Is it possible for the arrow of time to be observed in reverse?

[Octonion]

I know that in general the arrow of time does not itself reverse but is it possible for there to exist 'observers' in our universe that see its progression from, what we think as, finish to start?

I just read http://arxiv.org/pdf/hep-th/0611088v3" by Brett McInnes so I understand why the arrow of time is 'forward' but I can't seem to grasp how this rules out an observer that perceives time in reverse from existing or explaining our same universe in reverse.

 

[dimensionless]

If time went backwards, how would our brains work? How could we remember what had just happened in the 'future?'

 

[atyy]

predicting the future = remembering the future

That you remember the future less accurately than the past is the second law of thermodynamics.

Feynman: The Distinction of Past and Future
http://research.microsoft.com/apps/...data=4|d88d1dbd-a736-4c3f-b832-2b0df62e4eca||

 

[CyberShot]

If time went backwards, how would our brains work? How could we remember what had just happened in the 'future?'

In my opinion, it would be no different from the ordinary flow of time. The "arrow of time" is something that is not absolute (so long as you are consistent with your definitions); rather it is arbitrary, much like the choice of axes in a physics problem.

 

[Fra]

This is how I see it.

If we consider that and observers expectations of a arbitrary CHANGE, relative to whatever is the current state. Then there is typically always an expected peak evolution direction; this is then definition "forward". This is as much observer dependent as is the expectations.

But if the a priori "expected evolution" in retrospect proves to be violated, then obviously the expectations are revised so that the arrow of time always adjust. If not, it would mean that the observer "fails to learn" and such an observer would destabilise itself.

> If time went backwards, how would our brains work? How could we remember what had just happened in the 'future?'

I think by construction, the future always refers to the "expected" change. You need no global or predefine "arrorw of time". A natural direction is always implicit in the uncertainty of an information state; it's the direction in where you are doing the random walk. Backwards time would mean that your expectations were MAXIMALLY bad, and an observer can't maintain such expectations without beeing destructed because every single new information would be maximally destructive during "information updates".

Edit: I'd argue that this the OT is in face analogous to, "is it possible for an observer to maintain maximally poor inferences without beeing physically destabilized by it's environement". I.e would an information processing and gambling agent be able to completely fail to learn and still exist in some equilibrium? I think not. However transient deviations happen all the time. So a transient or temporary violation of expectations is just part of the game, just like the 2nd law is "only" statistical.

/Fredrik

 

[Fra]

It's quite interesting also talking about the brain, that as far as I know, a lot of current brain research indicates that the way the brain actually remembers the past (stores information about it's own history) is done in a way, that is optimized for fitness in an expected future. This is also why, quite often, "actual memories" are distored, or technically "wrong" in details. But there is an explanation for this, it's because the reason for remembering the past is not book keeping, it's a survivla value for hte future. And remembering the EXACT historical sequences and detail are not always optimal. And sometimes even a quite "deformed" to the point of incorrect remembering of history may be or survival value. From the point of view of the brain this still is not an error as there is a rational objective behind it.

Some savants do have remarkable and almost supernatural abaility to in face remember actual detailed histories - beyond that point that seem "relevant". And this is indeed considered a defect. A normal brain doesn't do this.

/Fredrik

 

[Demystifier]

If there are two systems with opposite directions of time, their interaction will cause their arrows to take the same direction:
http://xxx.lanl.gov/abs/1011.4173

 

[Fra]

If there are two systems with opposite directions of time, their interaction will cause their arrows to take the same direction:
http://xxx.lanl.gov/abs/1011.4173

Thanks for the link Demystifier, I'll skim it later.

Indeed that is an excellent way of putting it! and it sounds exactly how I think of it.

In terms of interacting inference systems, an "interaction" between two superficially disagreeing systems can be seen exactly as a negotiation process. The logic of a physical interaction, can there be understood as the logic of rational negotiation between two information processing agents.

One could even conjecture (like I do) that in this there is the origin of gravity as well. Because ANY two interacting systems will face an attractive force, regardless of the details of the systems. It only depends on system complexity acting as an effective inertia in the negotiation process. This has the exact same universal traits as does gravity. What is missing though, is to understand how 4D space emerges in such negotiation process and thus how the generic information divergence measures somehow in some classical limit allows emergence of a classical distance measure and metric.

/Fredrik