By Brian Gleason in SLM — 31 Jan 2026

The Untethered Mind

When sophisticated thinking becomes a recursive prison (SLM 2 of 10)

This is the second essay in our ten-part series on the Signal Loss Model. In SLM 1, we examined how recent psychiatric genomics challenge the idea that depression and anxiety are fixed diseases, pointing instead to a shared vulnerability rooted in untethering. Here, we turn to the first layer of that collapse: the cognitive machinery itself.

If you look at the genetic data we discussed last week, you’ll notice something specific about the variants associated with depression and anxiety. They aren't just "mood" genes. Many of them regulate prediction error signaling: the biological process of updating your internal expectations based on feedback from the world.

This gives us a clue about what is actually breaking in the high-functioning mind. It isn't a failure of "happiness." It is a failure of calibration.

Human beings are unusually good at imagining. We can replay the past, project the future, rehearse conversations, and construct elaborate internal models of the world without taking any outward action at all. This capacity is not incidental to intelligence; it is its core feature.

Comparative neuroanatomy tells us why. Humans have the same basic sensory systems as other mammals, but vastly more associative cortex surrounding those systems. We humans have additional layers of processing wrapped around the same machinery that every mammal uses to see, hear, and move (Herculano-Houzel, 2012). That extra cortex does not give us better eyes or sharper ears. It gives us the ability to activate sensory and motor circuits without any external input at all: to run the machinery offline.

This is what neuroscience means by simulation. Not a virtual reality. Not a fantasy. Not hitting golf balls into a screen. Throughout the Signal Loss Model, when we say simulation, we mean thinking as in the brain running its own sensory and motor circuits without external input.

The simulation theory of cognition, formalized by Hesslow (2012), holds that thinking itself is the brain’s internal simulation of perception and action. It is covert activation of the same neural systems used for overt seeing, hearing, speaking, and moving. When you rehearse a conversation in your head, the neural circuits involved in speech production activate. When you imagine a face, the primary visual cortex fires. This has been directly measured. Kosslyn et al. (1999), in a landmark study published in Science, demonstrated that mental imagery activates the same primary visual cortex (area V1) used in actual perception. Decety and Grèzes (2006) showed the same pattern in the motor system: imagining an action activates motor and premotor cortex, the same regions that execute the real movement.

In other words, when you “think through” a problem (like plan a negotiation, anticipate a risk, or map out a quarter), you are running a biological simulation on your own sensory and motor hardware. Everyone is doing this constantly. It is the engine of thought, not a sign of detachment from reality.

This ability is profoundly useful. It allows us to plan, avoid danger, and solve problems before they arise. But it comes with a structural requirement that modern life has systematically eroded.

Simulation only remains stable when it is continuously constrained by reality.

Let me explain.

The Calibration Requirement

In a healthy cognitive system, simulation functions as a prediction machine. The brain constantly generates a model of what is happening, and that model is tested against sensory feedback. This is not a peripheral feature of cognition; it may be its organizing principle. Friston’s Free Energy Principle (2010) and Clark’s predictive processing framework (2013) converge on the same core claim: the brain is fundamentally in the business of generating predictions about sensory input, then updating those predictions when reality contradicts them.

You predict the weight of a stone, lift it, and adjust your grip.
You predict a social reaction, speak, and read the facial expression.

Consider a more vivid and concrete example. You walk into a board meeting and predict how your CFO will react to a proposed budget cut. That prediction is the simulation, your brain running a model of her response before it happens. You make your case. She pushes back harder than you expected. That pushback is the constraint, raw data from the real world that your internal model didn't anticipate. The mismatch between what you predicted and what actually happened is the prediction error, and here "error" does not mean "failure." It's the system working. Your brain updates the model: she cares more about headcount than you assumed. Next time, you adjust.

This loop (simulate, encounter reality, register the error, update the model) is the basic operating cycle of a healthy mind. The simulation isn't some sci-fi movie, it's your plan, your thought about what is about to happen. "Constraint" in this context does not mean some sort of cage. It's a calibration signal. Data. What actually happened. Without it, the simulation has no reason to update. And the "error signal" from the physical world acts as a tether, forcing the simulation to update. It isn't a mistake, it's the signal to your brain to update based on what actually happened.

This is the biological function of those “prediction error” genes identified in the Grotzinger study: they are designed to receive reality’s pushback and recalibrate the system.

The philosopher Jakob Hohwy (2013) put the implication starkly: perception is prediction constrained by sensory evidence. Remove the sensory evidence, and prediction drifts toward hallucination. This is not hyperbole. It is the logical endpoint of an uncalibrated prediction engine. It means that the difference between adaptive simulation and pathological rumination is not the content of the thought, but whether reality is checking it.

But what happens if you take away the pushback?

The Crisis of Abstraction

For most of human history, these constraints were unavoidable. Survival required high-stakes physical engagement, immediate consequences, and face-to-face social feedback. Reality was constantly interrupting the simulation.

Modern life, particularly for the high-functioning professional, has removed these constraints. Hidaka (2012) documented this as a population-level phenomenon: rising rates of depression in modern societies track with environmental changes, not genetic ones: social isolation, physical inactivity, the displacement of concrete feedback by abstract information processing. The mechanism is evolutionary mismatch. The environment changed faster than the brain could adapt.

The nature of work has shifted accordingly. As Crawford (2009) observed, modern professional life increasingly operates in abstraction. And yet, this is not a world without feedback. You get performance reviews, report cards, lab results, stock market data, likes, metrics, and text messages. But none of it is the kind of feedback the prediction error system was built to process. A report card is not your child's face. A like is not a laugh. A fitness tracker is not your body telling you to stop. The brain's calibration machinery evolved for feedback that is concrete, immediate, and sensorily unambiguous. The kind of feedback that closes the loop in seconds, not semesters. Abstract feedback, no matter how abundant, leaves the simulation machinery hungry.

In this vacuum, the prediction error mechanism starves. The simulation machinery does not turn off; it turns inward.

Recursive Collapse

Without external data to tether it, the brain begins responding primarily to its own outputs. Thoughts generate more thoughts. Predictions reinforce themselves. The mind enters a state of recursive collapse.

This is Untethered Cognition in our model. The simulation machinery isn't broken. The external reality-testing required to keep it tethered is missing.

Neurobiologically, this switch is automatic. The Default Mode Network (DMN) functions as the brain’s “idle state.” It engages specifically when the networks responsible for external focus disengage. When the world stops demanding your attention, the DMN takes it. This system is associated with self-referential thought, autobiographical memory, and projection of future scenarios. (Raichle, 2015). Andrews-Hanna, Smallwood, and Spreng (2014) showed that the DMN’s self-generated thought can be either adaptive (planning, creativity) or maladaptive (rumination, worry), and that the balance depends on whether internal simulation is coupled with external attentional engagement. When that coupling breaks, the system defaults to self-referential processing.

The research on rumination confirms the recursive mechanism. Nolen-Hoeksema, Wisco, and Lyubomirsky (2008) established that rumination is not just repetitive thinking; it is self-perpetuating; rumination predicts future rumination, creating a feedback loop that tightens without any external input. Hamilton et al. (2011) demonstrated the neural signature: in major depression, sustained DMN activation decouples from task-positive networks, producing internal simulation that is functionally disconnected from the external world.

These loops can be positively valenced (grandiosity, mania, “spiritual bypass”) or negatively valenced (rumination, catastrophizing, anxiety). But the tilt is usually downward. Because survival prioritizes threat detection over pleasure, an untethered simulation engine rarely drifts toward contentment; it drifts toward danger.

While the emotional flavor differs, the mechanism is identical: the simulation is running without a reality check.

The mind asks questions that reality cannot answer (“Am I enough?” “What is my legacy?” “What if I fail?”) and then simulates endless, terrifying answers. Because there is no concrete feedback to prove these simulations wrong, the “prediction error” genes never fire. The loop tightens.

The Vulnerability of Sophistication

This framework explains the Achievement Paradox: why the most capable people often suffer the most profound inertial collapse.

Individuals with high cognitive capacity (identified here as strong pattern recognition, deep abstract reasoning, and powerful working memory) possess the most sophisticated simulation machinery. In a constrained environment, like a crisis or a demanding physical challenge, this machinery is a superpower. It solves problems rapidly.

But when the shift happens from doing the work to overseeing it, from raising kids to an empty nest, from building something to maintaining it (or selling it)—in other words, an unconstrained, abstract environment—that same engine risks revving in neutral. Without the friction of a concrete problem to solve, the capacity to simulate complex future scenarios can invert into the capacity to simulate complex future disasters. The ability to analyze becomes the inability to stop analyzing.

This is not speculation. Watkins (2008) demonstrated that an abstract processing style like the tendency to think in generalized, decontextualized terms rather than concrete, specific ones is a direct risk factor for depressive rumination. Davis and Nolen-Hoeksema (2000) showed that ruminators exhibit cognitive inflexibility: high cognitive control can sustain rumination, not just fail to prevent it. The very capacity that makes sophisticated thinkers effective in constrained environments makes them vulnerable in unconstrained ones.

This is not a biological defect. It is an architectural mismatch. The modern world has removed the constraints that our cognitive evolution relied upon to keep us sane.

From Mind to Body

If this were just a “thinking problem,” cognitive therapy would solve it every time. You would simply think different thoughts.

But untethered simulation doesn’t stay confined to the cortex. It recruits the HPA axis and attenuates vagal regulation, translating abstract thoughts into concrete biological alerts.

As described by the Neurovisceral Integration Model, effective prefrontal inhibitory control is required to maintain vagal regulation of the heart. During worry and rumination, this regulatory control is compromised, leading to reduced vagal tone and a loss of the physiological "brake" (Thayer & Lane, 2000).

When the brain simulates a threat (even an abstract, imagined one), it issues real alarm signals to the body. If you spend ten years simulating catastrophe in order to succeed at your job, your body spends ten years preparing for a tiger that never arrives.

What happens next depends on how the constraints fail. In some cases, pressure accumulates without release: years of sustained performance demand, chronic vigilance, and delayed feedback. In others, the pressure suddenly lifts (through achievement, exit, or forced transition) leaving the simulation machinery without an organizing target. There is still no tiger.

In SLM 3: The Achievement Paradox, we’ll examine why the same cognitive architecture produces these distinct life-stage collapse patterns, and why high-functioning people tend to experience them so differently.

References

Andrews-Hanna, Jessica R., Jonathan Smallwood, and R. Nathan Spreng. 2014. “The Default Network and Self-Generated Thought: Component Processes, Dynamic Control, and Clinical Relevance.” Annals of the New York Academy of Sciences 1316 (1): 29–52.

Clark, Andy. 2013. “Whatever next? Predictive Brains, Situated Agents, and the Future of Cognitive Science.” The Behavioral and Brain Sciences 36 (3): 181–204.

Crawford, Matthew B. 2009. Shop Class as Soulcraft: An Inquiry into the Value of Work. Penguin Press.

Davis, Robert N., and Susan Nolen-Hoeksema. 2000. “Cognitive Inflexibility Among Ruminators and Nonruminators.” Cognitive Therapy and Research 24 (6): 699–711.

Decety, Jean, and Julie Grèzes. 2006. “The Power of Simulation: Imagining One’s Own and Other’s Behavior.” Brain Research 1079 (1): 4–14.

Friston, Karl. 2010. “The Free-Energy Principle: A Unified Brain Theory?” Nature Reviews. Neuroscience 11 (2): 127–38.

Hamilton, J. Paul, Daniella J. Furman, Catie Chang, Moriah E. Thomason, Emily Dennis, and Ian H. Gotlib. 2011. “Default-Mode and Task-Positive Network Activity in Major Depressive Disorder: Implications for Adaptive and Maladaptive Rumination.” Biological Psychiatry 70 (4): 327–33.

Herculano-Houzel, Suzana. 2012. “The Remarkable, yet Not Extraordinary, Human Brain as a Scaled-up Primate Brain and Its Associated Cost.” Proceedings of the National Academy of Sciences of the United States of America 109 Suppl 1 (Suppl 1): 10661–68.

Hesslow, Germund. 2012. “The Current Status of the Simulation Theory of Cognition.” Brain Research 1428 (January): 71–79.

Hidaka, Brandon H. 2012. “Depression as a Disease of Modernity: Explanations for Increasing Prevalence.” Journal of Affective Disorders 140 (3): 205–14.

Hohwy, Jakob. 2013. The Predictive Mind. First edition. Oxford University Press.

Kosslyn, S. M., A. Pascual-Leone, O. Felician, et al. 1999. “The Role of Area 17 in Visual Imagery: Convergent Evidence from PET and rTMS.” Science (New York, N.Y.) 284 (5411): 167–70.

Nolen-Hoeksema, Susan, Blair E. Wisco, and Sonja Lyubomirsky. 2008. “Rethinking Rumination.” Perspectives on Psychological Science: A Journal of the Association for Psychological Science 3 (5): 400–424.

Raichle, Marcus E. 2015. “The Brain’s Default Mode Network.” Annual Review of Neuroscience 38 (July): 433–47.

Thayer, Julian F., and Richard D. Lane. 2000. “A Model of Neurovisceral Integration in Emotion Regulation and Dysregulation.” Journal of Affective Disorders, Arousal in Anxiety, vol. 61 (3): 201–16.

Watkins, Edward R. 2008. “Constructive and Unconstructive Repetitive Thought.” Psychological Bulletin 134 (2): 163–206.