card deck that will fit the description of looking “randomly shuffled”.

    Same with the egg. There are (relatively) few ways to arrange the molecules of an egg that will result in it looking unbroken, compared to the huge number of ways that will result in it looking broken. SO, unbroken egg…low entropy. Broken egg…high entropy.

    AND the same with the universe…there are (again, relatively) few ways to arrange the atoms of the universe in a way that makes it resemble what we see with people and trees and planets and stars and galaxies, compared with the gargantuan number of ways to arrange things so that it resembles a generic looking cloud of dust.

    OKAY. Now.

    Of the relatively few ways that the elementary particles of the universe can be arranged so as to resemble what we see around us today, only a tiny fraction of those particle arrangements will have values for momentum and position that are consistent with them having arrived at that state 13.7 billion years after something like the Big Bang.

    The vast majority of the particle arrangements that macroscopically resemble the world around us will *instead* have particles in states (e.g., with positions and velocities) that are consistent with the particles having previously been in something more like a giant dust cloud.

    By which I mean: If we take their current positions and velocities, and work backwards to see where they came from, and go back far enough in time, eventually we will not arrive at the Big Bang. Instead we will arrive at a state resembling a giant dust cloud (probably a very thin, spread-out dust cloud).

    SO, bottom line:

    Out of all the possible configurations that the universe could be in, ones that have people, and planets, and stars, and galaxies are extremely rare.

    Further, even if we then only consider those extremely rare possible configurations that have people, and planets, and stars, and galaxies – the ones with particles in states (e.g., with positions and velocities) that are consistent with having arrived at this configuration 13.7 billion years after something like the Big Bang are STILL rare.

    We don’t know the exact state of our universe’s particles, but in statistical mechanics the Principle of Indifference requires us to consider all possible microscopic states that are consistent with our current macroscopic state equally likely.

    So given all of the above, and our current knowledge of the laws of physics, the most likely explanation is that all of your current memories are false and that yesterday the universe was in a HIGHER state of entropy, not a lower state (as would be required by any variation of the Big Bang theory).

    Physical systems with low states of entropy are very rare, by definition. So it’s very improbable (but not impossible) that the unlikely low entropy state of the universe of today is the result of having evolved from an EVEN MORE UNLIKELY lower entropy universe that existed yesterday.

    Instead, statistically it’s overwhelmingly more probable that the unlikely low entropy state of the universe today is the result of a random fluctuation from a HIGHER entropy universe that existed yesterday.

    And thus your memories of a lower entropy yesterday are most likely due to this random fluctuation, not due to yesterday actually having had a lower entropy than today.
  2. jcsd
  3. Apr 25, 2010 #2
    ok, that's quite an idea... but doesn't every spontaneous action increase the entropy of the universe... For the universe to become less random, more ordered, it would have to be increasing in energy... Since thermodynamics tells us that energy cannot be created or destroyed, I'm going to have to say that the universe is becoming more and more disordered, presumably from a very ordered point, to eventually a thin dust cloud..
  4. Apr 25, 2010 #3
    Two wikipedia articles of interest:



    Also check out the references on the first article...
  5. Apr 25, 2010 #4


    User Avatar
    Gold Member

    Except brains and bodies are dissipative structures that pay for their order by their more rapid degradation of entropy gradients.

    The second law of thermodynamics has more complex consequences than those you are using in your argument here.
  6. Apr 25, 2010 #5

    Did he say human bodies violated the 2LOT? What does the fact that human bodies obey the second LOT have to do with how likely it is for something similar to a universe and a conscious human being with memories to emerge out of a quantum 'soup'?

    Again, this does not prove his idea is wrong. As far as i can see, RexAllen is trying to evaluate how likely it is for anything to emerge in a state that we might call a Big Bang and enfold to its current state. And as far as i can see his point, even the 2LOT is a statistical occurence that happens to make sense to us because if it had not, we wouldn't be here and definitely not made it that far development-wise.
  7. Apr 25, 2010 #6
    Anthropic principal, only in a universe where an intelligent entity exist is it possible for there to be questions and ideas such as these.

    Thermodynamics, I can take energy from a to make b more ordered, though in the end, the total disorder in a AND b after the transaction will be greater than before the transaction. To put it in words of someone who almost discovered the sun...
  8. Apr 25, 2010 #7
    Nothing in the original post contradicts the second law of thermodynamics. Read the wikipedia links and associated references.

    This idea isn't original to me. In fact, it's not even especially obscure...note the article in the New York Times listed in the references to the first wikipedia link.
  9. Apr 25, 2010 #8
    It's a well known problem and I mentioned it in the last thread you posted. The answer is that if all our memories are false then so is the second law of thermodynamics - it undermines itself. That's why we need to assume the universe started with incredibly low entropy.
  10. Apr 25, 2010 #9


    User Avatar
    Science Advisor

    In the case of memories, they are "false" for entirely different reasons than this. We are remarkable incapable of maintaining an accurate picture/description of a situation over time. Our memories are also changing over time, even new elements can be included.

    Taken in the general perspective, that something cannot be a "true" description/representation of something else over time, I will question this by asking what is a "true" description anyway?
  11. Apr 25, 2010 #10
    Well thats a different discussion entirely, more of a psychology argument... If people remember something different than what actually happened, did it actually happen? Who is to say that events are not defined by perception?
  12. Apr 25, 2010 #11
    Well my memorys are not false... And I personally wouldn't label the universe in terms of order/disorder because one mans order is the other ones disorder. As I understand it for the universe to be going into a more entropic state things would have to be getting more complex and well... they are imo.
  13. Apr 25, 2010 #12


    User Avatar
    Science Advisor

    I'd say it's more of a question of how our language and other forms of expressions relates to the world.
  14. Apr 25, 2010 #13
    If I might can I ask these questions of you all?
    How do you define order?
    How do you define chaos?
  15. Apr 25, 2010 #14


    User Avatar
    Gold Member

    Thanks, I'm familiar with the Boltzmann Brain argument. And the fact that it seems possible to extract ludicrous consequences from a premise is normally evidence there is something wrong with the premise.

    In this case, it is that the second law is being used too simplistically. Entropy degrading systems in fact have high probability of occuring where there is an entropy gradient.

    So if you truly want to calculate anthropic probabilities, you have to go back to the question of how such cosmological-scale gradients might arise.

    Even your "dust cloud" universe would be still expanding, still cooling. It would still be a gradient (even if a much weaker, duller heat death one).

    When thinking about the cosmological scale, you need to include both the global scale (the void, the vacuum, the de sitter event horizon) and the local (the dust, the atoms, the background radiation) in the calculations.

    The void is often treated as if it is "just nothing" and so can be left out of entropy maths. Yet clearly it is a very orderly and continually developing "something".

    Unless you take a completely solipsistic line with Boltzmann brain, saying that even all cosmological facts are the imaginings of a mind that fluctuates into momentary being. But then how would any entropic arguments be of interest if everything has been faked in this (now inexplicable) way?
  16. Apr 25, 2010 #15
    Sean Carroll's From Eternity to Here has a good discussion of these issues.

    Particularly this chapter:


  17. Apr 25, 2010 #16
    I can be rather dense. How did you get from this,
    to this?
  18. Apr 25, 2010 #17

    It might make more sense if you put this between the two paragraphs that you quoted:
  19. Apr 25, 2010 #18

    Char. Limit

    User Avatar
    Gold Member

    Well, I'm no philosopher, but order to me is having fewer possible microstates, and chaos is having more possible microstates.

    [tex]S=k \times ln{W}[/tex] FTW!
  20. Apr 25, 2010 #19
    OK. Got it. Not so very different from stray thoughts of my own, but in very different words. Say we are willing to put aside, for the moment, the concept of linear time and replace it with the concept of multiple pasts and multiple futures, what would we cognitively record in memory? It should be safe to say that the process of cognition, and therefore memory storage itself is a process that involves an increase in entropy from information loss.
  21. Apr 26, 2010 #20
    Aside from epistemological issues there is a known phenomenon of degradation of information storage in the brain. Our minds do not like gaps though so when we do not clearly remember something we often unconsciously mend the gaps in the degraded memory by patching it with what ever information seems most likely or appropriate there by altering the stored memory.

    edit: this is actually more along the lines of what I thought the thread was going to be about when I first clicked it.
  22. Apr 26, 2010 #21


    User Avatar
    Gold Member

    My question is still how do you get thermal fluctuations without a thermal gradient?

    And if we must presume a gradient, then the second law states it is probable that the past of this gradient was more orderly, and the future will be more disorderly.

    If our argument must invoke an eternity of time passing, then the universe must be asymptotically close to heat death. While the passage of time would seem to increase the changes of something very unlikely happening, it also increases the odds that it can't happen because there is less energy, less material, to generate candidate fluctuations.

    The curve of dissipation being an exponential, the time it spends in the fluctuation generation realm is in fact vanishingly short compared to the time in which it is too heat dead to do anything of note.

    All this is leaving aside the further issue of how any kind of arrangement of dust particles would mimic the complexity of a biological brain. What is this dust composed of? A suitable mix of atomic elements? Carbon, nitrogen and magnesium, etc.

    And is this dust world then a realm without supernova and so complex atoms also conveniently fluctuate into being via thermal fluctuations (rather extreme ones if the dust is hydrogen, helium and lithium - the heat released by the fusion needed to assemble the brain would surely fry it before it happened).
  23. Apr 26, 2010 #22
    The problem as I understand it, is that the law of increasing entropy works the same backwards and forwards in time, i.e. the disorder should increase backwards in time. This leads to the conclusion that the universe just appeared 1 second ago as a thermal fluctuation and all the dinosaur fossils and memories etc. are false. So the concept of heat death is inapplicable since the universe was always in that state, there is no asymptote.

    This is why we assume a low entropy beginning - it circumvents all of these problems. And as I mentioned previously, the conclusion that all our memories are false undermines the argument we used to get there, since we invoked scientific laws which rely on the validity of our experience and memory.
  24. Apr 26, 2010 #23

    I believe this is wrong. Can you explain why you think so? The 2nd LOT is still a great challenge to the idea that time does not flow, i.e. the thermodynamic arrow of time(low entropy to high entropy) is still valid, though there are hints from GR that the flow of time is not a fundamental ingredient of reality or even an illusion. This doesn't mean however, that "entropy works the same backwards and forwards in time", but just that the universe is in a sense "frozen" in one state that encompasses all possible states at once. Einstein had trouble understanding this weird consequesnce of his relativity and today it's still as incomprehensible as ever.

    This could be the case but for different reasons, and your premise seems faulty to me.

    The second law of thermodynamics is almost a fundamental law in physics(not that there exists such a law). If you want to circumvent it, you have to explain how the universe is always in one state and the 2nd law is applicable.

    If all our memories(or should i say "my memories"?) are wrong, we can't have access to truths. In that case, at a fundamental level everything that is known can be wrong.
    Last edited: Apr 26, 2010
  25. Apr 26, 2010 #24


    User Avatar
    Gold Member

    The essence of it is that at the micro-causal level of description - such as when tracking the individual motions and collisions of a collection of independently moving gas particles - things seem time symmetric. Physical descriptions - whether Newtonian, relativistic or even QM - are mechanical and make equal sense whether read forwards or backwards.

    The second law then arose as a macro-causal level description of nature. It captured the truth that the universe unwinds. It cools and expands. Work always results in waste heat. Time rolls everywhere forward. So at the macro-level, the global scale, nature is patently asymmetric.

    Now science wants to reduce all macro-descriptions to micro-descriptions. Boltzmann is celebrated for his microstates ensemble model which treated the second law as a probabilistic statement and allowed an apparent reduction to a local, mechanical, time symmetric, description of what was really going on.

    As a model, it is very good. But arguably incomplete because of what it left out to make a useful simplification of nature. And what it clearly leaves out is the question of the "shape of the container". When a collection of particles are doing their thing in a flask, someone has created the flask (a static box which reflects all the particles back on themselves, and the heat sink that allows the particles to cool to some max ent condition, yet not cool completely to the point they cease to move and become something else, like a bose condensate).

    It indeed leaves out the macro-causal factors (the flask, the heat sink, the person who originally confined the particles). These initial conditions and boundary conditions are not modelled. Instead, the macro-view is only of some emergent global-scale property - a measurement of temperature, pressure, or maximum entropy.

    When things get left out - especially the global causal factors, the downwards acting constraints - paradoxes of thinking must arise as a purely local, mechanical, micro-causal description is stretched to its limit.

    Boltzmann's statistical mechanics was faced with the problem of, well, who ordered the universe's initially ordered state? A micro-causal loophole seemed to be that the time reversibility of the local mechanics means that anything can happen in the "wrong direction" as a fluctuation in apparent defiance of the second law. As long as the excursions did not exceed the limits set by the probabilistic descriptions at the emergent macro-level, then both the macro and micro view seemed in harmony. You had perfect symmetry as apparently demanded by mechanical equations of motion and action, and perfect asymmetry as encoded by the second law.

    The error of thinking is then to try to do away completely with the global asymmetry in physical description. We see a world unwinding down an entropic gradient by expanding and cooling. But perhaps this is just a really big fluctuation in some greater equilibrium which has a timeless symmetry as its now global true state. So the second law, as a global statement, becomes reduced to merely a local law applying to some limited region of lucky fluctuation. And micro-causality or uber-reductionism wins.

    Boltzmann's brain is a recent thought experiment to dramatise this possibility.

    But there are plenty of people working on better ways of thinking about the second law and entropy issues.

    Personally, I favour a model based on vagueness and its dichotomisation (or max symmetry => max asymmetry).

    But generally, it pays not to confuse successful mechanical models with what might be actually true about reality. Better models - in the sense of more complete - await development.
  26. Apr 26, 2010 #25


    User Avatar
    Science Advisor

    I would ask OP, what constitutes a "true" memory?
    Last edited: Apr 27, 2010
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook