Transcendent Experience and The Bayesian Brain

Manuel Brenner
7 min readJul 19, 2020

“If the doors of perception were cleansed, every thing would appear to man as it is, Infinite. For man has closed himself up, till he sees all things thro’ narrow chinks of his cavern.”
William Blake

William Blake, Jacob’s Ladder/ Public domain

In our everyday life, we inhabit a low-resolution version of the world. Cognition is expensive, and so when we live in familiar surroundings, our brains go into energy-saving mode. We operate on the level of what Buzsáki calls the good-enough brain, and which can be roughly equated to Kahneman’s system one: a level of brain activity that is not particularily concerned with getting to the bottom of the truth, but rather optimized for a cognitive activity that lets us survive with the minimal energy cost.

But as William Blake says, alluding to Plato’s allegory of the cave: this makes us spend our daily lives closed off at the base of a cavern, out of touch with the light of Plato’s transcendental realm of ideas, with what Blake calls the Infinite.

The Bayesian brain hypothesis details how cognition lends itself to a statistical interpretation, grounded in Bayes’ law: every perceptual act consists of the integration of sensory data into a prior framework, defined by Bayesian priors, via posterior inference. The confidence in these priors is in technical terms given by their precision.

Photo by Robina Weermeijer on Unsplash

According to Kant (I wrote a short introduction here), our subjective perspective on the world is filtered through what he conceived of as an a priori perceptual structure. If we want it or not, we perceive everything filtered through the lens of the structure of our subjectivity. The parallels to Bayesian priors are no coincidence: our world model determines how we perceive the world, while the Kantian “Ding an sich”, the “Thing in itself”, eludes our grasp and lies beyond our perception.

The Bayesian brain hypothesis further proposes that our knowledge of the world is structured in hierarchical Bayesian models (as I explain in more detail here). Based on these models we can cast predictions about the future state of the world. Our every-day perception of the world is determined by our already existing…

--

--