vibe

I don't know quite how to articulate or formalize this, but I get a sense that there is something fundamentally analogue, 'periodic' or 'frequency domain'-ish, about human consciousness and mental processing. A few threads:

• the deep relationship that music seems to have with emotion and psychological state
• the existence of bodily tremors, as in tension release exercises, leg bouncing, autistic stimming, shivering from the cold
• brain waves
• meditatively sometimes it feels like sensation in different regions of my body has different 'frequencies', so that it's hard to keep both regions in consciousness at once, and when I do there's a sense of subtle dissonance between them
  • Andres Gomez Emilsson thinks that harmonizing these is 50% of the mechanism of action of meditation
• I've had an experience on 5-MeO-DMT where it just felt clear (in a way I can't really now remember or describe) that all the senses (not just hearing) were made up of the same kind of vibratory energy
• tons of linguistic metaphors:
  • parents being 'attuned' to children (an aspect of secure attachment)
  • relationships being 'harmonious'
  • people and groups having different 'vibes'
  • ideas or statements 'resonating' with our experience

Michael Johnson lays out a version of this view very explicitly:

• https://opentheory.net/2018/08/a-future-for-neuroscience/
• https://opentheory.net/2019/11/neural-annealing-toward-a-neural-theory-of-everything/

This is based on work on "connectome-specific harmonic waves" by Atasoy, Donnelly & Pearson. https://www.nature.com/articles/ncomms10340

And there is an older (2006) very extensive survey on "Rhythms of the Brain" by György Buzsáki: https://www.goodreads.com/en/book/show/1241576 Which was reviewed by ACX! https://www.astralcodexten.com/p/book-review-rhythms-of-the-brain

Also related:

• Music of Consciousness: http://commons.trincoll.edu/dlloyd/files/2012/07/Lloyd-2013-Music-of-Consciousness.pdf

See also work by Andres at QRI on quantifying the characteristic frequencies of different psychedelics (https://qualiacomputing.com/2020/10/09/modeling-psychedelic-tracers-with-qris-psychophysics-toolkit-the-tracer-replication-tool/):

LSD and DMT seem to produce strobe and replay patterns of markedly different frequencies. For DMT, the spatial and temporal frequency of the visual hallucinations is usually described as “very high”. Based on the replications thus far, along with personal reports from a musician I trust, DMT’s “characteristic frequency” seems to be in the 25 to 30 Hz range. In contrast, LSD’s frequency is more in the range of 15 to 20 Hz: both Dubois & VanRullen’s LSD tracer study and subjective reports I’ve gathered over the years point to the hallucinations of acid having this rough frequency.

What implications does this view have for artificial intelligence?

1. It's clearly not necessary for learning machines to have these sorts of dynamics, insofar as transformers seem to work quite well for language modeling.
2. Anything that aspires towards human-like emotional processing --- being 'attuned' with humans in a real way --- will need to implicitly or explicitly model human vibes. This can probably be done without itself being a 'vibes-based' system (just as digital computers can process music).