I’ve often noted that I find more consilience than disagreement between the empirically grounded theories of consciousness. They seem to be looking at the problem at differing levels of organization, and together they may present a growing scientific consensus about how the mind works.
In particular, a few weeks ago, when discussing higher order theories, I made the observation that another theory historically discussed on this blog, the attention schema theory, could be considered a higher order representation. It’s gratifying to know that my conjecture wasn’t hopelessly off base, as Michael Graziano, the author of the attention schema theory, along with coauthors, has the same idea.
In a new paper (which unfortunately appears to be paywalled), they take a shot at reconciling various theories and philosophies of consciousness, and reach a striking conclusion: Toward a standard model of consciousness: Reconciling the attention schema, global workspace, higher-order thought, and illusionist theories:
Here we examine how people’s understanding of consciousness may have been shaped by an implicit theory of mind. This social cognition approach may help to make sense of an apparent divide between the physically incoherent consciousness we think we have and the complex, rich, but mechanistic consciousness we may actually have. We suggest this approach helps reconcile some of the current cognitive neuroscience theories of consciousness. We argue that a single, coherent explanation of consciousness is available and has been for some time, encompassing the views of many researchers, but is not yet recognized. It is obscured partly by terminological differences, and partly because researchers view isolated pieces of it as rival theories. It may be time to recognize that a deeper, coherent pool of ideas, a kind of standard model, is available to explain multiple layers of consciousness and how they relate to specific networks within the brain.
As a quick reminder, global workspace theories (GWT) argue that consciousness results from mental content making into a global workspace, often thought to range across the fronto-parietal network. Higher order theories (HOT) posit that consciousness involves secondary or higher order thought or representations of first order sensory representations or other more primal processing, with the higher order representations though to be in the prefrontal cortex, although possibly in other regions as well. The attention schema theory (AST) posits that awareness is a model, a schema, of the messy emergent multilevel process of attention.
Before attempting a reconciliation, the authors make a distinction between two views of consciousness, which they label i-consciousness and m-consciousness. i-consciousness is the information processing view of consciousness, how information is selected, enhanced, and used. m-consciousness is the mysterious experiential essence version. (This distinction seems similar, but not exactly equivalent, to Ned Block’s distinction between access consciousness and phenomenal consciousness.)
The theories noted above all focus primarily on i-consciousness, often to the frustration of those concerned about m-consciousness. However, the authors argue that our impression of m-consciousness comes from an internal model the brain uses to track its own processing, a model that evolved to be effective rather than accurate, meaning that what it presents is a simplified and cartoonish picture of that processing, one that because of the simplifications, produces something that seems magical and incompatible with i-consciousness, but that nevertheless is a component of it. This description of m-consciousness resonates with illusionist theories.
For the authors, the internal m-consciousness model is the attention schema. However, while they think the attention schema uniquely tracks subjective experience, they admit that the brain probably produces numerous models of different aspects of its processing. All of these models could be considered the higher order representations of HOT.
They also put forth the attention schema as a bridge between HOT and GWT. The global workspace could be seen as the highest level of attention processing, which would make the attention schema essentially a higher order representation of the global workspace.
So we have a multilevel attention mechanism, which culminates in the global workspace, which is modeled by the attention schema, making it a higher order representation of the workspace. The attention schema, along with all the other higher order representations, produces m-consciousness, a simplified cartoonish model that is effective but not accurate, the contents of which we could describe as an illusion. All of this would make up i-consciousness.
My take is that this is a compelling view, although perhaps a bit too slanted toward the AST. For example, I think emotional feelings are higher order representations of lower level reflexive survival circuits, which seem just as central to subjective experience as attention. All of this, to me, strengthens HOT as the more fundamental view of what’s happening. (Which since HOT is a collection of theories, means we’re still far from a final theory, if we indeed ever get to one final one.)
But the convergence of these theories, if they are in fact converging, is starting to look like the rough outlines of a standard model, a collection of theories that together provide an account of the mind. Only time and additional research will tell if it actually is.
Unless of course, I’m missing something?