I don’t think it is too controversial to say that the basic HTM update cycle roughly corresponds to the Alpha wave.
That’s where I was and the articles @CollinsEM cited above seem to support this. There are also some video and audio files of Greek classicists ‘singing’ portions of the Illiad which are quite rhythmic and ‘sing-songy’. These songs were sung by wandering bards (aoidoi) over centuries of time. Now I need to go see if poetry and lyrics are easier to remember than prose.
There is no question in my mind that the human brain has a unique ability to memorise very long sequences with remarkable accuracy. A concert pianist playing a piano concerto will play many thousands of notes with virtually 100% accuracy. Most people (myself not included) can recite long poems. Stage actors likewise. You don’t need examples from centuries past to show this. There is a strong likelihood that HTM or a mechanism of that kind plays a major role.
However, this all has nothing to do with alpha/beta/delta/theta (‘greek’) cycles, which my question was about. Sorry if you misunderstood.
1 Like
Not a problem, we all get confused about things sometimes, or fail to see the inherent patterns that exist.
There is a lot of such cycle interactions, as direct synchronisation methods, in this book which may help.
(Chapter 6 - Brain Rhythms Provide a Syntax for Neural Syntax).
I haven’t fully digested it myself yet - but there seems to be much overlap with this thread.
1 Like
There is indeed. Check out Buzsáki’s lab, a treasure trove.
1 Like
Indeed, 7 seems to be very important. The horizontal cortical context of a cortical column is 6 (i.e. neighbor columns). Actually, a part of the input to that particular column (most likely at L4) has 7 cortical columns contributors (i.e. affected by axons that belongs to 7 different cortical columns) in the previous rung of the hierarchy.
I would guess that excitatory connections between neighboring columns are lateral, through L2-L3. Connection through thalamus / L4 is much slower and more expensive, unless you have evidence on that?
I was thinking in cortico-cortico connections (e.g. V1->V2). L4 at V2 receives both from V1 and Cortico-thalamic loop inputs. V1 outputs into V2 are somewhat retinotopic ( “less” retinotopic that LGN outputs into V1). I would guess that thalamic inputs are somewhat also retinotopic, but that’s another different beast. L2-L3 are mostly intra-cortical column connections… I think.
(my understanding of cortical column is in the sense o V.Mountcastle… not HTM terminology)
1 Like