Wanted: Architecture for changing minds: A comment on "The growth of cognition: Free energy minimization and the embryogenesis of cortical computation".

The discussion of what makes a neuron survive or instead be sacrificed by apoptosis is an excellent opportunity to remind oneself that the cortex is a network. A community were the fate of the individual is determined by the collective. This is demonstrated most convincingly in Wright and Bourke’s treatment of the role of neural synchrony during cortical embryogenesis, as well as by their impressive growth simulations [4]. While doing so, they provide a bridge between Friston’s powerful Free Energy principle [1] and Perlovsky’s elegant Dynamic Logic framework [3], and complement them both in encouraging ways. The proposal that the survival of neurons is promoted by their synchronized activity with other, sometimes distant, neurons fits well with previous data, as reviewed. Moreover, it can help explain long-standing conundrums, such as why the innate columnar selectivity in visual primary cortex still requires actual experience for maintaining that organized selectivity [4]. It is further proposed that this synchrony is spatially partitioned and transcends small-world assemblies, as well as that this model is applied in all cortical areas. Naturally, what dictates the brain’s architecture and how it evolves to suit its many functions is a monumental topic of prolonged scientific interest. In discussing the architecture, however, it is often apparent that we implicitly consider the brain as an organ that copes with varied circumstances with the same, fixed machinery. But as will be argued below, the human brain is able to accommodate diverse contexts, which can be radically different, by changing states of mind dynamically. Therefore, an important step towards our understanding of the underlying architecture, extending from the work of Wright and Bourke’s, is thinking how the same architecture can subserve different states. States of mind, as we have recently defined them [2], can vary along multiple dimensions. Here are a few examples. Perception can be based on bottom-up information from the senses, or lean more on top-down expectations that are based on experience as stored in memory. When attending a physical scene, we may be focused on the local details (i.e., the “trees”) or on the global information in that scene (i.e., the “forest”). We can be thinking in a broad and associative manner, or we may be focused with our thoughts narrowly on the same topic for an extended amount of time, perhaps even ruminating in more clinical cases. Our mood can be more positive or more negative. And when it