neuroscience - Is the minicolumn the unit of the neocortex?

There are many arguments for what the unit of the neocortex is. "Columns" seem to be the standard, but what exactly those are is extremely contradictory between individuals, cortical regions, and species. Often times, when a column is referred to, it's actually a functional column without any anatomical borders (such as Hubel and Wiesels ocular dominance columns). Sometimes, the "column" is semi anatomical, such as the rat barrel cortex. Other times, these functional columns are confabulated into anatomical units, without any evidence for a border.

So my question is, could Mountcastle's "minicolumns" be the actual anatomical unit of the cortex? I've heard arguments that they are mere developmental relics. But they seem like the only reliable and consistent unit in the cortex.

Answer

I am skipping the background information (I will amend my question later) for now.

So, what is known about minicolumn?

1. The term was coined by Ramon y Cajal who saw narrow stripes, running from white matter to the surface (reference) in his preparation with Nissl staining.


2. The hypothesis about minicolumns to be anatomical units was proposed by Mountcastle in 1957, known as "columnar hypothesis" in neuroscience (reference).


3. It has been shown that minicolumns grow from progenitor cells within the embryo (reference), and don't require brain functions to emerge.


4. Minicolumns contain neurons within multiple layers of the cortex (reference).


5. One of the latest review on the topic (Buxhoeveden & Casanova, 2001) contains 121 references to experimental papers on this topic.

I wouldn't say that these columns are developmental relic, for there is a known heterogenity among these structures (like hypercolumns, macrocolumns and segregates) and it is rather improbable that all these types are relics.

So, as long as we stick to the term "anatomical unit", I see no problems why columns (with all possible types and sizes of them) should not be it.