Ep. 31 - Awakening from the Meaning Crisis - Embodied-Embedded RR as Dynamical-Developmental GI

Study Notes
, ,
1

topic

discussion

2

my notes

Embodied/Embedded

  • Embodied:
    • Trying to argue there is a deep dependency between cognitive agency as an intelligent general problem solver and fact brain exists within a bio-economy.
    • Body is not Cartesian-clay that we drag around and shape into the desires of our self-enclosed immaterial minds, not a useless appendage
    • Body is an auto-poeitic bioeconomy that makes cognition possible
    • Without it don’t have the machinery necessary for ongoing evolution of relevance realization
    • Body is constitutive of cognitive agency
  • Embedded:
    • Biological fittedness of a creature not a property of the creature per-se. It’s a real relation between the creature and the environment
    • Adaptivity not a property intrinsic to it per se. It’s an affordance between it and the environment
    • Relevance not a property in the object, not a property of subjectivity of the mind nor of objectivity
    • Property that is co-created by how environment and embodied brain fitted together in a dynamic evolving fashion
    • We don’t subjectively project relevance
    • We don’t detect it either empirically
  • Transjective: a real relationship between the organism and the environment
    • Relevance is realized.
    • Objective sense to realization: to make real.
    • Subjective sense: coming into awareness
  • Anti-cartesian: connection mind and body is one of dependence, mind needs the body.

(this is also the case in Integrated Information Theory which posits that information integration must be in 3d space. Embodied)

  • The world and the organism are co-creating, co-determining the fittedness
  • Need to transcend the objective and subjective - changes how we interpret spirituality: functional sense of connectedness that affords wisdom, self-transcendence, etc.

Efficiency-Resiliency Tradeoffs

  • Initially the machinery of RR has to be internal to an embodied embedded system of adaptivity
  • Want a way to optimize efficiency/resilience tradeoffs
  • Logistical norms (constraints) can be realized in particular virtual engines
  • Internal Bioeconomic Properties:
  • External Interactional Properties:
    • Eventually give rise to goals in the world as opposed to constitutive goals in the system.
  • Go back and forth
  • If want to be adaptive, a general problem solver - two kinds of machines can be:
    • General Purpose Machine:
      • Ex: hand: can be used in many contexts and many tasks.
      • Jack of all trades, master of none
      • Not as good as a hammer or screwdriver
      • Use the same thing over and over again
      • More efficient
    • Special Purpose Machine:
      • Outperform GPM in some contexts
      • More resilient
    • Sometimes you’re adaptive by being a GPM, other times by being SPM
    • Want to be able to move between these.

  • Want to be able to use the same set of functions over and over again. SPM don’t use as often, gets inefficient - carry it around but not using it. But makes me resilient. When a lot of new things that GPM can’t handle ready for it to deal with SPM
  • Want to constantly trade between them.

Information Processing

  • How would you make information processing more efficient?
  • Want functions using to be as generalizable as possible - use the same functions in many places
  • Relevance Realization in the Emerging Cognitive Science: paper
  • People who are trying to make AI already implementing these strategies

Data Compression (Application)

  • In science, take two variables, scatterplot w line of best fit
  • Line may touch none of data points. Allows to interpolate and extrapolate. Go beyond the data. But taking a chance.
  • Popper: all good science takes good chances
  • Allows to make predictions, generalize the function
  • Data compression
  • Trying to pick up on what’s invariant. Information always contains noise want to pick up on what’s invariant and extend that.
  • If the data compression lets me generalize my function, feeding through the sensory-motor loop, protecting and promoting auto-poeitic goals will be reinforced

Particularization (Scope)

  • Trying to create a function that overfits with the data
  • Emphasize pick up on variations
  • Make the system more context sensitive

  • Make the organism adaptive going back and forth. Though skewed one way or the other. Some are more biased to overfitting/particularizing, others to generalizing/compressing

  • If get scope going the right way will get coupled to this pattern of interaction that will fit you well to pattern of change/stability in the environment

Exploiting vs. Exploration

  • Deals with timing. Should stay here get as much as can (exploiting) or move and try and find new things. Trade off.
  • Different strategies being considered.
    • Reward when system doesn’t make an error, punish when makes an error. Makes it more curious and conscientious
    • Reward error reduction, punish error increase
  • Cognitive tempering vs. Projectibility of processing.

  • Not claiming these are exhaustive. Exemplary of ways in which can trade efficiency/resiliency and create virtual engines that will adapt through system of constraints
  • Exploitation efficient: don’t have to expend much, can just stay here. But things have to stay the same.
  • Exploration: have to expend a lot of energy but only rewarding if significant difference.
  • Sometimes what makes something relevant is how its the same, sometimes it’s what makes something different
  • Have to constantly shift between them because that’s what reality is doing
  • There are many of these trade-offs. All regulated by the opponent processing of efficiency/resiliency
  • Bioeconomics: cost of processing
  • Might also need to trade between scope/temporary and applicability/projectability

  • Flexible gambling: Have to be flexible in how gamble. May decide to hedge bets and activate as many functions as can, or go for the big thing and give priority to just this function
  • Sometimes focusing, sometimes diversifying
  • All of it can be represented mathematically

  • Scope, temporing, prioritizing between them
  • RR is always taking place in this space.
  • Moving around in a state space.
  • Developmental
  • Data compression - integrate information vs. differentiation
  • Can argue that these map onto two fundamental processes that Piaget said drive development: Assimilation
  • Assimilation: have a cognitive schema (set of constraints) - integrate assimilate new information - compression.
  • Opposite: accomodation: opens you up. Causes you to change structure, schemas.
  • If we just assimilate machinery gets brittle and distortive
  • RR is inherently dynamic and developmental - it develops by functioning and by functioning it develops
  • Complexification: simultaneously integrating and differentiating
  • If highly differentiated can do many different things but if not integrated fly apart
  • As systems complexify they self-transcend - go through qualitative development
  • Ex: start out life as zygote, initially cells just reproduce, then get cellular differentiation, also integrate, self-organize into heart, eye, etc. Biological complexification
  • Give emergent abilities: transcend yourself as a system
  • If we are RR we are inherently:
    • Dynamical
    • Self-organizing
    • Autopoeitic
    • Developmental
    • Self-transcending

RR as a Unified Phenomena

  • Spierman: general factor of intelligence
    • How kids doing in math predictive of how doing in english and sports
    • Strong positive manifold: interpredictability in different tasks.
    • Underwrites being a general problem solver

  • When testing intelligence often testing for general intelligence
  • Test:

1600×1510 275 KB

  • When trace these back:

1600×1285 217 KB

  • All the places need RR
  • RR is actually the underlying ability of general intelligence
  • RR is a good candidate for general intelligence.
  • GI is a unified thing. Keeps getting replicated.
  • If had to know one thing about you to predict: one thing that tells more than anything else. GI crushes how well do in an interview in predicting if get a job
  • So therefore RR is a unified thing. RR = GI.
  • GI is a dynamic developmental evolution of your sensory motor fittedness, regulated by virtual engines regulated by the logistical normativity of the opponent processing between efficiency and resiliency
  • Spending so much time on this because this is the lynchpin argument on the cognitive science side of the whole series. Everything leads to RR. Naturalistic explanation.
  • Argue: how RR embedded in procedural, perspectival participatory knowing, embedded in transjected dynamic coupling to the environment and the affordance of the agent/arena relationship, mind/body, central to mind/consciousness
  • Allows to explain so much at the center for human spirituality
  • How we can integrate cognitive science and human spirituality.
3

A post was merged into an existing topic: Ep. 32 - Awakening from the Meaning Crisis - RR in the Brain, Insight, and