How to Curate Your Information Inputs for Better Thinking

The problem of information overload is usually framed as a volume problem. There is too much information available; how do you manage the volume? This framing leads to productivity-adjacent solutions: RSS aggregators, read-later apps, newsletter digests, AI summaries. All of these address the volume problem by reducing the amount you have to engage with. None of them addresses the more fundamental problem: the quality of what you are selecting from.

Curation as a cognitive practice is about quality, not quantity management. It is asking not "how do I consume less" but "what should I be consuming in order to become better at thinking?" These are different questions with different answers.

The Cognitive Architecture Argument

To understand why curation matters at the level it does, you need a basic model of how thinking works.

Thinking is not computation on raw data. It is pattern-matching, analogy-drawing, and model-applying. When you encounter a problem, your brain searches for patterns in existing experience and knowledge that the problem resembles, applies models from those domains, and generates solutions by analogy. The quality of this process depends entirely on the richness and diversity of the pattern library you have built.

A person who has read deeply in history, economics, biology, and psychology has a richer pattern library than one who has stayed within a single domain. When they encounter a novel problem, they have more models to draw on, more analogies to attempt, more structural patterns to recognize. This is why generalists often outperform specialists at strategic problem-solving even when the specialists have more domain-specific knowledge: the generalist's broader pattern library enables connections the specialist cannot see.

This has a direct implication for curation: the intellectual range of what you consume is a direct input to your cognitive range. Narrow inputs produce narrow thinking, regardless of the depth within that narrow range.

The Signal Quality Problem

Not all information is equal in its contribution to thinking quality. Understanding the dimensions of signal quality allows you to evaluate potential information sources more accurately.

Explanatory power. Does this source explain why things happen, or only report that they happened? Reporting provides facts. Explanation provides models. A source that consistently offers structural explanations for events — not just "X happened" but "X happened because of these mechanisms operating in this context" — builds your explanatory capability over time. A source that only reports is providing data without the framework for making sense of it.

Falsifiability. Does this source make predictions that can be tested? Does it update when its predictions turn out to be wrong? Sources that never make falsifiable predictions and never acknowledge error are not producing knowledge — they are producing unfalsifiable narratives. Your information diet should include sources with track records you can evaluate, not just sources whose positions feel true.

Epistemic standards. What counts as evidence in this source? What kinds of claims require what kinds of support? A source that treats anecdote as equivalent to systematic data, or that presents correlation as causation without qualification, is not modeling good epistemic standards. Consuming it regularly trains you to accept lower epistemic standards in your own thinking.

Temporal depth. Does this source situate current events and phenomena within longer historical context? Short-horizon sources produce short-horizon thinking. Sources that consistently bring historical perspective to current events build a temporal depth of understanding that improves pattern recognition across time.

Designing the Diet

A deliberately curated information diet has architecture — intentional composition across several dimensions.

Foundational versus current. Foundational material is the works — books, papers, essays — that have shaped a domain's understanding over decades or centuries. Current material is what is being produced and discussed now. Both matter, but most people's diet is overwhelmingly current. Foundational material builds the vocabulary and conceptual frameworks through which current material is interpreted. Without a sufficient foundation, current material is processed without context. The ratio should be closer to even than most people's actual consumption.

Within-field versus cross-field. Within-field material deepens expertise in domains where you need depth. Cross-field material — reading biology when you work in finance, reading military history when you work in software — builds the pattern library that enables analogical thinking. The cross-field material is frequently described by people who think well as disproportionately valuable: it consistently provides frameworks that illuminate their core domain in ways that staying within the domain does not.

Affirmative versus challenging. Affirmative material confirms and refines positions you already hold. Challenging material contests them. Both are necessary. Without affirmative material, you lose the depth and confidence in your core positions that makes them actually useful. Without challenging material, you develop intellectual brittleness — positions that feel strong but have not been stress-tested. The challenging material is the harder discipline; it requires actively seeking out well-argued positions you disagree with and engaging with them seriously.

Fast versus slow. Fast material — news, newsletters, social media — is processed quickly and continuously. Slow material — books, long-form essays, dense technical work — requires sustained engagement over days or weeks. Fast material keeps you current; slow material builds depth. Without slow material in the diet, your understanding of everything is at the fast-material level of processing: broad, current, and shallow.

The Concept Collection Practice

One of the highest-value practices associated with deliberate curation is active concept collection. Rather than simply consuming sources and hoping that useful ideas stick, the concept collector actively identifies and records mental models, frameworks, and structural insights encountered across sources.

The collection is not a reading log. It is an organized repository of concepts that have proven useful in thinking about how things work. When you encounter a concept from evolutionary biology that explains an organizational dynamic, it goes in the collection. When you find a historical pattern that recurs across different civilizations in similar circumstances, it goes in the collection. When you read a framework that clarifies a class of problem you frequently encounter, it goes in the collection.

Over time, this becomes a personalized library of thinking tools. Consulting it when facing a problem produces cross-domain insights that would not be available without the collection practice. It also reveals gaps — domains where your concept library is thin — which then inform your curation decisions.

Pruning as Practice

Curation is not only about what to add. It is equally about what to remove. Information sources that are not contributing to thinking quality consume the finite time and attention available for those that would. Identifying and removing low-value sources is as important as identifying and adding high-value ones.

The assessment framework for removal: Has consuming this source changed my thinking in the past six months? Has it introduced a concept or framework I have actually used? Has it provided information that turned out to be accurate and actionable? Has it challenged an assumption that needed challenging? If the answers are consistently no, the source is occupying space without earning it.

Removal is resisted by two forces: habit and FOMO. The habit is mechanical — you have been checking this source for years, and stopping feels like a loss even when the marginal value is near zero. The FOMO is cognitive — the fear that the one important thing you miss will be the thing you most needed. Both of these forces are manageable when the removal is based on an honest assessment of past value rather than speculation about future potential.

The Meta-Cognitive Feedback Loop

The ultimate test of your curation is not the quality of your inputs — it is the quality of your thinking. Periodically, assess whether your thinking has improved: Are you making better predictions? Recognizing patterns earlier? Generating more creative solutions? Building better arguments? Finding more elegant connections between domains?

If yes, the curation is working. If not, the diet needs revision. This feedback loop — curate, think, assess, revise — is Law 5 applied to the cognitive infrastructure of the self. The mind you think with is built from what you read, watch, and discuss. It is worth building deliberately.