t, there’s a newness or “surprise” factor in almost every result as I connect it to my own private akashic record or Platonic realm as it were that I consider my list of rules rather than forever-objects as every rule is subject to change. Building on firm sand as I don’t believe in solid.

My take at the bottom:
Bayesian cognitive science, predictive brains, and the nativism debate -Matteo Colombo
Conclusion
Bayesianism offers a fertile source of modelling tools rather than a well-understood and empirically supported theory of the innate architecture of the human mind. A substantial account of the acquisition of a given psychological trait should be able to answer a number of questions. For example, are we talking of an evolved trait, or of a trait culturally acquired during development, or of a trait acquired through a developmental process triggered by a narrow range of variation in environmental conditions? And what kinds of learning mechanisms and representations are necessary for the acquisition of the trait? While Bayesianism alone cannot answer these questions, it can frame them in a precise and transparent way, combining aspects of both Connectionism and Classicism. Like Connectionism, Bayesianism shows that an innate language of thought is not required to account for the acquisition of high-level psychological traits. Like Classicism, Bayesianism offers a transparent way of evaluating the character of the innate structure in the human cognitive architecture. Combining these aspects, what Bayesianism brings to the table is not a vindication of either nativism or empiricism, but one flexible and precise way to transparently capture and evaluate the character of the innate structure in the human cognitive architecture.
—Ken’s takeaway—
Short: My fascination with Connectionism since I was a teenager continues to be useful while Hierarchical Bayesianism as a cognitive modeling (NOT for pointing to physical structures) gives a precision to descriptions that Connectionism cannot.
I loathed Bayesian for decades because I did not understand BELIEF UNTIL a few weeks ago when I found HIERARCHICAL BAYESIAN models of ADHD, Autism, OCD, psychosis and other belief-related things. One for either ADHD or Autism (I forgot which) show that much of the thinking can be successfully modeled by increasing present and future DETAIL (or error rate depending how you look at it), which effectively mutes the involvement of the PRIORS.
But not all priors because there’s a hierarchy and priors at a broader level in the hierarchy CAN work to constrain belief just as regular priors BUT because this process uses different priors HIDDEN to people accustomed to using standard social/cultural priors, their behavior can seem odd or strange, “fixated”, or off the wall when stressed, or unable to make conclusions or making analogies that appear odd to people accustomed to standard sociocultural priors.
So for me, that means I have long lists of rules I follow and instead of beliefs I consider myself as having degrees of certainty related to how well something follows a rule or does not follow a rule. These rules are generally explicit although there’s what I call “having a strong intuition” which I trust which likely serves as my superordinate in my hierarchy of priors but when evaluating something in the present moment, there’s a newness or “surprise” factor in almost every result as I connect it to my own private akashic record or Platonic realm as it were that I consider my list of rules rather than forever-objects as every rule is subject to change. Building on firm sand as I don’t believe in solid.
“Short: My fascination with Connectionism since I was a teenager continues to be useful while Hierarchical Bayesianism as a cognitive modeling gives a precision to descriptions that Connectionism cannot. One for either ADHD or Autism show that much of the thinking can be successfully modeled by increasing present and future DETAIL , which effectively mutes the involvement of the PRIORS. But not all priors because there’s a hierarchy and priors at a broader level in the hierarchy CAN work to constrain belief just as regular priors BUT because this process uses different priors HIDDEN to people accustomed to using standard social/cultural priors, their behavior can seem odd or strange, “fixated, or off the wall when stressed, or unable to make conclusions or making analogies that appear odd to people accustomed to standard sociocultural priors.”

Leave a comment

Your email address will not be published. Required fields are marked *


× 9 = thirty six

Leave a Reply