Predictive Theories (Top-Down)

Top-down predictive theories highlight brain-based, central-to-peripheral, efferent influence on sensory organs more than peripheral-to-central, afferent sensory perceptions—and while top-down predictive models may or may not be themselves explanations of consciousness, they give insight into the nature of consciousness and its evolutionary development. Top-down is a fundamental principle of how brains work, and it would be surprising if it were not relevant for understanding consciousness.

According to Anil Seth and Tim Bayne, there are two general approaches to understanding consciousness via the centrality of top-down signaling in shaping and enabling conscious perception.

The first is reentry theories where recurrent, reentrant pathways are in some sense conscious perceptions—and thus reentry theories are theories of consciousness per se.

The second approach, broadly described as predictive processing, starts instead from a foundation principle of how the brain works—in terms of prediction as a core principle underlying perception, action, and cognition, and therefore does not directly specify theories of consciousness. Nonetheless, the “core claim of reentry theory and predictive processing (PP) is that conscious mental states are associated with top-down signaling (reentry, thick arrows) that, for PP, convey predictions about the causes of sensory signals (thin arrows signify bottom-up prediction errors), so that continuous minimization of prediction errors implements an approximation to Bayesian inference”(Seth and Bayne, 2022).

Reentry Theories and Predictive Processing

Cognitive philosopher Andy Clark puts it succinctly: Rather than your brain perceiving reality passively, your brain actively predicts it. Your brain is a powerful, dynamic prediction engine, mediating our experience of both body and world. From the most mundane experiences to the most sublime, reality as we know it is the complex synthesis of predictive expectation and sensory information, “sculpting” all human experience. Thus, the extraordinary explanatory power of the predictive brain (Clark, 2023).

Leveraging the work of Karl Friston (9.5.4), Clark states that in predictive processing, perception is structured around prediction, which he suggests is the fundamental operating principle of the brain (Musser, 2023a, Musser, 2023b).

While the rudimentary evolutionary driver of the predictive brain is simply survival, staying alive, the emergence of consciousness can be seen as facilitating the predictive capabilities in terms of awareness, responsiveness, and conformity to external realities.

Extended Mind and Evolutionary Perspectives

Clark stresses that even though biological brains are increasingly cast as “prediction machines” this should not constrain us “to embrace a brain-bound ‘neurocentric’ vision of the mind.” The mind, such views mistakenly suggest, consists entirely of the skull-bound activity of the predictive brain, an inference from predictive brains to skull-bound minds that Clark rejects.

Predictive brains, he argues, can be apt participants in larger cognitive circuits. The path is thus cleared for a new synthesis in which predictive brains act as entry-points for extended minds (9.7.1), and embodiment and action contribute constitutively to knowing contact with the world (Clark, 2017a; 2017b.)

Cognitive psychologist Richard Gregory pioneered conceptualizing the brain as actively shaping perception, not the assumed inert receptacle of sensory signals. (Gregory himself credited Herman von Helmholtz for realizing that “perception is not just a passive acceptance of stimuli, but an active process involving memory and other internal processes.”)

Gregory's key insight was that “the process whereby the brain puts together a coherent view of the outside world is analogous to the way in which the sciences build up their picture of the world, by a kind of hypothetico-deductive process.” Although timescales differ, Gregory advocated the guiding principle that perception shares processes with the scientific method.

In particular, Gregory incorporated “explicitly Bayesian concepts” into our understanding of how sensory data is combined with pre-existing beliefs ("priors") to modify and mold perceptions. Consciousness evolved, according to Gregory, to enable rapid comparisons between real-world events and counterfactual simulations in order to make optimum decisions (Gregory, 2023).

Neuroscientist Rodolfo Llinás traces the evolution of the "mindness state" to enable predictive interactions between mobile creatures and their environment, arguing that the nervous system evolved to allow active movement in animals. Because a creature must anticipate the outcome of each movement on the basis of incoming sensory data, the capacity to predict is most likely the ultimate brain function. Llinás even suggests that Self is the centralization of prediction (Llinás, 2002).