Cortical Columns and Our Thoughts

How Millions of Microprocessors Shape the Architecture of Consciousness

The Hidden Parliament of Thought

There are millions of cortical microcircuits behind your every conscious thought.

Every conscious thought you’ve ever had—whether a flash of insight, a pang of doubt, or a moment of moral reckoning—has passed through a vast and silent parliament. Not metaphorically vast, but anatomically: your neocortex contains millions of cortical columns, each a microprocessor of meaning. These columns do not merely relay sensory data. They interpret, compare, anticipate, and negotiate.

(Note: Early estimates placed the number of cortical columns around 100–200k. More recent anatomical analyses indicate the cortex contains millions of smaller functional columns. The conceptual point remains the same: thought emerges from vast distributed microcircuits.)

Thought, in this view, is not a single spark but a layered convergence. It begins with raw input—light, sound, touch—and ascends through tiers of interpretation and integration. At the summit of this pyramid lies the prefrontal cortex: the brain’s executive suite, where competing signals are weighed, goals are set, and decisions are made.

This essay explores how cortical columns shape our conscious experience—not just as passive filters, but as active agents of cognition. We’ll trace their layered roles, map their convergence in the prefrontal cortex, and examine how the brain’s bilateral architecture creates a dynamic tension between intuition and deliberation.

*** image Pyramid of Cognition ***

The Pyramid of Cognition offers a high-level map of this terrain as a structural hierarchy of imperatives, emotions, and goals that scaffold cognition itself. Conscious thought occupies only the capstone, where working memory and executive function reside. But the real work begins below, after sensations become perceptions and our bodily needs are reshaped into emotions, the penultimate meanings are drawn in the cortical columns that populate the neocortex and deliver their messages to the executive areas in the prefrontal lobes.

These cortical columns—each a six-layered microprocessor—do not merely relay sensory input. They receive sparse semantic content which they aggregate with memory, emotion, context, and prior outcomes as they travel the neural pathways. Within this architecture, they perform specialized operations—detecting features, recognizing patterns, anticipating outcomes, sequencing events, and encoding memory. Their outputs are sculpted with additional meaning, provided by the emotional and motivational currents flowing upward through the pyramid.

When this meaning is delivered to the prefrontal cortex, the right and left lobes often have distinct interpretations. One seeks precision and alternatives with facts at hand; the other responds with immediacy and emotional salience, ignoring inconvenient details. The prefrontal cortex adjudicates between the two thought modes.

Cortical Columns

Ninety percent of the cortex shares a repeating six-layer microcircuit, replicated millions of times across the cortical sheet. Cortical columns receive much of their input as categorized patterns from other cortical areas. The input is aggregated with content of memory, knowledge, emotions, and previous results, not merely sensory data.

Though the six-layer cortical is from a lab animal, it vividly shows the six layers we also have. These layers allow columns to operate both locally (within its neural pathway) and globally (with feedback loops).

It also reveals that the number of connections within a layer is fixed but can vary. James Kalat ¹ mentions that “Prefrontal neurons have up to 16 times as many dendritic spines as neurons in the primary visual cortex. … The result is that the prefrontal cortex can integrate an enormous amount of information.”

That’s the reason we’ll discuss the special role of the prefrontal cortex in more detail in a subsequent essay.

Semantic Toolkit

Cortical columns provide a semantic toolkit of seven general functions. Each column has a relatively isolated package of input it is responsible for and transforms their input into sculpted meaning to pass on a neural pathway towards the prefrontal executive area, its executive suite.. The cortical column functions are automatic, distributed, and shaped by both structure and experience. Let’s begin with mechanistic clarity—mapping each function to its laminar origin and cognitive role.

• Contextual Modulation (Layer I + Layer VI feedback loops)
  Inputs from other cortical and subcortical areas modulate columnar activity, adjusting interpretation based on emotional tone, social context, or motivational state. Columns adjust their response based on neighboring activity or top-down input.

• Feature Detection (Layer II)
  Neurons in Layer II specialize in identifying salient features—edges, contrasts, deviations from expectation. This is the column’s initial signal flagging system.

• Temporal Sequencing (Layer II + Layer III)
  Neurons across these layers encode event order and rhythm, supporting causal inference and narrative structure.

• Pattern Recognition (Layer III → Layer V)
  Horizontal connections across Layer III support comparison and integration, enabling recognition of familiar configurations. This function contextualizes current input against stored patterns.

• Predictive Coding (Layer III + Layer V projections)
  Columns generate anticipatory models based on partial input, projecting likely outcomes. These signals often ascend toward orbitofrontal regions for emotional weighting.

• Memory Encoding (Layer V → Subcortical targets)
  Deep-layer neurons project to hippocampal and thalamic structures, contributing to long-term encoding and retrieval.

• Decision Biasing (Layer VI → Thalamus → PFC)
  Columns influence executive processing by biasing attention, salience, and response selection. These signals shape the input received by the prefrontal cortex.

The outputs are not decisions but semantic interpretations, delivered upward for use by the lobes of the prefrontal cortex.

Interhemispheric Integration via the Corpus Callosum

Although cortical columns in each hemisphere primarily process local inputs, the corpus callosum serves as a vital bridge—enabling dynamic exchange between hemispheres. As the brain’s largest white matter tract, spanning roughly 10 cm and comprising 200–300 million axonal projections, it orchestrates the interhemispheric transfer of sensory, motor, and associative data.

This exchange isn’t merely passive relay—it supports real-time comparison, substitution, and integration of semantic patterns and verbal constructs. Whether parsing a word or refining a perceptual schema, each hemisphere can adjust its internal model based on input from the other. Crucially, this negotiation often occurs midstream: within cortical columns or intermediary layers, before content reaches the prefrontal cortex for higher-order synthesis.

Although cortical columns in each hemisphere primarily process local inputs, the corpus callosum serves as a vital bridge—enabling dynamic exchange between hemispheres. As the brain’s largest white matter tract, spanning roughly 10 cm and comprising 200–300 million axonal projections, it orchestrates the interhemispheric transfer of sensory, motor, and associative data.

This exchange isn’t merely passive relay—it supports real-time comparison, substitution, and integration of semantic patterns and verbal constructs. Whether parsing a word or refining a perceptual schema, each hemisphere can adjust its internal model based on input from the other. Crucially, this negotiation often occurs midstream: within cortical columns or intermediary layers, before content reaches the prefrontal cortex for higher-order synthesis.

Once bilateral sensory inputs are aligned—visual fields merged, tactile comparisons resolved—the corpus callosum remains active, facilitating semantic integration across hemispheres. This same white matter bridge enables patterns to be matched with memory, and words to be reconciled with conceptual knowledge. Whether resolving linguistic ambiguity or aligning emotional tone with verbal content, the hemispheres negotiate meaning through callosal exchange. The mechanism scales: from sensory coherence in posterior cortices to semantic synthesis in frontal and associative regions, culminating in unified representations within the prefrontal cortex.

Cortical Capstone: The Bilateral Executive

All of the semantic refinement performed by cortical columns—feature detection, pattern recognition, predictive coding, and more—occurs within the layered folds of the cerebral cortex, the brain’s highest cognitive terrain. These sculpted meanings are not decisions. They ascend toward the final adjudicator: the prefrontal cortex (PFC), where executive function resides.

At the apex of the Pyramid of Cognition (image in the first section), a dividing line appears—signaling a fundamental distinction in how the right and Left PFC lobes handle incoming meaning. This division is not merely anatomical; it reflects two cognitive stances, each shaped by its own circuitry, priorities, and temporal orientation. When necessary, working memory mediates between them, allowing deliberation to override impulse or vice versa.

This chart is a summary of the contrary features of the dual thought streams. We will investigate the Hot and Cold, Fast and Slow mechanisms further in the next post.

For now, let’s lay some groundwork for accepting that these thoughts coexist, followed by a couple of examples to show their interaction.

Associative Mindworks is an attempt to balance the scale that Einstein warned about.

“The intuitive mind is a sacred gift and the rational mind is a faithful servant. We have created a society that honors the servant and has forgotten the gift.”

We focus especially on the Right PFC—an intuitive processor that operates through patterns, associations, and emotional salience. This is the same operation built on Neural Thresholds, Almost Gates, and Self Organizing Maps throughout the brain which explains how bodily imperatives are converted into emotions and now cortical columns transforming the world we experience with our emotions into meaningful relationships that we ultimately use to decide how to act.

The Right PFC receives patterns and associations rapidly, in context, and is always ready with an immediate response to the present situation, although by its inherent abstraction, it can ignore important details and offer a bad suggestion. The Left PFC uses language, and rule-based logic making it slow, exact, and always searching for a better solution. It can be stymied by a lack of facts.

Let’s consider the different Prefrontal Cortex network targets that earlier cortical processes along the way to sent forward.

In keeping with the common hemispheric nomenclature—left for language, right for intuition—the following table outlines the known functional distinctions across PFC subregions.

The Right PFC attends to emotional relevance, immediacy requirements, social norms, and risk evaluation. It weighs what feels urgent, appropriate, or dangerous. The Left PFC, by contrast, receives signals for goal setting, alternative evaluations, rule retrieval, and moral reasoning. It is slower, more effortful, and tuned to long-term outcomes.

Together, these lobes form a bilateral executive—one intuitive, one deliberative—each interpreting the semantic drafts delivered by their cortical columns in its own inimitable way.

Example: Successful Behavioral Response

Scenario: After a day at the beach, I returned to my apartment to shower and change. Then I was hungry. I wanted to take my swimming partner to dinner, but she didn’t answer the doorbell. I notice her car is there. She must be back at the beach. I find her there. We return to my car and go out for a relaxing meal.

Abbreviated cognition flow:

3S Imperative: Hunger arises—hours since last meal.

Emotional Desire: Motivation for social connection.

Behavioral Intent: Decide to invite swimming partner.

Sensory Input: Car observed → processed by cortical columns.

Left PFC Evaluation: Registers absence—no answer at door.

Right PFC Inference: Pattern-based conclusion—she’s likely at the beach.

Motivational Override: Emotional salience delays eating → pursue social goal first.

Example: From Unsatisfactory Action to Informed Resolution

Scenario: An incandescent 60-Watt light bulb burnt out in a wall sconce. I wanted a brighter bulb, but the sconce warned it had a 60W maximum. Incandescent bulbs are wasteful, so I bought an LED equivalent to 60W. And, yes, I was disappointed that it wasn’t brighter.

Then in a random article, I read that the warning on older incandescent fixtures was due to the heat the bulbs emitted, not the actual wattage. The 60W maximum actually meant the heat generated by the bulb, not the wattage.

LEDs produce far less heat. I upped the LED to equivalent to 75W. Now, I am happier with the ambient light.

Cognitive Flow

Phase 1: Initial Decision (Unsatisfactory Outcome)

1. 3S Imperative: Need for ambient light → triggers behavioral intent to replace bulb.

2. Memory Retrieval (Columnar): Recollection of fixture’s “60W max” warning.

3. Emotional Desire: Preference for brighter light, tempered by safety concern.

4. Columnar Semantic Integration:

   • Feature Detection: Burnt-out bulb.

   • Pattern Recognition: “60W max” interpreted as wattage limit.

   • Decision Biasing: Safety over brightness.

5. Left PFC Evaluation: Logical adherence to label → selects LED “60W equivalent.”

6. Right PFC Emotional Salience: Hopes for brightness, but accepts constraint.

7. Sensory Feedback: Installed bulb → light is dimmer than desired.

8. Cognitive Dissonance: Emotional disappointment vs logical compliance.

9. Hung Resolution: No immediate corrective action—unsatisfactory but “safe.”

Phase 2: Semantic Update and Behavioral Revision

10. New Semantic Input (Columnar): Article reveals “60W max” refers to heat, not brightness.

11. Columnar Reprocessing:

    • Contextual Modulation: Reframes prior interpretation.

    • Memory Encoding: Updates fixture constraint as thermal, not luminous.

12. Left PFC Reevaluation: Recognizes logical permission to exceed “equivalent wattage” with LED.

13. Right PFC Emotional Relevance: Reignites desire for brighter light.

14. Behavioral Revision: Selects LED “75W equivalent” → installs bulb.

15. Sensory Feedback: Improved ambient light → emotional satisfaction.

16. Memory Consolidation: New rule encoded: “LED wattage ≠ heat output.”

Interpretation

This flow illustrates how cognition is not static. It’s iterative and reentrant. The initial behavior was constrained by a misinterpreted semantic frame. Once new meaning entered the system, the columns reprocessed, the PFC lobes reevaluated, and behavior adapted. The architecture didn’t fail—it evolved.

Closing Thoughts and Further Explorations

This post developed the ways in which cortical columns add cognitive meaning and deliver it up to the executive areas of our brains.

Because we have two techniques, associations in the Right PFC and logic in the Left PFC, to assess our situation and potential responses, it is not always the case that an answer will be found that is satisfactory to both. Yet the two thought modes are not at war, they coexist and overwhelmingly work together.

In the next post, I will go from this post’s gloss on the distinctions between the two thought streams to a fuller explanation. The focus will be on elaborating on the characteristics and interactions of the two thought streams at the highest level of cognition.

Let me close with a question. Since our brains are composed of neurons and they operate with neural thresholds and Almost Gates, where do you think logic comes from? A genetic mutation or a genius cave dweller above a placid river eons ago?

---

Congratulations! You now have the background to appreciate the cognitive hand-and-glove relationship of patterns-and-logic.

---

Citation

Image of Cortical Columns. By Bodysurfinyon - Own work, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=125106144

1. Biological Psychology, James W. Kalat. Thomson Wadsworth. 2004. Belmont, CA 94002