Consciousness and the fallacy of misplaced objectivity

Abstract

Objective correlates-behavioral, functional, and neural-provide essential tools for the scientific study of consciousness. But reliance on these correlates should not lead to the 'fallacy of misplaced objectivity': the assumption that only objective properties should and can be accounted for objectively through science. Instead, what needs to be explained scientifically is what experience is intrinsically-its subjective properties-not just what we can do with it extrinsically. And it must be explained; otherwise the way experience feels would turn out to be magical rather than physical. We argue that it is possible to account for subjective properties objectively once we move beyond cognitive functions and realize what experience is and how it is structured. Drawing on integrated information theory, we show how an objective science of the subjective can account, in strictly physical terms, for both the essential properties of every experience and the specific properties that make particular experiences feel the way they do.

Keywords: consciousness; contents of consciousness; functionalism; integrated information theory; space.

Figure 1.

Panels A to F portray the visual experience of an observer looking at a dark screen on which dashed ovals appear. The illustration serves to guide the reader through introspecting the phenomenal structure of space, as described in the text.

Figure 3.

Extendedness in phenomenology and cause–effect structures. A) The phenomenal properties that characterize spatial extendedness and B) the physical properties that correspond to them. The polytope in the bottom center depicts the unfolded cause–effect structure of a seven-unit grid. The distinctions “d,” “cd,” “de,” and “cde” are singled out to illustrate the four types of causal relations that make up extendedness. Cyan edges indicate partial overlap between causes and/or effects, purple edges asymmetric full overlap, and magenta edges symmetric full overlap [see Haun and Tononi (2019)]. “Reflexivity”: each phenomenal spot overlaps itself; similarly, distinctions in the cause–effect structure fully overlap (by virtue of the overlap between their cause and effect). “Connection”: spots partially overlap other spots, and the overlap corresponds to another spot; similarly, distinctions partially overlap with their causes and effects, and the overlap corresponds to the cause and effect of other distinctions. “Fusion”: the union of two connected spots corresponds to another spot; similarly, the union of two connected distinctions is also a distinction. “Inclusion”: every spot includes and is included by other spots; similarly, every distinction includes and is included by other distinctions. For a more detailed analysis, see a companion paper (Grasso et al. 2021).

Figure 2.

Unfolded cause–effect structures for three “brain-like” physical substrates. A) A highly integrated eight-unit network in state ABcdEfgH (where uppercase indicates ON and lowercase OFF) unfolds into a rich cause–effect structure with many distinctions and relations. The cause–effect structure of this eight-unit network is meant to illustrate a small part of a much larger cause–effect structure (grayed out in background) that is presumably specified by the physical substrate of human consciousness. The overlay highlights how a distinction (brown) specified by a mechanism (black) comprises a cause (red) and/or an effect (green). Relations among causes and/or effects (overlaps among them) are shown as lines, faces, and volumes, with colors denoting different types of relations. The figure indicates that during wakefulness, the physical substrate of consciousness—constituted of a large number of neurons arranged as “pyramids of grids” and located predominantly in posterior cortical areas (outlined in blue)—can specify a cause–effect structure with high Φ (Tononi et al. 2016). B) When eight units are connected pairwise as parallel modules with minimal intermodular connectivity, the network specifies four separate, minimal cause–effect structures, each with very low Φ. Modular connectivities are found, for instance, in the cerebellum. C) Even when units are interconnected as in A, their causal interactions can be disrupted by changes in neuromodulation and excitability (dashed connections) that lead to a breakdown of causal interactions, severely reducing the network’s capacity for integrated information. As a consequence, the single, rich cause–effect structure of high Φ (Panel A) “disintegrates” into multiple disjoint ones, each with low Φ (Panel C). According to IIT, this breakdown in the capacity for integrated information, consistent with the breakdown of perturbational complexity, accounts for the loss of consciousness during deep sleep early in the night (Massimini et al. 2005; Pigorini et al. 2015). While the purpose of this figure is purely illustrative, the cause–effect structures shown were obtained by computationally unfolding the associated networks of units in their particular state, obtaining their irreducible causal mechanisms with their causes, effects, and relations [for details, see Haun and Tononi (2019)].