top of page

Mathematical description of conscious access.

Updated: May 29, 2022

Studying the "Study of Consciousness" (Stanislas Duanne) can further deepen your understanding of coaching theory.

I am studying to add a unique flavour to "unconscious rewriting".


This series of blog posts are my study notes. This time, the theme that follows the unconscious and conscious

I will write a "sign of consciousness".


 

Modern science requires a description of phenomena and a mathematical explanation.


"Conscious Access engraves thoughts in the Global Workspace Network by engraving patterns of active and inactive neurons." Gaston Bachelard, The Formation of the Scientific Spirit (1938).


This expression gives an intuitive understanding of what consciousness is, but "how does a neural network work" and "why it is shown by macroscopic patterns of EEG recordings, neurophysiological. It must be replaced by mathematical theory, which explains "does it produce a sign?"


Jean-Pierre Change and Duanne simulated on a computer a large number of sections of the visual cortex, parietal region, and prefrontal cortex involved in sub-threshold and conscious processing.


The four hierarchical regions are connected by feedforward and long-range feedback neural connections.




Each region comprises cortical cells organized in layers and connected to neurons in the thalamus.


The activation propagated from low to high levels with a short stimulus and then disappeared. This phase is a state of short-term activation of the cortical pathway when subconscious (unconscious) perception occurs.

After a little longer stimulus, a global ignition arose. The top-down coupling amplifies the input, creating a long-lasting second wave of activity. This state of activation is in which conscious perception is occurring.

They made the virtual neurons into local cortical columns, and the subdivision of the cortex into interconnected cell layers was reproduced.

This model is designed with this biological composition in mind. The neurons inside the column tend to support each other and respond to similar inputs.


They have small-scale reproductions of the thalamus, consisting of multiple nuclei strongly connected to specific parts of the cortex and different parts.


We constructed a rough model that simulates the thalamic cortex column, the basic calculation unit of the primate brain.

With no input, virtual neurons were programmed to spontaneously fire and form brain waves similar to those produced by the human cortex.


The neurons that encode sound and the neurons that encode light are activated simultaneously without interfering with each other.

But at higher levels of the cortical hierarchy, neurons can actively suppress each other and have only one state, the only "thinking" that integrates firing.


What is essential is that the network has a mechanism to send information back to the lower regions via long-range feedback projection so that the higher regions are the significantly lower sensory cortex that stimulated themselves. It is given support by stimulation.


When a stimulus pulse is given, it slowly climbs the cortical hierarchy from the primary region to the secondary region to the tertiary and quaternary regions. This phenomenon is the state of global ignition.


This wave of feedforward activity mimics the propagation of neuronal activity up the hierarchy of the visual cortex.

After a while, the entire column encoding the perceptual object ignites.

Through feedback coupling, neurons encoded for the same sensory input interact and enhance by exchanging stimulatory signals.

The signals cause the ignition of activity to occur suddenly. Meanwhile, the alternative perceptual representation is positively suppressed. This activity lasts for hundreds of milliseconds.


This length is independent of the length of the initial stimulus given, and even a very short external stimulus will continue to have a persistent echoing state.


These computer-simulated experiments capture the brain's ability to produce long-lasting images of flashed images, similar to the results of previous real-world experiments.


In running the simulation, they gained the following mathematical insights:



Conscious access is what theoretical physicists call a "phase transition." For example, the sudden transition from one state to another in the physical phase of water turning into ice.




Thus, when a phase transition occurs, the system's physical properties suddenly change discontinuously.


Higher-dimensional neurons activated by stimuli from lower dimensional neurons send stimuli to the lower dimensional neurons, so the system has two stable states.


It is unpredictable what state an intermediate intensity stimulus will result in, and activity will either diminish quickly or suddenly jump to an elevated state.


The unconscious process corresponds to neurons that propagate from region to region without causing global ignition.

On the other hand, conscious access suddenly transitions to a higher dimensional state in which synchronized brain activity takes place.

"It will take a considerable amount of time to build a theory that properly explains the phase transitions that occur under the dynamics that govern real-life neural networks," Duanne believes.


It will be essential to investigate the cause of

the "phase transition" in the future.

4 views0 comments

Comments


bottom of page