Information theorists have been preoccupied with how the brain might compress or ‘chunk’ information to enhance the information transfer rates (Miller 1956). On this point, it is known that a non-trained person can discriminate six levels of pitch (2.5 bits), but a trained person is known to discriminate 50 to 60 levels of pitch (~ 5.8 bits, Miller 1956). Once discriminations are based on multidimensional parameters, information increases with the magnitude of dimensionality (Miller 1956). The discrimination of faces or words for language is a multidimensional process, and in the case of speech it has up to ten dimensions (Miller 1956). After the 1950’s the world made it its business to educate children en masse, which means speakers of language throughout the world are now communicating at high information transfer rates. Indeed, accomplished communicators deliver verbal language at a rate of ~ 40 bits per second (a trillion possibilities per second) irrespective of language-type (Coupé et al. 2019; Reed and Durlach 1998).
We now accept that movement and thinking are mediated by common circuits in the brain (Andersen and Mountcastle 1983; Darlington and Lisberger 2020; Jame 1890; Kimura 1993; Sacks 1976, 2012; Schiller and Tehovnik 2015; Vanderwolf 2007) despite decades of opposition to this idea (e.g., Chomsky 1965, 2012, 2023; also see: Bushnell, Goldberg, and Robinson 1981). It is known that when spatial information is consolidated during slow-wave sleep based on a rodent having run along a track repeatedly that the stored information is compressed temporally (Wilson and McNaughton 1994). However, consolidation during REM sleep occurs in real time (Louie and Wilson 2001), perhaps because REM is a model of waking state which operates in real time.
We know that every form of accomplished behavior requires serious dedication whether it is the understanding of physics by Albert Einstein, or computing the best move on a chess board by Gary Kasparov, or assessing the football pitch by Edson Do Nascimento (Pelé), or defeating all opponents on a 100-meter track by Usain Bolt. All these accomplishments have one thing in common: years of focused effort dedicated to a single goal, which is secured by strengthening the connections between the neocortex and cerebellum through much study and repetition (Hasanbegović 2024). This permits for decisions about relationships to be made at lightning speed and at the highest level of performance, i.e., at the highest level of automaticity. When Gary Kasparov played the supercomputer Deep Blue, he was able to transmit some 28 bits per second, namely, ~ 300 million possibilities per second (Tehovnik, Hasanbegović, Chen 2024). And finally, the way to appreciate ‘E = mc^2’ is to think of it as a chunk of information (Miller 1956), which is based on years of reflection and data analysis by Albert Einstein’s nervous system. Surely, someone will need to determine how many bits of information are represented by ‘E = mc^2’.
One chunk of information, which can be addressed by a symbol, is represented by about 30 bits.
Regards,
Joachim
Most people, including information theorists, usually regard 'information' and 'data' as being different things. Data is a physical item which takes up some space on a physical medium measured in binary digits, whereas information is the result of interpreting some message. So the phrases like, 'bits of information', or, "a 'chunk' of information" is 'ontologically challenging'.
The question goes on to confuse human memory (neurons) and computer memory (bits) e.g. "... data analysis by AE's nervous system....". Again, bits are a measurement of data, whereas neurons perform neural activity.
(Miller, 1956) talks about the entropy co-firing of neurons, using phrases like "an increased tendency to fire together", but if computer memory worked like this computers simply could not work.
Sorry if this is not what you wanted to hear.
Kind regards,
Martin
I am not a computer guy, but I understand how the nervous system works. So, in the case of the neocortex/cerebellum, the information is stored in the form of objects and sound bites in explicit form. This explicit form (also referred to as declarative information, Corkin 2002) comes about by interacting with the external world: as when memorizing a list of sentences so that they can be delivered during the performance of a school play. The hippocampus is central to the consolidation process (which is finalized during sleep, Marr 1971) and which has information (or data) stored in specific locations of the brain (locations that are related to the sensorimotor attributes of the stimuli in the neocortex: language is stored in Wernicke’s and Broca’s areas, faces are stored in the temporal and orbital cortices, and so on). To retrieve this information, the learned sentences are re-concatenated (in the same way they were during the consolidation process) and delivered at a particular rate of information transfer, e.g., 40 bits per second through the skeletomotor system (Reed and Durlach 1998).
Neuroscientists have come up with two values regarding information: the amount stored and expressed in trillions of bits (and based on synaptic connectivity, Hebb 1949), and the amount transferred at any one time (based on the content transferred through speech or other movements; note that even if you are immobile information is still being transferred to maintain the autonomic nervous system). The neocortex and cerebellum are central to the transfer process (together these structures represent over 95% of the neurons in the mammalian nervous system, Herculano-Houzel 2009). The cerebellum converts the explicit coding of neocortex into executable code expressed as a rate code, which is the code used by the muscles (Tehovnik, Patel, Tolias et al. 2021).
Other details of how the information is linked during a stream of execution whether through thinking or actual delivery will be provided once the book is published, since this has been a seven-year investigation that covers over 2,000 pieces of published peer-reviewed research (Tehovnik, Hasanbegović, Chen 2024).
I am a computer guy and you seem to revert to my sort of language (information, bits, storage, transmission, executable code etc.) to describe the workings of the mind. Surely, its more usual for computer guys to try to use this analogy, no? This seems to promote the computational paradigm way above its status, computers are complex but completely understood devices, so the mind/brain=software/hardware analogy fails on any inspection.
- If a computer were to record the sequence of a shuffled playing card deck, would require an array of 52 6-bit values; however, if someone were to memorise this they wouldn't simply use 312 bits.
- A typical technique employed to do this cognitively is to form a narrative, such as the King of Spades was in his counting house but he only had Three Diamonds, so he had to rethink how he could impress the Queen of Hearts, and so on.
You say that sound bites are stored in the neocortex/cerebellum in explicit form. Does it follow that pictures are stored in the visual cortex, as a 'raster display', so to picture the first 6 bits (the King of Spades) you'd 'see' a picture of him in a counting house?
Does your analogy not lead to, or at least suggest, the possibility of uploading a mind onto the internet?
I'm intrigued!
Here is our course on the brain:
https://www.researchgate.net/profile/Edward-Tehovnik/research
"The Fantastic Voyage: from the Retina to the Eye Muscles. A Lecture Series. Third Edition."
Enjoy,
PS. A nice recent paper to check out available on RG: Dijksterhuis, D.E., Self, M.W., Possel, J.K., Peters, J.C., van Straaten, E.C.W.,…Roelfsema, P.R., 2024. Pronouns reactivate conceptual representations in human hippocampal neurons. bioRxiv doi:10.1101/2024.06.23.600044.
- Badges
- Science topic
- Similar topics
- Psychology
- Cognitive Science
- Mind