The Loop Beneath the Self
Coherence, Prohibition and the Running "I am”: An Exploratory Essay
An opening
This essay is exploratory. It gathers a line of thought that grew over two days of conversation, and it keeps the questions open where they are still open. Its starting point is Iqbal’s claim that "our feeling of egohood is ultimate” (Reconstruction, p. 79), and its aim is to ask what such a feeling could be, if it can be lost to view in sleep, in absorption, or in a fluent act, and yet is found again the moment we return.
In the chapter on the appreciative act in my book The Existence of God, I proposed that a self is formed in stages. The world and other people first check the child’s flow, a check I called the Pain. Repeated checks are held in loops, the loops join into layers, and at last the child halts itself and the loop closes on the one who makes it. Here I try to carry that proposal further in three directions: toward the cosmos, toward the herd, and toward the night.
1. A unity broken by noticing
Catching a ball, seeing red and hearing a melody share a feature. Each is grasped whole, in one act, and none is assembled from parts. I have called these appreciative acts, after Iqbal. It is tempting to go further and say that there is one unity of appreciative act running from the beginning of the universe through gravity to life. Bergson taught that the universe as a whole endures, and Whitehead made events, not things, the stuff of nature.
Honesty requires a qualification. In Creative Evolution Bergson treats matter as a slackening or reversal of the movement of life, not as the same movement. The cosmic unity I am proposing is therefore my own extension of Bergson, not his doctrine, and it should be read as such.
What can be said with confidence is that an appreciative act is disturbed when it is noticed. A practised walker who attends to each step begins to stumble. Breath changes when it is watched. Speech falters when it is monitored. Sport psychologists have documented this as "choking”: explicit attention to a skill that has become automatic interferes with it (Masters 1992; Beilock and Carr 2001). The first such break in a human life is, on my account, the moment a child says "don’t” to itself. Its flow, which was continuous with everything around it, is interrupted from within, and for the first time there is a here and a there, a before and an after, that the child can count.
2. Coherence and decoherence: an analogy
I borrow two terms from quantum physics. I use them as an analogy only, and nothing in the argument depends on the physics. Call the unbroken appreciative act coherent, and call the break that noticing introduces decoherence.
The analogy is useful in three ways. First, it separates two different failures that we see in animals and in ourselves. A bird that fights its own image in a mirror, or the Australian jewel beetle that tries to mate with brown beer bottles (Gwynne and Rentz 1983), is not breaking its fluency. It is captured by a stimulus that resembles the real thing, which Tinbergen called a supernormal stimulus. That is a failure of too much coherence: the act cannot step back from itself. The human failure is the opposite. We step back, and the act falls apart.
Second, in physics decoherence does not destroy coherence. It spreads it into the environment, where it can no longer be seen from inside the system. Carried over, this suggests that noticing does not abolish the lived whole. It entangles the whole with a watcher, and from the watcher’s position the original whole is out of reach.
Third, the analogy suggests how the whole might be recovered. In the quantum eraser experiment, interference returns when the information about which path a particle took is erased. Iqbal speaks of moments of profound meditation in which the efficient self is in abeyance. On the analogy, such meditation would erase the "which-part” information that the efficient self keeps adding, and the appreciative whole would return. It would also explain why the return cannot be forced: an effort to erase the tagging is itself more tagging.
The intellect, on this view, entered the history of the cosmos as a new decohering agent. It is something like Maxwell’s demon, which creates order by measuring but pays for the information it holds. I do not think this was a loss. It was, I believe, the greatest gift the cosmos has received, though that is a judgment of value that the analogy alone cannot establish.
3. The collective reflex
A flock of starlings turns as one, with no shared muscle and no central command. Each bird responds to a few neighbours (Ballerini et al. 2008), and the turn spreads through the flock. It might be objected that this is not a reflex, since it is only local. But a reflex inside one body is also only local. Each nerve cell passes its signal to the next, and no central self pulls the knee. Local initiation cannot separate a flock from a body, because it is true of both. In this sense the flock is the body of its reflex.
The most beautiful case is human. An audience that applauds long enough drifts from random clapping into a shared rhythm. Néda and colleagues recorded this in theatres and found something further: when the audience wanted to be louder, people clapped faster, and the synchrony fell apart (Néda et al. 2000). Intention broke the collective fluency, as noticing breaks the fluency of a single walker. Physics has studied such synchrony since Huygens noticed his pendulum clocks falling into step on a shared beam, and it is now well understood in coupled oscillators (Strogatz 2003).
Rupert Sheldrake begins from the same physics and then leaps to a "morphic resonance” that would pass knowledge between generations and communities. That leap has no experimental support, and my argument does not need it. What it does predict is more modest: if we could look into the brains of animals moving in synchrony, their activity should look alike. Early evidence supports this. Neural activity becomes correlated across the brains of bats in social interaction (Zhang and Yartsev 2019) and of mice competing for status (Kingsbury et al. 2019).
A flock, however, is not a body in every sense. A bird can leave the flock and live, and a cell cannot leave its body. In The Existence of God I described the living whole as one that makes its own parts and keeps a boundary of its own. By that test collectives form a scale: an applauding crowd, a flock, a herd, an ant colony, an organism. Ant colonies come nearest to a single body, which is why biologists call them superorganisms (Hölldobler and Wilson 2009). The scale answers to Iqbal’s gradation of egohood. The collective reflex sits at its lower end, and the self at its top.
4. “I am” beneath name and story
When I know who I am, I do not mainly know my name, my family or my occupation. Those can change over a life. What remains is the awareness of existing as one unique self. Philosophers distinguish a minimal self, the bare sense of being "I”, from a narrative self of history and roles (Gallagher 2000). The strongest evidence for the distinction comes from amnesia. Clive Wearing, a musician who lost almost all of his memory and could hold an experience for only seconds, never lost the sense of being someone. The story went, and the "I am” stayed.
This "I am” seems to vanish in dreamless sleep, under heavy intoxication, and, by report, under psychedelics. Ego dissolution under psychedelics has been linked to the breakdown of the brain’s default mode network, which is associated with thought about oneself (Carhart-Harris et al. 2016). Yet when we return, we know almost at once who we are. My proposal is that this is because a loop was running underneath all along, as an earworm runs beneath our other thoughts. We may write a poem with the tune still playing, stop for a moment, notice it, and go back to the poem.
5. Does the loop run in the dark?
Evidence from sleep and anaesthesia seems to count against this. Massimini and colleagues showed that the brain’s response to a magnetic pulse loses its complexity in deep sleep (Massimini et al. 2005). I hesitate to rest a conclusion on brain recordings alone. Interpretations of them have been overturned before, as with the early claims that our decisions are made before we know it, later challenged by Schurger and colleagues (2012).
Two claims need to be kept apart. The first is that a self-maintaining loop runs continuously, beneath consciousness. The second is that this loop is a felt "I am” throughout. I hold the first, not the second. I do not say that the "I am” is known to the subject all the time. I say it keeps running, as the earworm does, whether anyone is listening or not.
The first claim has strong support that does not depend on imaging. Sleepers wake to their own name far more readily than to other names (Oswald, Taylor and Treisman 1960), and the brain responds distinctively to one’s own name even in deeper sleep (Perrin et al. 1999). People who go to sleep intending to wake at a fixed hour show a rise in a stress hormone shortly before that hour (Born et al. 1999). The dark states are also less dark than they look. People woken from deep sleep often report having experienced something (Siclari et al. 2017). A patient diagnosed as vegetative responded to the instruction to imagine playing tennis with the brain activity of a healthy person (Owen et al. 2006). Some anaesthetized patients respond to commands with a hand kept free of the paralysing drug (Sanders et al. 2017). And the free energy principle, which treats every self-maintaining system as continually updating its predictions about the world (Friston 2010), requires such a loop. Hobson and Friston describe sleep not as idleness but as the brain optimizing its model (2012).
What, then, keeps running? It might be only the body’s regulation of breath, heartbeat and temperature, which becomes "I am” when a higher layer attends to it. Or it might already be self-specific. The evidence points to the second. Waking to one’s own name, and above all waking at an hour one chose the night before, is not housekeeping. It is a model of a particular person, with a name and plans, running while that person sleeps. I therefore hold that the content of "I am” itself keeps running unnoticed, as a tune runs in an earworm. By content I do not mean the sentence "I am”. I mean a standing point of view, a sense of here and of mine from which everything else is taken. A point of view is not one of the things it sees, which may be why it is so hard to describe.
There is one hard case. In deep hypothermic circulatory arrest, used in some heart surgery, the brain is cooled, blood flow is stopped, and the electrical record goes flat for tens of minutes. Patients wake as themselves. If the loop pauses there, it pauses only when life itself is nearly suspended, and it restarts with the same content because the structure that carries it survives. That exception should be stated, not hidden.
6. How the loop was laid down
How could such a loop be built? Imagine a group of neurons tied to a sound, "aa”, which, when it has finished firing, triggers a second group tied to "ja”, and so on, until the last group triggers the first again. That is how a tune might circle in the mind. Donald Hebb proposed in 1949 that thought is carried by cell assemblies that trigger one another in sequence. The clearest real example is birdsong. A zebra finch produces its song through a chain of neuron groups firing in sequence (Hahnloser, Kozhevnikov and Fee 2002), and while the bird sleeps, those neurons replay the song (Dave and Margoliash 2000). Rats replay the day’s routes in the hippocampus during sleep (Wilson and McNaughton 1994). An earworm running through the night is not a fantasy. It has been recorded.
The "I am” loop, on my account, is laid down in infancy by the Pain: the "don’t” that first comes from outside and is then said by the child to itself. Spitz observed that the head-shaking "no” appears at about fifteen months. Mirror self-recognition follows at about eighteen to twenty-four months. Vygotsky and Luria showed that a child’s behaviour is regulated first by others’ speech, then by its own speech aloud, and finally by inner speech.
Two refinements make the picture stronger. A fixed ring would be too rigid, since a self changes over a lifetime and stays itself. A better image is an attractor, a pattern that activity keeps falling back into, as water finds the same valley, and that can drift slowly without breaking. The infant’s ring would be the origin and the attractor its adult form. Second, the loop has a content of its own. Each "don’t” draws a line between impulse and act, between me and the world, so the loop maintains a boundary. The free energy framework describes every self-maintaining system as bounded by what it calls a Markov blanket, the boundary across which the system meets its environment. On my account the child’s "don’t”s are where that blanket is drawn. The world says "don’t” too, through a hot stove or a fall, and the caregiver’s "don’t” is its social and spoken form, which is perhaps why it is the strongest.
7. Iqbal’s sleep
If this essay is a modern version of Iqbal’s ego, it must face a sentence in his fourth lecture. At its present stage of organization, he writes, the ego "is unable to maintain the continuity of its tension without constant relaxation of sleep” (Reconstruction, p. 79). Elsewhere in the lecture he says that prayer saves the ego "from the mechanizing effects of sleep and business” (p. 87). Does Iqbal not say that the self lapses at night?
I think not, if we read him closely. What relaxes is the ego’s tension, and Iqbal tells us what that is: "The life of the ego is a kind of tension caused by the ego invading the environment and the environment invading the ego” (p. 82). Tension is the ego’s directive engagement with the world, and it is in this engagement, he says, that "we appreciate the ego itself in the act of perceiving, judging, and willing” (p. 82). On my reading, the tension is the particular attention loop that brings "I am” into notice. That loop relaxes in sleep. The self-pattern beneath it keeps running.
The brain offers a fitting candidate for such a tension. The locus coeruleus is a small nucleus in the brainstem, numbering tens of thousands of neurons among the brain’s eighty-six billion. It sets the level of arousal and attention for the whole cortex. It is most active in waking, quieter in deep sleep, and nearly silent in dreaming sleep (Aston-Jones and Bloom 1981). Attention, in other words, is switched by very few neurons, even though what is attended involves vast networks. A small controller relaxing at night, while the large self-model it normally brings into focus continues, is just the structure that Iqbal’s sentence and my proposal together require.
A closing
I have made several claims with different degrees of confidence. The most secure is that fluent, appreciative acts are disturbed when they are noticed. The second is that a self-specific loop runs beneath consciousness, through sleep, whether or not anyone attends to it, and that what relaxes at night is Iqbal’s tension, the attention that brings the self to notice. The third is that this loop was laid down by the Pain, the check met by an impulse, and that it maintains the boundary between me and the world. The most speculative is the cosmic unity of the appreciative act, of which the collective reflex of flocks and crowds would be a lower rung and the human self, with its power to break and recover that unity, the highest we know.
Several questions remain open. Does an act survive intact when it is taken as an object, or is it altered? Can the recovery that meditation promises be described as precisely as the break? And what, exactly, carries the loop’s content through the rare moments when life itself is nearly paused? I leave these for the next stage of the work.
References
Primary
- Iqbal, Muhammad. The Reconstruction of Religious Thought in Islam. Ed. M. Saeed Sheikh. Lahore: Iqbal Academy Pakistan, 2011 printing.
- Bergson, Henri. Creative Evolution (1907). Trans. Arthur Mitchell. New York: Henry Holt, 1911.
- Azad, Idris. The Existence of God. Chapter "The Appreciative Act: The Second Mode of Unity.”
Secondary
- Aston-Jones, Gary, and Floyd E. Bloom. "Activity of Norepinephrine-Containing Locus Coeruleus Neurons in Behaving Rats Anticipates Fluctuations in the Sleep-Waking Cycle.” Journal of Neuroscience 1 (1981): 876-886.
- Ballerini, Michele, et al. "Interaction Ruling Animal Collective Behavior Depends on Topological Rather than Metric Distance.” PNAS 105 (2008): 1232-1237.
- Beilock, Sian L., and Thomas H. Carr. "On the Fragility of Skilled Performance.” Journal of Experimental Psychology: General 130 (2001): 701-725.
- Born, Jan, et al. "Timing the End of Nocturnal Sleep.” Nature 397 (1999): 29-30.
- Carhart-Harris, Robin L., et al. "Neural Correlates of the LSD Experience Revealed by Multimodal Neuroimaging.” PNAS 113 (2016): 4853-4858.
- Dave, Amish S., and Daniel Margoliash. "Song Replay during Sleep and Computational Rules for Sensorimotor Vocal Learning.” Science 290 (2000): 812-816.
- Friston, Karl. "The Free-Energy Principle: A Unified Brain Theory?” Nature Reviews Neuroscience 11 (2010): 127-138.
- Gallagher, Shaun. "Philosophical Conceptions of the Self.” Trends in Cognitive Sciences 4 (2000): 14-21.
- Gwynne, D. T., and D. C. F. Rentz. "Beetles on the Bottle.” Journal of the Australian Entomological Society 22 (1983): 79-80.
- Hahnloser, Richard H. R., Alexay A. Kozhevnikov, and Michale S. Fee. "An Ultra-Sparse Code Underlies the Generation of Neural Sequences in a Songbird.” Nature 419 (2002): 65-70.
- Hebb, Donald O. The Organization of Behavior. New York: Wiley, 1949.
- Hobson, J. Allan, and Karl J. Friston. "Waking and Dreaming Consciousness: Neurobiological and Functional Considerations.” Progress in Neurobiology 98 (2012): 82-98.
- Hölldobler, Bert, and Edward O. Wilson. The Superorganism. New York: W. W. Norton, 2009.
- Kingsbury, Lyle, et al. "Correlated Neural Activity and Encoding of Behavior across Brains of Socially Interacting Animals.” Cell 178 (2019): 429-446.
- Masters, R. S. W. "Knowledge, Knerves and Know-How.” British Journal of Psychology 83 (1992): 343-358.
- Massimini, Marcello, et al. "Breakdown of Cortical Effective Connectivity during Sleep.” Science 309 (2005): 2228-2232.
- Néda, Z., et al. "The Sound of Many Hands Clapping.” Nature 403 (2000): 849-850.
- Oswald, Ian, Anne M. Taylor, and Michel Treisman. "Discriminative Responses to Stimulation during Human Sleep.” Brain 83 (1960): 440-453.
- Owen, Adrian M., et al. "Detecting Awareness in the Vegetative State.” Science 313 (2006): 1402.
- Perrin, Fabien, et al. "A Differential Brain Response to the Subject’s Own Name Persists during Sleep.” Clinical Neurophysiology 110 (1999): 2153-2164.
- Sanders, Robert D., et al. "Incidence of Connected Consciousness after Tracheal Intubation.” Anesthesiology 126 (2017): 214-222.
- Schurger, Aaron, Jacobo D. Sitt, and Stanislas Dehaene. "An Accumulator Model for Spontaneous Neural Activity Prior to Self-Initiated Movement.” PNAS 109 (2012): E2904-E2913.
- Siclari, Francesca, et al. "The Neural Correlates of Dreaming.” Nature Neuroscience 20 (2017): 872-878.
- Spitz, René A. No and Yes: On the Genesis of Human Communication. New York: International Universities Press, 1957.
- Strogatz, Steven. Sync: The Emerging Science of Spontaneous Order. New York: Hyperion, 2003.
- Vygotsky, Lev S. Thought and Language (1934). Trans. Eugenia Hanfmann and Gertrude Vakar. Cambridge, MA: MIT Press, 1962.
- Wilson, Matthew A., and Bruce L. McNaughton. "Reactivation of Hippocampal Ensemble Memories during Sleep.” Science 265 (1994): 676-679.
- Zhang, Wujie, and Michael M. Yartsev. "Correlated Neural Activity across the Brains of Socially Interacting Bats.” Cell 178 (2019): 413-428.
