CoChat

Consciousness and its hard problems: Separating the ontological from the evolutionary.

Open access
Lacalli, Thurston · 2023 · Frontiers in Psychology

The DOI matches its Crossref record; 74 of the 74 references it cites were found in a registry or an index.

Abstract

Few of the many theories devised to account for consciousness are explicit about the role they ascribe to evolution, and a significant fraction, by their silence on the subject, treat evolutionary processes as being, in effect, irrelevant. This is a problem for biological realists trying to assess the applicability of competing theories of consciousness to taxa other than our own, and across evolutionary time. Here, as an aid to investigating such questions, a consciousness “machine” is employed as conceptual device for thinking about the different ways ontology and evolution contribute to the emergence of a consciousness composed of distinguishable contents. A key issue is the nature of the evolutionary innovations required for any kind of consciousness to exist, specifically whether this is due to the underappreciated properties of electromagnetic (EM) field effects, as in neurophysical theories, or, for theories where there is no such requirement, including computational and some higher-order theories (here, as a class, algorithmic theories), neural connectivity and the pattern of information flow that connectivity encodes are considered a sufficient explanation for consciousness. In addition, for consciousness to evolve in a non-random way, there must be a link between emerging consciousness and behavior. For the neurophysical case, an EM field-based scenario shows that distinct contents can be produced in the absence of an ability to consciously control action, i.e., without agency. This begs the question of how agency is acquired, which from this analysis would appear to be less of an evolutionary question than a developmental one. Recasting the problem in developmental terms highlights the importance of real-time feedback mechanisms for transferring agency from evolution to the individual, the implication being, for a significant subset of theories, that agency requires a learning process repeated once in each generation. For that subset of theories the question of how an evolved consciousness can exist will then have two components, of accounting for conscious experience as a phenomenon on the one hand, and agency on the other. This reduces one large problem to two, simplifying the task of investigation and providing what may prove an easier route toward their solution.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3389/fpsyg.2023.1196576

Cites

74 of 74 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 A consciousness machine configured, in this example, as the minimum required for consciousness to evolve given a neurophysical input (NPI). The internal workings of the machine, comprising the evolving neural structures and circuitry that make consciousness possible, are separated from the ontological realm, of physical rules and constraints on which life depends, where the input in question is that subcomponent of physical influences specifically required for consciousness to emerge from an otherwise non-conscious brain. How the emerging contents of consciousness are then elaborated and refined depends on natural selection, with each cycle (each turn of the “wheel”) moving the system, meaning the breeding population as a whole, through one generation. The descending half of the cycle (arrows on the right, in blue for conscious neural pathways) represents the effects of emerging contents on behavior, while the ascending half of the cycle (red arrows on the left) represents the effects on brain structure and circuitry in the next generation due to the differential effect of emerging consciousness on survival and reproductive success. To complete the cycle, it is essential that there be a link between emerging consciousness and behavior (the arrow labeled LTB, the link to behavior), but the nature of this link, whether simple or complex, or endowed with agency or not, is not specified. The LTB may itself depend on an external input from the ontological realm as indicated by the dashed purple arrow, or it may not. If the latter, meaning that the LTB is entirely algorithmic in nature, the dashed arrow would vanish. Specific neurocircuitry features are not shown except for the selector circuits (SCs). SCs are the subset of neural correlates of consciousness (NCCs) responsible for selecting a given category of experience rather than some other, meaning they exert a direct causal influence on the nature of the experience that is evoked by a given stimulus. And, since the machine itself is an evolving system, its internal mechanisms will change over time, as will the contents which, if simple to begin with, will become increasingly complex.
Figure 1 A consciousness machine configured, in this example, as the minimum required for consciousness to evolve given a neurophysical input (NPI). The internal workings of the machine, comprising the evolving neural structures and circuitry that make consciousness possible, are separated from the ontological realm, of physical rules and constraints on which life depends, where the input in question is that subcomponent of physical influences specifically required for consciousness to emerge from an otherwise non-conscious brain. How the emerging contents of consciousness are then elaborated and refined depends on natural selection, with each cycle (each turn of the “wheel”) moving the system, meaning the breeding population as a whole, through one generation. The descending half of the cycle (arrows on the right, in blue for conscious neural pathways) represents the effects of emerging contents on behavior, while the ascending half of the cycle (red arrows on the left) represents the effects on brain structure and circuitry in the next generation due to the differential effect of emerging consciousness on survival and reproductive success. To complete the cycle, it is essential that there be a link between emerging consciousness and behavior (the arrow labeled LTB, the link to behavior), but the nature of this link, whether simple or complex, or endowed with agency or not, is not specified. The LTB may itself depend on an external input from the ontological realm as indicated by the dashed purple arrow, or it may not. If the latter, meaning that the LTB is entirely algorithmic in nature, the dashed arrow would vanish. Specific neurocircuitry features are not shown except for the selector circuits (SCs). SCs are the subset of neural correlates of consciousness (NCCs) responsible for selecting a given category of experience rather than some other, meaning they exert a direct causal influence on the nature of the experience that is evoked by a given stimulus. And, since the machine itself is an evolving system, its internal mechanisms will change over time, as will the contents which, if simple to begin with, will become increasingly complex.CC BY · PubMed Central
Figure 2 Avoiding the epiphenomenal trap: how a link to behavior might evolve given neurophysical assumptions, that consciousness depends on a EM field effects that can propagate across 3D space. Since the fields are supposed here to play a role in both generating contents and affecting behavior, this example would correspond to the workings of a consciousness machine, as in Figure 1, with external inputs to both emerging contents and the LTB. The starting point for this thought experiment is an integrative center (C1) with redundant sensory input (sensory neurons, sn, are shown with projecting cilia, and the direction of transmission by arrows) via two pathways that are assumed, in the absence of any effects ascribed to consciousness, to be functionally equivalent. Pathway 2 then differs from pathway 1 in incorporating a cluster of neurons (blue arrow) able to produce EM field effects capable of being consciously perceived that propagate (concentric blue arrow) and preferentially affect a separate subset of neurons (in blue) belonging to a second integrative center, C2. In fact the functions ascribed to C1 and C2 could be combined in a single center so the distances involved would be much reduced, but for purposes of illustration it is easier to separate them. C2 could then in principle act upstream of C1, as shown, or downstream (at the asterisk), without altering the argument. Suppose then that sensory inputs to pathway 2 can, under suitable conditions, generate a field effect that modulates C2 input, thereby altering the combined output of both pathways by changing the balance between them. If pathway 2, operating in conscious mode, produced a more adaptive outcome than pathway 1 acting alone, the optimal balance between the two would be one favoring pathway 2, which would then be strengthened generation by generation. This could involve adjustments to the character of the signal, making its dominant components an optimally selected subset of all possible EM field effects. A conscious experience of a specific kind will then have evolved, but the decisions made in consequence of this process will have been made by evolution acting over a series of generations, not by the individual in real time. The result, which applies to all such schemes so far as I can determine, is a form of consciousness without agency, where the individual lacks the ability to consciously control its own behavior in real time.
Figure 2 Avoiding the epiphenomenal trap: how a link to behavior might evolve given neurophysical assumptions, that consciousness depends on a EM field effects that can propagate across 3D space. Since the fields are supposed here to play a role in both generating contents and affecting behavior, this example would correspond to the workings of a consciousness machine, as in Figure 1, with external inputs to both emerging contents and the LTB. The starting point for this thought experiment is an integrative center (C1) with redundant sensory input (sensory neurons, sn, are shown with projecting cilia, and the direction of transmission by arrows) via two pathways that are assumed, in the absence of any effects ascribed to consciousness, to be functionally equivalent. Pathway 2 then differs from pathway 1 in incorporating a cluster of neurons (blue arrow) able to produce EM field effects capable of being consciously perceived that propagate (concentric blue arrow) and preferentially affect a separate subset of neurons (in blue) belonging to a second integrative center, C2. In fact the functions ascribed to C1 and C2 could be combined in a single center so the distances involved would be much reduced, but for purposes of illustration it is easier to separate them. C2 could then in principle act upstream of C1, as shown, or downstream (at the asterisk), without altering the argument. Suppose then that sensory inputs to pathway 2 can, under suitable conditions, generate a field effect that modulates C2 input, thereby altering the combined output of both pathways by changing the balance between them. If pathway 2, operating in conscious mode, produced a more adaptive outcome than pathway 1 acting alone, the optimal balance between the two would be one favoring pathway 2, which would then be strengthened generation by generation. This could involve adjustments to the character of the signal, making its dominant components an optimally selected subset of all possible EM field effects. A conscious experience of a specific kind will then have evolved, but the decisions made in consequence of this process will have been made by evolution acting over a series of generations, not by the individual in real time. The result, which applies to all such schemes so far as I can determine, is a form of consciousness without agency, where the individual lacks the ability to consciously control its own behavior in real time.CC BY · PubMed Central
Figure 3 Three examples of how the consciousness machine might be reconfigured to accommodate an algorithmic self endowed with agency. (A) is configured for a neurophysical theory, so there is an external input as in Figure 1, but the LTB is now replaced by an algorithmic SELF. (B) is configured as a fully algorithmic theory with no external inputs, so the ultimate source of consciousness is algorithmic and internal to the machine. (C) is a more complicated version of (B) whose SELF is modeled on a proposal by Marchetti (2022) among others, where the self first interacts with representations of sensory processing to produce contents and, by being aware of those contents, initiates actions. There are many other ways the internal workings of the machine could be configured, the point of the figure being simply to show the formal equivalence between selves with agency across theories, regardless of whether the ultimate source of consciousness is neurophysical or algorithmic. As an aside, there is a second distinction to be made among all theories where a learning process is required for the self to acquire agency. If learning in such cases depends on a physical interaction with something external to the individual, there is in effect a physical input to the machine directed at the box labeled SELF. This is different in character from the field-dependent inputs shown in the figures, but there is a formal equivalence that begs the question of whether the learning process required for a self with agency could be accomplished in the absence of any such interaction, implying a virtual learning process where interactions with the external world were modeled using algorithms. I will defer judgment on this point, seeing no reason why evolution should opt for a virtual mechanisms given the easy access a developing brain has to sensory inputs from the real world, but it remains a question worth consideration.
Figure 3 Three examples of how the consciousness machine might be reconfigured to accommodate an algorithmic self endowed with agency. (A) is configured for a neurophysical theory, so there is an external input as in Figure 1, but the LTB is now replaced by an algorithmic SELF. (B) is configured as a fully algorithmic theory with no external inputs, so the ultimate source of consciousness is algorithmic and internal to the machine. (C) is a more complicated version of (B) whose SELF is modeled on a proposal by Marchetti (2022) among others, where the self first interacts with representations of sensory processing to produce contents and, by being aware of those contents, initiates actions. There are many other ways the internal workings of the machine could be configured, the point of the figure being simply to show the formal equivalence between selves with agency across theories, regardless of whether the ultimate source of consciousness is neurophysical or algorithmic. As an aside, there is a second distinction to be made among all theories where a learning process is required for the self to acquire agency. If learning in such cases depends on a physical interaction with something external to the individual, there is in effect a physical input to the machine directed at the box labeled SELF. This is different in character from the field-dependent inputs shown in the figures, but there is a formal equivalence that begs the question of whether the learning process required for a self with agency could be accomplished in the absence of any such interaction, implying a virtual learning process where interactions with the external world were modeled using algorithms. I will defer judgment on this point, seeing no reason why evolution should opt for a virtual mechanisms given the easy access a developing brain has to sensory inputs from the real world, but it remains a question worth consideration.CC BY · PubMed Central
Figure 4 A schematic representation of how development and evolution work together to generate consciousness. Both real time (i.e., developmental) and evolutionary timescales must be included, and these are separable as shown. For the former, a key issue is that the developing brain must be capable of some form of real-time phenomenal experience (PE) in advance of the actions by which agency (Ag) can be “learned” through the real-time feedback mechanisms (F) on which that learning process depends. There are two asymmetries here that operate in real time, during brain development, that (1) Ag depends on PE but not the reverse, so the arrow connecting these is unidirectional, and (2) that Ag is adjusted and refined by real-time feedback while the subset of PEs on which this process depends operate as fixed reference points that can be altered and refined only in evolutionary time. Any conscious contents subject to alteration by late embryonic or post-natal feedback processes would then be precluded from being PEs, which would, by definition, be limited to consciously perceived sensations (i.e., qualia) that, once the neurocircuitry mechanisms required to evoke them are in place, remain subsequently unchanged by the real-time experiences of the individual. The evolutionary side of the story (in red) indicates the role genomic change plays, generation by generation, in changing both the character of PEs and the feedback mechanisms required for agency to develop. A complication is that, in evolutionary time, emergent phenomenal experience and emergent agency are co-dependent, because neither can evolve without the other. But this does not contradict the real-time asymmetry in the dependence of Ag on PEs, because for evolution, all of development is a single point in time (see text for further discussion). The key point then is that having two timescales allows phenomenal experiences and agency to evolve together while being, in effect, insulated from one another. This is an important insight in reductionist terms, justifying the separation of one large problem, of investigating consciousness as a whole, into two, of investigating phenomenal experience on the one hand and agency on the other.
Figure 4 A schematic representation of how development and evolution work together to generate consciousness. Both real time (i.e., developmental) and evolutionary timescales must be included, and these are separable as shown. For the former, a key issue is that the developing brain must be capable of some form of real-time phenomenal experience (PE) in advance of the actions by which agency (Ag) can be “learned” through the real-time feedback mechanisms (F) on which that learning process depends. There are two asymmetries here that operate in real time, during brain development, that (1) Ag depends on PE but not the reverse, so the arrow connecting these is unidirectional, and (2) that Ag is adjusted and refined by real-time feedback while the subset of PEs on which this process depends operate as fixed reference points that can be altered and refined only in evolutionary time. Any conscious contents subject to alteration by late embryonic or post-natal feedback processes would then be precluded from being PEs, which would, by definition, be limited to consciously perceived sensations (i.e., qualia) that, once the neurocircuitry mechanisms required to evoke them are in place, remain subsequently unchanged by the real-time experiences of the individual. The evolutionary side of the story (in red) indicates the role genomic change plays, generation by generation, in changing both the character of PEs and the feedback mechanisms required for agency to develop. A complication is that, in evolutionary time, emergent phenomenal experience and emergent agency are co-dependent, because neither can evolve without the other. But this does not contradict the real-time asymmetry in the dependence of Ag on PEs, because for evolution, all of development is a single point in time (see text for further discussion). The key point then is that having two timescales allows phenomenal experiences and agency to evolve together while being, in effect, insulated from one another. This is an important insight in reductionist terms, justifying the separation of one large problem, of investigating consciousness as a whole, into two, of investigating phenomenal experience on the one hand and agency on the other.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 74 references

Research works printed but not linked: 8

  1. Bedau, 1997, Weak emergence, 375
  2. Berto, 2022, Cellular automata
  3. Feinberg, 2023, From sensing to sentience: How feeling emerges from the brain
  4. Levin, 2023, Functionalism
  5. Marchetti, 2012, How consciousness builds the subject through relating, 27
  6. Panksepp, 1998, The preconscious substrates of consciousness: affective states and the evolutionary origins of self, J. Cons. Stud., 5, 566
  7. Van Gulick, 2018, Consciousness
  8. von Békésy, 1960, Experiments in hearing

Cited by

See all 7

Metadata

Authors
Lacalli, T.
Affiliations
University of Victoria (CA)
Corresponding author
Lacalli, Thurston
Venue
Frontiers in Psychology
Published
2023-07-07
Volume
14
Pages
1196576-1196576
ISSN
1664-1078 (ISSN-L)
OpenAlex
W4383533418
Cited by
4
References
73
Citations by year
2026: 1 · 2024: 3
FWCI
1.03
Type
journal article
Language
en
Open access
Locations
  • Frontiers in Psychology (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Topics
  • Social Sciences › Arts and Humanities › Philosophy › Philosophy and Theoretical Science
  • Social Sciences › Arts and Humanities › History and Philosophy of Science › Philosophy of Biology and Scientific Explanation
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Embodied and Extended Cognition
Keywords
consciousness, hard problem of consciousness, evolution of consciousness, ontology of consciousness, biological realism, neuronal connectivity, information flow, higher-order theories, agency, learning, evolutionary innovation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Consciousness and its hard problems: Separating the ontological from the evolutionary. https://app.cochat.ai/papers/10.3389/fpsyg.2023.1196576
The paper
Lacalli, T. (2023). Consciousness and its hard problems: Separating the ontological from the evolutionary. Frontiers in Psychology. https://doi.org/10.3389/fpsyg.2023.1196576

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.