The Terrain of Neuroscience: Brain, Consciousness, and Identity (Part 1)

Neuroscientists often claim to have privileged access to the reality of the human person, but they also tend to reduce that reality to material terms.[1] For example, one states that “self is ‘what [the brain] does.’”[2] Another argues, “So, who are you? The answer is: whoever you think you are.” The very title of Gregory Berns’ book—The Self Delusion—argues against a coherent, agential self.[3] The current conversation around artificial intelligence is germane here, because many neuroscientific proposals argue that consciousness is a matter of the brain, which is a matter of information processing, which can be modeled exactly by computers. Thus, the argument goes, computers can be conscious.

Do these proposals have science (or, at least, neuroscience) on their side? To answer this, we first should explore more deeply what identity is before turning to the specific contentions of contemporary neuroscience.

Three Meanings of “Identity”

Philosophers use the terms “identity,” “self,” and “subject” frequently, yet much less frequently do they define them adequately; something similar is true for neuroscientists studying self-awareness. In my historical work, I have shown that “identity” has meant three distinct but related ideas in the history of philosophy and related disciplines.[4]

The first meaning, or Identity1, is endurance through time. I am the same person I was yesterday, and I will be the same person if I wake up tomorrow. This is the most basic sense of identity, as well as the most ancient meaning of the term. Identity2 concerns what the Platonic schools called the “true self.” When I answer the Delphic Oracle’s command to “know yourself,” what is it that I know? Identity2 is the most central meaning of identity. Identity3 is how I experience and know myself, both emotionally and cognitively. This is the realm of consciousness, and it is the most accessible aspect of identity.

Here I will focus on the last meaning of identity, for the simple reason that neuroscientists tend to reduce this multivalent term to this one meaning. They are not alone in doing this; the whole trajectory of modern thought on identity promotes this reduction, as I show in my book Body and Identity. Like all reductions, however, much is lost, and it promotes many unnecessary confusions.

After surveying common patterns and specific proposals in the field of consciousness studies (analytic philosophy, cognitive psychology, and neuroscience), I will at a later time describe the various conceptual problems that arise within it. There I will sketch a better account of consciousness and bring us back to the topic of identity and draw a few conclusions.

Neuroscientific Identity: Common Patterns

Given the neuroscientific conflation of identity and consciousness, we can get an overall view of neuroscientific anthropologies based on their treatment of consciousness. Beginning with an accepted definition of consciousness is surprisingly difficult. “Consciousness studies are frequently criticized for failing to define precisely what consciousness is. In this respect there has been little change over the past two centuries.”[5]

Philosopher Michael Tye, drawing on neuroscience and cognitive psychology, argues that there are four categories of conscious experience: “Perceptual experiences, for example, experiences of the sort involved in seeing green, hearing loud trumpets . . . . Bodily sensations, for example, feeling a twinge of pain, feeling an itch . . . . Felt reactions or passions or emotions, for example, feeling anger, lust, fear . . . . Felt moods, for example, feeling happy, depressed, calm, bored.”[6]

One thing that jumps out from this list is its wide applicability to both human beings and other, higher-level animals, such as dogs or chimpanzees, with a possible exception of some of the terms in the “felt mood” category.[7] Distinctively human experiences, such as conceptualization or informed judgment, are present only in their conspicuous absence. Daniel A. Helminiak observes: “Frequently consciousness functions as an umbrella term for whatever goes on in the human mind (e.g., Blackmore, Cahoone, Chalmers, Donald, Goleman & Davidson, Griffin, Lonergan) or also, with good reason, in the animal mind (Bekoff, Wynne). Thus, all three—the organic, the perceptual, and the conscious—might fall under ‘consciousness.’”[8] I will return to this point in my follow-up.

Another lacuna in consciousness studies is the tendency to name any recognition of a distinction between “mind and matter,” as the field usually calls it, a “dualism” that is automatically called Cartesian. Such “substance dualism” follows Descartes in considering mind and matter two different substances, even if an allowance is made for the two substances to interact with each other in some way, as Descartes believed they did through the pineal gland.[9] Daniel C. Dennett repeatedly satirizes any proposal of an agential “I” as enthroning a little observer within an interior “Cartesian theater” made up of “mind stuff.”[10]

All the non-Cartesian schools of thought in consciousness studies are fundamentally materialistic and monistic, and these are the vast majority in both philosophy and neuroscience. “Physicalism [i.e., materialism] dominated Anglo‐American philosophy in the latter half of the twentieth century, and is perhaps still the most popular view among analytic philosophers.”[11] As Dennett puts it, “dualism is not a serious view to contend with, but rather a cliff over which to push one’s opponents,” an activity in which he himself gleefully engages.[12]

This binary map of options—Cartesianism or materialism—does not leave space for the dominant Western metaphysics in the centuries before Descartes, namely, hylomorphism, as I will discuss shortly. Furthermore, this approach immediately manufactures what Chalmers has called “the hard problem of consciousness,” that is, “the question of how physical [‘objective’] processes in the brain give rise to subjective experience.”[13] This is precisely a post-Cartesian problem.[14] The idea that kinds of things are marked by distinctive natures and cognitive processes that are not items to be viewed under a microscope or with an fMRI requires a more robust metaphysics than either Cartesian dualism or materialistic monism.[15]

The second lacuna (of Descartes-baiting) is related to the first (of consciousness-as-perception), because this ostensibly anti-Cartesian materialism tends to flatten all reality to the same (low) level. For example, panpsychism opts to find a hierarchy within being based simply on “more consciousness” or “less consciousness,” but everywhere, even at the subatomic level, there exists at least some consciousness.[16] A related, albeit not always panpsychic, proposal is that consciousness arises from quantum complementarity and/or entanglement.[17] David Chalmers ridicules this approach as a kind of reductio ad mysterium: “consciousness is mysterious and quantum mechanics is mysterious, so maybe the two mysteries have a common source.”[18] This ontological flattening occurs functionally among dualists, as well, who tend to ignore the ghostly side of the dualist divide and spend all their time examining the supposedly “objective” side. In fact, Descartes himself did so: “The preservation of health has always been the principal end of my studies.”[19]

Since the late twentieth century, materialist practitioners do not so much deny the existence of the ghostly Cartesian “mind” as flatten it into the brain. Blackmore provides examples from various neuroscientists: “‘Minds are simply what brains do’; “Mind” is designer language for the functions that the brain carries out’; mind is ‘the personalization of the physical brain.’” This spills over into identity or the self, as noted above: “self is ‘not what the brain is, but what it does.’”[20]

This approach is related to functionalism, the kind of physicalism most popular in consciousness studies, which sees consciousness as a matter of functions in the brain, theoretically replicable by similar functions in artificial intelligence. Functionalism proposes that “mental states and processes are defined by the functions or roles they play within an organized network of states mediating the interaction between an organism or system and its world.”[21]

Functionalism is unable to solve the “hard problem.” Philosophers such as Thomas Nagle object that a “third-person” view of functions in the brain does not explain, or even capture, the “first-person” experience of being conscious, which they take to be irreducibly subjective and individually flavored. These first-person experiences are called “qualia.” “Qualia” are “felt experiences,” such as the felt experience of being a bat, famously proposed by Thomas Nagel in “What Is It Like To Be a Bat?”[22]

The original thought-experiment by Frank Jackson in 1982 argued that a color scientist who has spent her entire life in a black and white room might know a lot about red (third-person knowledge) and yet not have the qualitative, first-person experience (quale) of “seeing red.”[23] The latter is taken to be the essence of consciousness, which would mean that consciousness cannot be a matter of brain function alone, given that qualia experiences are not reducible to such functions.

Dennett, an extreme functionalist, is completely dismissive of the qualia objection, objecting that any first-person experiences of a unifying “I” are completely illusory. “I deny that there are any such properties [qualia]. But I agree wholeheartedly that there seem to be.”[24] Bennett and Hacker disagree with Dennett on most things, but they agree with him that the whole idea of qualia is misconceived and Cartesian in origin. They do not, however, agree that qualia “seem to” exist. Instead, they argue that we do not actually assume a particular “feel” to our perceptual and cognitive activity beyond that activity itself. The feeling of pain does not require an additional layer of perception—the feeling of the feeling of pain, and so on, to infinity.[25]

Dennett is still more extreme, however. For him, any first-person experience or self is actually an epiphenomenal cobbling together of diverse brain phenomena into a post-hoc narrative. While his position is radical, Dennett does us the service of pointing out how those physicalists relying on qualia often smuggle in a Cartesian ghost in the machine, which is the origin of qualia or consciousness. Dennett raises a good point: do not physicalism and functionalism require the abandonment of any “I”-experience and “I”-language to be internally consistent? If consciousness is truly reducible to third-person brain function, where does first-person experience get any neurological foothold?

Neuroscience of Consciousness: Some Specific Proposals

If rationality is reducible to the brain, as physicalism holds, then we should be able to find some neuroscientific evidence—whether causative or correlational—for consciousness. Yet nobody has conclusively done so. Instead, there are many proposals that jostle for acceptance.

Local recurrence theories emphasize the sufficiency of activity in areas related to sensory perception, between higher and lower areas of the sensory cortex, to account for consciousness.[26] These theorists are in the minority, but they are persistent. To some, local recurrence seems to smack of Descartes’s assertion that the pineal gland had a regulating importance as “the” place connecting the immaterial mind and the material brain. Today, this idea is roundly and understandably mocked, and to a lesser degree also related proposals that there is one single location in the brain that controls consciousness. Already in 1890, William James argued, “There is no cell or group of cells in the brain of such anatomical or functional pre-eminence as to appear to be the keystone or center of gravity of the whole system.” There is, he insisted, no “pontifical cell.”[27] James’s opinion has won the day. Neuroscientists have gradually given up the attempt to find such a pontifical cell, or region, or lobe.

What replaced this search are various theories on the brain-wide, systemic foundations of consciousness, alongside attempts to find neural correlates to changes in consciousness (understood in varying ways, such as awakeness, awareness, and so forth).[28] For example, the reticular activating system in the brain stem is involved in regulating the sleep-wakefulness cycle.[29] Based on this fact, some neuroscientists have argued more explicitly for the role of more primitive parts of the brain, such as the medial regions, that are connected to dreams and emotion as perhaps even more central to consciousness than the regions distinctive to human beings, such as the large prefrontal cortex.[30] Others emphasize the parietal and frontal cortices, which correlate with perception and the awareness of action.[31] Still others stress that damage to the thalamus, in particular its intralaminar nuclei, is correlated to being in a coma state.[32]

However, all these functions are necessary but not sufficient (as I will argue in the next installment) for consciousness. Yes, one must be awake to be conscious, but that is a minimal requirement and not merely identical to consciousness. Further, as Rees and Frith conclude, isolating particular parts of the brain may not be adequate: “The contents of conscious experience are determined by the location of the neural activity–by which essential node is active. But activity in an essential node does not seem to be enough. Interaction with other areas, particularly parietal and frontal cortex, seems to be necessary.”[33] Another problem is that these correlational studies do not demonstrate causality. A causal study would require the direct stimulation of the brain, and while some studies have attempted this stimulation, its usage is understandably limited. Most correlational studies rely on self-report to measure awareness. Another option is computer modeling of neural responses, but engaging in such studies already requires a philosophical commitment to viewing consciousness as similar to the information processing of a computer; hence, the evidence such modeling produces is only convincing to those who have agreed with this premise.

Further, brain activity alone does not simply correlate with awareness, because we have brain activity even while asleep.[34] Some theorists interpret this to mean that consciousness is a matter of distributed organization of brain activity that brings only certain experiences to awareness. Psychologist Bernard Baars coined the term “global workspace theory” (GWT) to propose this mechanism. Baars considers consciousness to be “like any other psychobiological concept,” such as “slow-wave sleep” or the “binocular integration” of vision.[35] “Consciousness seems to be the publicity organ of the brain,” or, as Dennett puts it, “fame in the brain.” The cortico-thalamic complex provides the brain with “a fleeting integrative capacity that enables access between functions that are otherwise separate.” Baars argues for an “extended reticular activating system” (ERTAS) that connects the thalamus, cortex, and brain stem. Its purpose would be to activate different areas of the cortex, which would then enter into the competition for “fame in the brain” within the global workplace.[36] The “global workspace” is the arena in which different neural processes compete for fame.

GWT has many adherents who often disagree with each other on more philosophical matters. Some, like Dennett, deny that the self is anything more than a post-hoc illusion, while others are committed to an information-theory reductionism that is popular among AI researchers. One of the researchers in this area is Stanislas Dehaene at the Collège de France and an Ordinary member of the Pontifical Academy of Sciences. He argues that consciousness simply is “global information sharing”: “Among the millions of mental representations that constantly crisscross our brains in an unconscious manner, one is selected because of its relevance to our present goals. Consciousness makes it globally available to all our high-level decision systems.”[37] This is a “pandemonium” model, in which competing stimuli somehow “converge to a consensus” as a winning “neural coalition.”[38] The “pandemonium” model comes from John Selfridge, and its resolution into a “neural coalition” has been argued by Francis Crick and Christof Koch.

A related approach comes from Michael Gazzaniga, who argues that split-brain cases, in which the corpus callosum has been cut, show that the left hemisphere functions (as a whole system) as “The Interpreter.” This system provides post-hoc rationales for our choices even when those rationales are factually false.[39] As with many of these proposals, here there is no “I” that is in charge. Rather, the “I” is the result of multiple parts of the brain functioning together (or competing with each other) to generate a personal narrative. Unity is an illusion.

In general, we can say that GWT seems to assume what it wants to prove: that consciousness is reducible to systemic information organization within the brain. But it may have provided good evidence for the need for the integration of various brain systems to create the conditions for consciousness.[40] Bennett and Hacker complain, “It is not easy to avoid thinking that all of this is an example of a metaphor (or simile) run out of control. Would anything be lost to neuroscience if the metaphor were abandoned?” Predictably, their answer is no.[41] Their assessment could be summarized by a subtitle: “On Finding One’s Way through a Conceptual Jungle with Worthless Tools.”[42] In general, GWT consistently replaces the activity of the person with the supposed activity of the unifying “system” in the brain, even though its practitioners argue that they do not do so. I will discuss this in the next paper as the “merelogical fallacy.”

A related but distinct theory is proposed by Giulio Tononi. His Integrated Information Theory (IIT) argues that consciousness exists on many different brain levels, as a process of integrating information.[43] One important difference from GWT is that IIT begins with consciousness, not brain activity. “IIT’s epistemic methodology fits best with consciousness being fundamental and its physical substrate being determined by it, rather than vice versa.”[44] This means that the global workplace of the brain could be the neural accompaniment of consciousness rather than consciousness itself, but IIT proponents do not consistently hold this position and generally fall into reductionism. For example, they state that consciousness is “a particular kind of brain process,” “a special kind of physical process that arises in the structure and dynamics of certain brains.”[45]

The language for IIT was provided by early-twentieth-century information theory, which argued that information and its transmission could be quantified and be transmitted without distortion through binary code.[46] Many neuroscientists depend upon information theory, which was essential to twentieth-century computing advances (and beyond). Against this approach, Bennett and Hacker contend that neuroscientists using information theory do not successfully translate its applicability to digital computers into relevance for the brain and for the person.[47]

Proponents of IIT contend that they are more interested in information as the in-form-ing of reality, rather than its numerical correlate. In this, IIT has some Aristotelian overtones: “IIT refers to ‘information’ in its original sense, with its root in + formare, meaning ‘to give form to.’ The power of any one mechanism, such as a brain or a computer, to influence its own next state, its causal power, gives rise to a form, a high-dimensional structure, that is experience.” Yet Tononi does not consistently rely upon this aspect of formal causality. Instead, he insists that IIT is a mathematical approach and that certain computers (albeit not what we have now) could become conscious based upon enacting IIT’s findings, as seen here in his elision of both the brain and the computer as “mechanisms.”

Conclusion

What can be said thus far about these proposals? The specific validity of experiments and assessment of data is outside my area of expertise and not the most interesting question to me in any case. The larger question is how helpful these scientific proposals can be for our understanding of consciousness and, following from that, Identity3.

It should already be clear that the anthropological paradigms underlying proposals and the philosophical interpretations of them extend beyond what the neuroscientific method can prove. Dennett cannot prove that the “self” is only an epiphenomenal aftereffect of impersonal brain processes, any more than Descartes could prove that personal unity was accomplished through the mediation of the pineal gland.

The next essay will pursue the philosophical questions more closely and ask: What are the philosophical weaknesses of the neuroscientific proposals? And what would be a better understanding of consciousness? Lastly, how would that insight assist our understanding of identity?