Consciousness is oblivious to [the] complex operation[s of the brain]. Subjectively, we see only the end result[s], (...) without ever realizing that our initial sensations were ambiguous and that our neuronal circuits had to work hard to make sense of them.'
Fascinatingly, the convergence process that leads our neurons to agree on a single interpretation vanishes under anesthesia. The loss of consciousness is accompanied by a sudden dysfunction of the neuronal circuits that integrate our senses into a single coherent whole. Consciousness is needed for neurons to exchange signals in both bottom-up and top-down directions until they agree with one another. In its absence, the perceptual inference process stops short of generating a single coherent interpretation of the outside world.
The role of consciousness in resolving perceptual ambiguities is nowhere as evident as when we purposely craft an ambiguous visual stimulus. Suppose we present the brain with two superimposed gratings moving in different directions. The brain has no way of telling whether the first grating lies in front of the other, or vice versa.
Subjectively, however, we do not perceive this basic ambiguity. We never perceive a blend of two possibilities, but our conscious perception decides and lets us see one of the two gratings in the foreground. The two interpretations alternate: every few seconds, our perception changes and we see the other grating move into the foreground. Alexandre Pouget and his collaborators have shown that 'when parameters such as speed and spacing are varied, the time that our conscious vision spends entertaining an interpretation is directly related to its likelihood, given the sensory evidence received. What we see, at any time, tends to be the most likely interpretation, but other possibilities occasionally pop up and stay in our conscious vision for a time duration that is proportional to their statistical likelihood. Our unconscious perception works out the probabilities—and then our consciousness samples from them at random.'
The existence of this probabilistic law shows that even as we are consciously perceiving an interpretation of an ambiguous scene, our brain is still pondering all the other interpretations and remains ready to change its mind at any moment. Behind the scenes, an unconscious Sherlock endlessly computes with probability distributions: as Peirce inferred, “the whole fabric of our knowledge is one matted felt of pure hypothesis confirmed and refined by induction.” Consciously, however, all we get to see is a single sample. As a result, vision does not feel like a complex exercise in mathematics; we open our eyes, and our conscious brain lets in only a single sight. Paradoxically, the sampling that goes on in our conscious vision makes us forever blind to its inner complexity.
Sampling seems to be a genuine function of conscious access, in the sense that it does not occur in the absence of conscious attention. Consider binocular rivalry, the unstable perception that results from presenting two distinct images to the two eyes. When we attend to them, the images ceaselessly alternate in our awareness. Although the sensory input is fixed and ambiguous, we perceive it as constantly changing, as we become aware of only one image at a time. Crucially, however, when we orient our attention elsewhere, the rivalry stops.
Discrete sampling seems to occur only when we consciously attend. As a consequence, unconscious processes are more objective than conscious ones. Our army of unconscious neurons approximates the true probability distribution of the states of the world, while our consciousness shamelessly reduces it to all-or-none samples.
The whole process bears an intriguing analogy to quantum mechanics (although its neural mechanisms most likely involve only classical physics). Quantum physicists tell us that physical reality consists in a superposition of wave functions that determine the probability of finding a particle in a certain state. Whenever we care to measure, however, these probabilities collapse to a fixed all-or-none state. We never observe strange mixtures such as the famed Schrödinger’s cat, half alive and half dead. According to quantum theory, the very act of physical measurement forces the probabilities to collapse into a single discrete measure. In our brain, something similar happens: the very act of consciously attending to an object collapses the probability distribution of its various interpretations and lets us perceive only one of them. Consciousness acts as a discrete measurement device that grants us a single glimpse of the vast underlying sea of unconscious computations.'
(Uit: Dehaene, S, Consciousness and the Brain, hfd.3)
-------
Commentaar: dit is een markante passage uit het boek van Dehaene. Hij beschrijft hier de functie van bewustzijn (en daar draait de global workspace theory om: de achterliggende gedachte is dat bewustzijn een functie moet hebben, anders rijmt het bestaan van dit vermogen niet met het naturalisme. Global Workspace Theorie is dus een zogenaamde 'functionalistische' theorie van de geest.)
De functie van het bewustzijn is om de neurale producten (sample) van alle afdelingen in het brein -indien noodzakelijk- te integreren. Waarom deze passage mij sterk is bijgebleven: omdat het bewustzijn slechts één product toelaat per keer. In het geval van 'rivalry' alterneren de mogelijke neurale producten -je ziet of het ene product of het andere- en nooit staat het bewustzijn twee cognities tegelijkertijd 'voor de geest'.
Deze functionalistische theorie stemt overeen met mijn indruk dat de mens een logische, binaire machine is: alle cognitieve apparaten staan in dienst van het lichaam, dat handelingen slechts één voor één optimaal kan uitvoeren (het lichaam is zo gebouwd dat het onmogelijk handelingen optimaal tegelijkertijd kan uitvoeren: er is sprake van 'rivalry' tussen de verschillende doelen die we willen bereiken en de beschikbare lichamelijke middelen die we nodig hebben om die doelen te bereiken).
Logisch denken en doen is -volgens mij- een biologisch schema met een duidelijke biologische functie, dat nauwelijks verder reikende epistemologische implicaties kan hebben. Uit het feit dat wij onze cognities logisch ordenen kan dan ook niet volgen dat de werkelijkheid logisch geordend is (wat er wel uit volgt is dat wij de werkelijkheid logisch moeten ordenen). Met andere woorden, wij ordenen de werkelijkheid niet logisch omdat zij logisch is, maar zij schijnt logisch omdat wij haar logisch moeten ordenen.
(Zie de bijdrage van 18-09 voor een eenvoudige weergave van Global Workspace Theory).
Geen opmerkingen:
Een reactie posten