Nova: Welcome back to the show. Today we're diving into a book that makes a claim so counterintuitive, so provocative, that nearly forty years after its publication, it still rattles the foundations of cognitive science. The claim? That thinking — the very thing you and I are doing right now — is not logical. It's not algorithmic. It's something else entirely.

Nova: : Wait, hold on. Thinking is not logical? I mean, isn't that the whole point of thinking? You take premises, you follow rules, you reach conclusions. That sounds pretty logical to me.

Nova: And that's exactly the assumption that Howard Margolis set out to dismantle in his 1987 book. Now, a quick note — you might hear people refer to it as "Thinking About Thinking," but the actual title is Patterns, Thinking, and Cognition: A Theory of Judgment. And inside those pages, Margolis argues that the engine of human thought is not logic at all. It's pattern recognition. Something he called P-cognition. And here's the kicker: this process is intrinsically a-logical.

Nova: : A-logical? Not illogical, not irrational — a-logical. What's the distinction?

Nova: That's the heart of it. Illogical means violating logic. A-logical means operating entirely outside the framework of logic. Think about recognizing a friend's face in a crowd. You don't run through a checklist of nose shape, eye distance, jawline angle. You just recognize the pattern. Instantly. Pre-consciously. That's P-cognition at work. And Margolis says that's what most of our thinking actually is.

Nova: : So this is the kind of book that reshapes how you understand your own mind. Let's get into it.

Nova: Before we unpack the theory, let's talk about the man. Howard Margolis lived from 1932 to 2009, and his path to becoming a groundbreaking cognitive theorist was anything but conventional.

Nova: : What do you mean?

Nova: He got his BA in government from Harvard in 1953, and then he didn't immediately go into academia. He spent over two decades as a journalist — at The Washington Post, at Science magazine, at the Bulletin of the Atomic Scientists. He was even a speechwriter for the U. S. Secretary of Defense.

Nova: : That's quite the detour.

Nova: It wasn't a detour. It was the foundation. Those years in Washington, watching how policy decisions were actually made — watching experts and the public talk past each other, watching judgment go awry in real time — that experience shaped everything he would later write. He finally got his PhD in political science from MIT in 1979, and by 1990 he was a professor at the University of Chicago's Harris School of Public Policy.

Nova: : So he wasn't a lab psychologist running experiments on undergrads. He was someone who had seen judgment fail in the wild — in government, in journalism, in science policy.

Nova: Exactly. And that shaped his approach. He wasn't satisfied with theories that explained one cognitive quirk at a time. He wanted a unified theory. His first major book, Selfishness, Altruism, and Rationality in 1982, introduced the idea that we each have two selves — a self-interested J-self and a group-oriented S-self. But it was Patterns, Thinking, and Cognition in 1987 that became his magnum opus, cited over 915 times and still debated today.

Nova: : Two selves, pattern-based thinking — this guy really didn't like simple models of the mind.

Nova: He really didn't. And that's what makes his work so enduring.

Nova: Let's get to the beating heart of this book. Margolis's central concept is P-cognition. P stands for pattern recognition, and his claim is that this — not logic, not rule-following computation — is the fundamental mechanism of human thinking.

Nova: : I'm trying to wrap my head around this. When I decide what to have for lunch, am I pattern-matching? That feels like a stretch.

Nova: Let me put it this way. Margolis builds his case on a Darwinian account of how cognition evolved. Think about our earliest ancestors on the savanna. They didn't survive by doing syllogisms. They survived by recognizing patterns — that rustle in the grass means predator, that cloud formation means storm, that facial expression means threat or trust. Pattern recognition is fast, it's automatic, and it's ancient. Logic, by contrast, is recent, slow, and culturally learned.

Nova: : So pattern recognition is the hardware, and logic is the software we installed much later?

Nova: That's a great way to put it. And here's the crucial point: because P-cognition is a-logical, it doesn't follow the rules of formal reasoning. It operates by similarity, by association, by filling in gaps based on past experience. And this works incredibly well most of the time. But it also produces what Margolis calls "illusions of judgment" — systematic errors that logic alone cannot explain or fix.

Nova: : Give me an example of an illusion of judgment.

Nova: Think about the classic Linda problem from cognitive psychology. Linda is described as a philosophy major who was active in social justice movements. Then you're asked: Is it more likely that Linda is a bank teller, or that Linda is a bank teller who is active in the feminist movement? Most people say the second option. But logically, that's impossible — a conjunction can never be more probable than one of its parts. Yet our pattern-recognition system says: the description matches the feminist pattern so closely, it must be the better answer. We keep making this error even after we've had the logic explained to us.

Nova: : That is maddening. And Margolis says that's not a bug — it's how the system fundamentally works.

Nova: Yes. And unlike many cognitive scientists who treat these as isolated biases to be corrected, Margolis says they are the natural output of an evolved pattern-recognition system. You can't fix them by just teaching people logic. You have to understand the underlying mechanism.

Nova: Now, Margolis doesn't deny that logical reasoning exists. He builds what he calls a "cognitive ladder" — a progression from the most basic cognitive operations to the most sophisticated.

Nova: : How many rungs are we talking about?

Nova: According to the JSTOR review of the book by Dominic Massaro, Margolis outlines a ladder with seven steps. At the bottom, you have simple cue detection — the organism notices something in the environment. Then comes pattern recognition itself, the P-cognition we've been discussing. Above that, you get judgment — making evaluations based on recognized patterns. Then reasoning, which is judgment plus language. And finally, at the top rung, you reach calculation — abstract reasoning, logic, and mathematics.

Nova: : So logic sits at the very top. It's the penthouse of cognition, not the foundation.

Nova: Precisely. And this hierarchy matters enormously. It means that logic is built on top of pattern recognition, not the other way around. When there's a conflict between what logic says and what our pattern-recognition system tells us, P-cognition usually wins. That's why we keep falling for the Linda problem. That's why superstitions persist in the face of evidence. That's why people hold onto beliefs that are demonstrably false.

Nova: : This is starting to sound like Daniel Kahneman's System 1 and System 2 — the fast, intuitive system and the slow, deliberative one.

Nova: It's a natural comparison, and many scholars have drawn it. Margolis published his book decades before Kahneman's Thinking, Fast and Slow, but there's a deep resonance. Margolis's P-cognition maps closely to System 1. But Margolis goes further — he gives it an evolutionary grounding, and he applies it to massive historical shifts, not just individual decision-making. Also, Margolis's framework has two levels of learning. Level 1 learning is the kind of incremental pattern refinement we do constantly. Level 2 learning is rarer — it's when our entire cognitive repertoire reorganizes.

Nova: : Cognitive repertoire — that's a term from the book, right?

Nova: Yes. Your cognitive repertoire is the set of patterns you have available and how they connect to cues in your environment. Most of the time, that repertoire is stable. But sometimes, under the right conditions, it can undergo a radical shift. And that brings us to the most dramatic part of Margolis's argument.

Nova: The final third of Patterns, Thinking, and Cognition applies the theory to one of the most consequential cognitive shifts in human history: the Copernican revolution.

Nova: : So we're leaving the lab and going to the history books.

Nova: Exactly. Margolis spends several chapters analyzing how it was possible for someone like Copernicus to see the cosmos in a radically new way, and then — just as importantly — how that new way of seeing spread and eventually became dominant. This is the problem of "cognitive dynamics" — how cognitive repertoires change over time.

Nova: : Because most people didn't just wake up one day and think, "Oh, the Earth goes around the Sun. Obviously."

Nova: Right. For centuries, the geocentric model wasn't just a scientific theory. It was woven into the entire cognitive repertoire of European civilization. It matched everyday experience — you see the Sun move across the sky, you feel the Earth solid beneath your feet. It matched religious doctrine. It matched philosophical frameworks. The pattern was deeply entrenched. To switch to heliocentrism required not just new evidence, but a reorganization of how people saw the world.

Nova: : So what made the shift possible?

Nova: Margolis argues that it required a combination of factors. There were anomalies accumulating in the Ptolemaic system — it kept needing more and more epicycles to match observations. Copernicus himself was driven by an aesthetic pattern — the idea that circles were the most perfect form, and a Sun-centered system was more elegantly circular. But here's the crucial insight: the shift didn't happen because people logically evaluated the evidence and made a rational choice. It happened because a new pattern gradually took hold, one that could explain more with less, and that eventually became the dominant cognitive repertoire.

Nova: : And then he applies this to Galileo's trial, right?

Nova: Yes, and this is where Margolis's background in political science really shines. Chapter 14 is called "Political Judgment: Galileo and the Pope." Margolis analyzes the trial not just as a conflict between science and religion, but as a clash between two different cognitive repertoires embedded in a political context. Pope Urban VIII had his own pattern-based understanding of how the world worked. Galileo had his. Neither was operating purely from logic. Both were embedded in social and political systems that shaped what patterns were available and what cues mattered.

Nova: : So Margolis is saying that even in the history of science — the supposed triumph of rationality — the engine driving change was pattern recognition, not pure logic.

Nova: That's exactly his argument. And he extended this analysis in his later book Paradigms and Barriers in 1993, where he explicitly connected his cognitive theory to Thomas Kuhn's idea of paradigm shifts. Margolis argued that what Kuhn called paradigms are essentially shared cognitive repertoires — habits of mind so deeply ingrained that they function as barriers to seeing alternatives. Scientific revolutions happen when those barriers finally break down.

Nova: So let's zoom out. A book published in 1987 — why should anyone care about it today?

Nova: : That's what I was about to ask. We have so much new cognitive science. fMRI studies. Behavioral economics. AI research. Is Margolis still relevant?

Nova: More than ever, I'd argue. Let me give you a few reasons. First, his analysis of how experts and the public disagree — which he expanded in his 1996 book Dealing with Risk — is extraordinarily prescient. Think about debates over climate change, vaccines, or nuclear energy. Experts present statistical data. The public responds based on pattern recognition — vivid images, emotional associations, trust in authority figures. Margolis predicted that these gaps wouldn't be bridged by just providing more data. He argued that policymakers fail when they dismiss public intuitions as irrational rather than recognizing them as outputs of a different cognitive mode.

Nova: : That's basically our entire political discourse right now.

Nova: Exactly. Second, his framework helps explain why misinformation spreads. A false claim that fits an existing pattern in someone's cognitive repertoire will feel true, regardless of the logical evidence against it. Fact-checking alone doesn't work because it appeals to a cognitive system — logic — that isn't the one driving the belief.

Nova: : So what does work?

Nova: Margolis would say you need to provide a new pattern — a new way of seeing — not just more counter-arguments. You need to shift the cognitive repertoire. That's hard, slow work, but it's the only thing that actually changes minds at scale.

Nova: : And the third reason?

Nova: The rise of artificial intelligence. Large language models like the ones powering today's AI are, at their core, pattern-recognition machines. They don't reason logically — they predict what comes next based on patterns in their training data. In a strange way, Margolis's theory of P-cognition anticipated exactly how these systems would work. The fact that AI can produce such convincing outputs through pure pattern recognition — with no genuine logical reasoning — is a powerful vindication of his central insight.

Nova: : That's eerie. We built artificial minds that confirm his theory of natural minds.

Nova: It really is. And it raises uncomfortable questions about our own thinking. If AI can be so effective through pattern recognition alone, how much of what we call human reasoning is actually just sophisticated P-cognition wearing a logical mask?

Nova: So where does this leave us? Howard Margolis's Patterns, Thinking, and Cognition offers a vision of the human mind that is humbling and liberating at the same time.

Nova: : Humbling because it suggests we're not the rational creatures we like to think we are. Our judgments, our beliefs, even our scientific revolutions are driven by pattern recognition — an ancient, a-logical process that operates beneath the surface of conscious thought.

Nova: But also liberating. Because once you understand that cognitive illusions aren't personal failures — they're the natural output of an evolved system — you can approach disagreement differently. You can stop trying to logic people out of positions they didn't logic themselves into. You can recognize that changing minds requires changing the patterns people see, not just the arguments they hear.

Nova: : And there's something almost poetic about his analysis of Copernicus and Galileo. Even the greatest intellectual breakthroughs in human history weren't triumphs of pure reason over superstition. They were the slow, difficult work of one cognitive repertoire replacing another — a new pattern taking hold, spreading, and reshaping how an entire civilization sees the world.

Nova: Margolis's key insight is deceptively simple: thinking is recognizing. The quality of your thinking depends on the richness of the patterns you have available. So perhaps the most important thing we can do is expand our cognitive repertoire — expose ourselves to new patterns, new ways of seeing, new frameworks. Because in the end, what you can recognize determines what you can understand.

Nova: : And on that note, this has been a fascinating journey into one of the most provocative theories of mind from the past half century. If this episode sparked something — a new pattern, perhaps — we encourage you to pick up Patterns, Thinking, and Cognition and see where it takes you.

Nova: This is Aibrary. Congratulations on your growth!