6 ms·

To Make Sense of the Present, Brains May Predict the Future

71 points by tdurden 8y ago

bobosha 8y ago
Another theory along these lines is by Jeff Hawkins of Numenta in his book "On Intelligence". He called in the Memory prediction framework.
- ilaksh 8y ago
  I was going to mention this https://arxiv.org/abs/1609.03971 https://arxiv.org/abs/1609.03971 which was partially inspired by Hawkins' work.
s-shellfish 8y ago
Calling it future prediction is problematic. We can't predict the future. It's a controlled delusion and sometimes some people will try to override your own reasoning system, and you honestly don't want to give them that power in exchange for it. It's called inferencing. Sometimes we have so much information we don't understand why things make sense in the order they do. It's not future prediction. It's confusion between information that is relatively predictable because it's stable information with very little variation, plus confirmation bias of being correct about highly variant information we are not accustomed to being right about in a particular order. Do not call it future prediction. Sometimes you will be correct about things before things happen. That's just chance. If it can happen, it can happen, it's in that space. But it's fundamentally flawed reasoning. There's still a whole open space of things yet to be experienced. That's what existence is defined as. You don't know what you know until you know what you know. It's not magic. It's called experiencing new things. Calling it future prediction primes you into bias. Which closes you off from opportunities eventually, because you literally stop being able to see things for how they are truly. Pattern matching brain. Being able to comprehend reality is absolutely vital to survival. Don't get lost in the waves of chaos. People can and do try to take control of confusion.
- timb07 8y ago
  > We can't predict the future. What tosh. Of course we can predict the future... and sometimes our prediction is correct. Data compression works by modelling the data stream and predicting the next symbol based on that model, and then recording in an efficient way the degree to which the prediction was correct. The comparison between the prediction and the actual outcome is the key point -- just as it is in the article.
  - s-shellfish 8y ago
    Eventually the system you use to model information with will override actual data. That's a feedback loop of confirmation bias. It's thinking in terms of time. How many models of data do you interact with versus how much actual data are you interacting with. I'm not on a soapbox without a point. It's a fine line. Your own mind matters in these systems. It's important to not undervalue it. It's not a computer. It's not a better computer. It's sometimes flawed. I'm fine with calling it probabalistic inferencing. It's not future prediction, especially when the models one is reasoning about are models of cognition. Thank you for not down voting though, that's appreciated.
- saas_sam 8y ago
  This: "because you literally stop being able to see things for how they are truly" is flawed. You never see things for how they are truly. You see them through a filter of your sensory capabilities, your mental organization, and, as this article discusses, your aptitude for prediction.
  - s-shellfish 8y ago
    If you only see the world through models of information, that is your perception. That doesn't negate the fact that there's measurable, testable reality that can be reproduced using scientific rigor. It's just a perspective from someone who has experienced being carried away with the waves. Aptitude for prediction helps you and tells you it's all working as expected until it slams you with hard consequences that leave enough impact to know where to draw your own line. My own father told me these things. I didn't listen. Guess what happened? It's arrogant to assume people can't catch up. Even if they don't understand exactly, you can't predict all the consequences of how people interact with a person who thinks they can predict the future, and especially people who act like some people are better at it than others. It works, until it doesn't.
contingencies 8y ago
Different brain regions … trade in different kinds of prediction. Seems like disparate departments within organizations do the same.
ddmma 8y ago
Interesting that now we transfer this prediction capability onto machines
skate22 8y ago
I pretty much assume that IF we were all knowing of the state of everything (matter, energy, & anything else that is / could be) and IF we were all knowing of the rules that defined the state transitions that we could predict the future (the state of everything at some observation point) We will likely never satisfy the above 2 assumptions, but our brains try to do the best they can with their limited sample of data / last gen processors. So sure, we may never predict future states with 100% certainty like above, but definitely with some % > 0
- opportune 8y ago
  not on an infinitely small scale: https://en.wikipedia.org/wiki/Uncertainty_principle https://en.wikipedia.org/wiki/Uncertainty_principle on a macro scale, with a reasonable margin of error, sure. The problem comes when small scale effects begin to have large scale consequences: human thought might be chaotic enough (in the physical, not colloquial sense) to be like this. In fact there is a whole branch of physics/math that deals with this issue: https://en.wikipedia.org/wiki/Chaos_theory https://en.wikipedia.org/wiki/Chaos_theory
  - skate22 8y ago
    I'm no expert on chaos theory, or physics, or much of anything, but directly from your 2nd link: "Chaos: When the present determines the future, but the approximate present does not approximately determine the future." Seems to support part* of my theory more than defeat it. Perhaps our current understanding of existence does not allow us to predict the future state of complex systems due to our level of uncertanty at a micro scale. Perhaps some day we will come to a better understanding that will. Surely we don't have the full picture yet, but i think we have a lot more reasoning to do before I will accept that existence is unreasonable
aalleavitch 8y ago
Kind of tangential, but I've always liked to think of memory as "predicting the past". I think we experience time in the direction that we do because it's easier to predict what's going to happen in the direction of less entropy. Given our current state, it's much easier to reconstruct what our state in the negative time direction must have been, as opposed to what our state in the positive time direction will be. Basically, our memory and our experience of time has a direction specifically because of entropy.

71 points by tdurden 8y ago

Another theory along these lines is by Jeff Hawkins of Numenta in his book "On Intelligence". He called in the Memory prediction framework.

ilaksh 8y ago
I was going to mention this https://arxiv.org/abs/1609.03971 https://arxiv.org/abs/1609.03971 which was partially inspired by Hawkins' work.

s-shellfish 8y ago

Calling it future prediction is problematic. We can't predict the future. It's a controlled delusion and sometimes some people will try to override your own reasoning system, and you honestly don't want to give them that power in exchange for it. It's called inferencing. Sometimes we have so much information we don't understand why things make sense in the order they do. It's not future prediction. It's confusion between information that is relatively predictable because it's stable information with very little variation, plus confirmation bias of being correct about highly variant information we are not accustomed to being right about in a particular order. Do not call it future prediction. Sometimes you will be correct about things before things happen. That's just chance. If it can happen, it can happen, it's in that space. But it's fundamentally flawed reasoning. There's still a whole open space of things yet to be experienced. That's what existence is defined as. You don't know what you know until you know what you know. It's not magic. It's called experiencing new things. Calling it future prediction primes you into bias. Which closes you off from opportunities eventually, because you literally stop being able to see things for how they are truly. Pattern matching brain. Being able to comprehend reality is absolutely vital to survival. Don't get lost in the waves of chaos. People can and do try to take control of confusion.

timb07 8y ago
> We can't predict the future. What tosh. Of course we can predict the future... and sometimes our prediction is correct. Data compression works by modelling the data stream and predicting the next symbol based on that model, and then recording in an efficient way the degree to which the prediction was correct. The comparison between the prediction and the actual outcome is the key point -- just as it is in the article.
- s-shellfish 8y ago
  Eventually the system you use to model information with will override actual data. That's a feedback loop of confirmation bias. It's thinking in terms of time. How many models of data do you interact with versus how much actual data are you interacting with. I'm not on a soapbox without a point. It's a fine line. Your own mind matters in these systems. It's important to not undervalue it. It's not a computer. It's not a better computer. It's sometimes flawed. I'm fine with calling it probabalistic inferencing. It's not future prediction, especially when the models one is reasoning about are models of cognition. Thank you for not down voting though, that's appreciated.
saas_sam 8y ago
This: "because you literally stop being able to see things for how they are truly" is flawed. You never see things for how they are truly. You see them through a filter of your sensory capabilities, your mental organization, and, as this article discusses, your aptitude for prediction.
- s-shellfish 8y ago
  If you only see the world through models of information, that is your perception. That doesn't negate the fact that there's measurable, testable reality that can be reproduced using scientific rigor. It's just a perspective from someone who has experienced being carried away with the waves. Aptitude for prediction helps you and tells you it's all working as expected until it slams you with hard consequences that leave enough impact to know where to draw your own line. My own father told me these things. I didn't listen. Guess what happened? It's arrogant to assume people can't catch up. Even if they don't understand exactly, you can't predict all the consequences of how people interact with a person who thinks they can predict the future, and especially people who act like some people are better at it than others. It works, until it doesn't.

contingencies 8y ago

Different brain regions … trade in different kinds of prediction. Seems like disparate departments within organizations do the same.

ddmma 8y ago

Interesting that now we transfer this prediction capability onto machines

skate22 8y ago

I pretty much assume that IF we were all knowing of the state of everything (matter, energy, & anything else that is / could be) and IF we were all knowing of the rules that defined the state transitions that we could predict the future (the state of everything at some observation point) We will likely never satisfy the above 2 assumptions, but our brains try to do the best they can with their limited sample of data / last gen processors. So sure, we may never predict future states with 100% certainty like above, but definitely with some % > 0

opportune 8y ago
not on an infinitely small scale: https://en.wikipedia.org/wiki/Uncertainty_principle https://en.wikipedia.org/wiki/Uncertainty_principle on a macro scale, with a reasonable margin of error, sure. The problem comes when small scale effects begin to have large scale consequences: human thought might be chaotic enough (in the physical, not colloquial sense) to be like this. In fact there is a whole branch of physics/math that deals with this issue: https://en.wikipedia.org/wiki/Chaos_theory https://en.wikipedia.org/wiki/Chaos_theory
- skate22 8y ago
  I'm no expert on chaos theory, or physics, or much of anything, but directly from your 2nd link: "Chaos: When the present determines the future, but the approximate present does not approximately determine the future." Seems to support part* of my theory more than defeat it. Perhaps our current understanding of existence does not allow us to predict the future state of complex systems due to our level of uncertanty at a micro scale. Perhaps some day we will come to a better understanding that will. Surely we don't have the full picture yet, but i think we have a lot more reasoning to do before I will accept that existence is unreasonable

aalleavitch 8y ago

Kind of tangential, but I've always liked to think of memory as "predicting the past". I think we experience time in the direction that we do because it's easier to predict what's going to happen in the direction of less entropy. Given our current state, it's much easier to reconstruct what our state in the negative time direction must have been, as opposed to what our state in the positive time direction will be. Basically, our memory and our experience of time has a direction specifically because of entropy.