R0 for Ideas

Epidemiologists have a number that captures everything about a disease's spreading power: R0, the basic reproduction number. It answers a simple question: on average, how many people will one infected person infect?

If R0 is less than 1, the epidemic dies out. Each infected person generates less than one new case; the chain of transmission eventually breaks.

If R0 is greater than 1, the epidemic grows. Each infected person generates more than one new case; the infection spreads exponentially until it runs out of susceptibles or something intervenes.

This framework has become so powerful that researchers began asking: can we calculate R0 for ideas?

The answer is yes, sort of. But the complications reveal as much as the calculations.

The Basic Model

For biological pathogens, R0 is determined by three factors:

Transmissibility (β): The probability of infection given contact between an infected and susceptible individual. A pathogen that spreads 50% of the time on contact has higher β than one that spreads 10% of the time.

Contact rate (c): How many contacts an infected person has during their infectious period. More contacts = more opportunities for transmission.

Duration (d): How long an infected person remains infectious. Longer duration = more time to transmit.

R0 = β × c × d

This multiplicative structure creates the range we observe: measles has an R0 of 12-18 (highly transmissible, spreads through respiratory droplets). Ebola has an R0 of 1.5-2.5 (requires bodily fluid contact, but very high case fatality triggers containment behaviors). COVID-19 original strain had an R0 around 2.5-3.5; newer variants pushed toward 5-7.

Can we map these parameters onto ideas?

Idea Transmissibility

What's the "transmissibility" of an idea? It's the probability that someone who encounters the idea will adopt it—or at least hold it long enough to potentially spread it.

This depends on multiple factors:

Simplicity. Simple ideas transmit more reliably than complex ones. "Taxation is theft" transmits better than "the optimal level of taxation depends on elasticities and welfare functions." The lossy telephone game favors compression.

Emotional valence. Ideas with strong emotional content—fear, outrage, awe, humor—are more memorable and more likely to prompt sharing. Neutral information gets filed; charged information gets forwarded.

Social utility. Ideas that make you look good when you share them—funny, insightful, well-informed, morally righteous—have higher transmission rates. Ideas that make you look bad get suppressed.

Narrative fit. Ideas that fit existing mental models transmit better than ideas that require new frameworks. "X confirms what you already believe" is more transmissible than "everything you believe is wrong."

Novelty. Ideas that are surprising or counterintuitive get attention—but only if they're not too counterintuitive. There's an optimal novelty level: surprising enough to be interesting, familiar enough to be understandable.

Some ideas are measles-level transmissible. "This is outrageous" spreads almost perfectly—everyone who hears it passes it on. Others are rabies-level: scary but hard to catch. "Hegel's dialectic illuminates the structure of history" is true or false, but it's not going viral.

Idea Contact Rate

How many "contacts" does an idea-holder have? This maps directly to network connectivity, but with complications.

In the biological case, contacts are physical interactions. In the ideational case, contacts are communication events—conversations, posts, shares, forwards, broadcasts.

Social media radically increased the contact rate for ideas. A pre-internet individual might share an idea with dozens of people through conversation. A post-internet individual can share with thousands through a single tweet.

But the quality of contacts differs:

Strong-tie contacts: Conversations with close friends, family, trusted sources. High engagement, high attention, high probability of adoption.

Weak-tie contacts: Posts seen by followers, articles skimmed, headlines glimpsed. Low engagement, low attention, low probability of adoption.

The internet massively increased weak-tie contacts while potentially decreasing strong-tie contacts. The net effect on R0 is ambiguous: more total contacts, but perhaps lower average transmission probability per contact.

This explains why online ideas spread fast but shallow. The high contact rate drives rapid diffusion; the low transmission quality means most exposures don't stick. Ideas go viral (high contact) without necessarily becoming beliefs (low adoption).

Idea Duration

How long does someone "carry" an idea before they stop spreading it?

For biological pathogens, infectious duration is determined by the disease course. For ideas, the equivalent is something like "interest duration"—how long does the idea remain salient enough that you keep talking about it?

Most ideas have very short durations. Yesterday's trending topic is gone today. The rage-inducing tweet from this morning is forgotten by evening. The contact rate may be high, but the duration is measured in hours.

Some ideas have longer duration:

Identity-constituting beliefs: Ideas that become part of how you see yourself persist. Political identities, religious commitments, foundational worldviews. You don't stop spreading "I'm a conservative" after a week.

Utility-providing ideas: Ideas that solve ongoing problems stick. "How to cook rice" persists because you keep needing to cook rice. "What Trump said yesterday" fades because you don't need it.

Socially reinforced ideas: Ideas your community keeps circulating don't fade. You're re-exposed constantly. The duration extends through repeated social contact.

The viral ideas with highest R0 often have short duration but explosive contact rates—they burn through the population fast. The ideas that actually shape belief systems have longer duration even if their initial spread is slower.

Calculating Idea R0

Some researchers have attempted explicit calculations.

Jonah Berger at Wharton studied what makes content viral. His analysis of New York Times articles found that content was more likely to make the "most emailed" list if it was surprising, emotionally activating (especially anger and anxiety), and practically useful.

The implicit R0 calculation: take an article that's shared by 100 people. Track how many of those shares generate reshares. Multiply out. You can estimate how far an article will spread based on its initial transmission patterns.

Studies of Twitter have found that most tweets have an effective R0 well below 1—they get seen but not retweeted. A small fraction have R0 above 1, and these drive the platform's visible content. The attention economy is an epidemic of the few surviving ideas with R0 > 1.

Meme researchers have tracked the "life cycles" of internet memes. Most memes never reach 1% of their eventual maximum spread—they're the epidemiological equivalent of infections that burn out before becoming epidemics. A few memes achieve massive spread, with R0 calculations in the dozens for their peak period.

Why the Model Breaks

But here's where the epidemiological analogy starts to fail:

Ideas aren't pathogens. You can hold contradictory ideas simultaneously. You can be "infected" by multiple competing ideas at once. The SIR model (Susceptible → Infected → Recovered) assumes discrete states; idea-holding is continuous and combinatorial.

Immunity is weird. For pathogens, infection usually generates immunity—you can't catch measles twice. For ideas, prior exposure doesn't create immunity; it might create either resistance (you've heard this before, it's boring) or reinforcement (you believe this more strongly now). The "recovered" state doesn't map cleanly.

Transmissibility isn't constant. A pathogen's β is largely fixed by biology. An idea's transmissibility depends on context—who's sharing it, what platform, what's competing for attention, what events are happening. The same idea might have R0 of 10 in one week and 0.5 in the next.

Network structure matters differently. For pathogens, the network structure matters (some structures accelerate spread), but the pathogen doesn't care about semantics. For ideas, the meaning of the connection matters. A political idea spreads differently through political networks than through family networks. The content of the idea interacts with the content of the relationship.

Mutation is creative. Pathogens mutate randomly. Ideas mutate creatively—people modify them for effect, clarify them for understanding, distort them for advantage. The mutation process isn't noise; it's signal.

These complications don't invalidate the R0 framework—they complicate it. The framework provides a useful first approximation. It just can't be the whole story.

Super-Spreaders and Amplification

One finding does transfer cleanly: super-spreaders exist for ideas just as they do for pathogens.

In disease epidemics, a small fraction of infected individuals generate a large fraction of secondary cases. The 80/20 rule often applies: 20% of cases cause 80% of transmission.

For ideas, the distribution is even more extreme. A small number of accounts, platforms, and influencers generate most of the idea-spread. The median tweet reaches dozens; the top tweets reach millions. The distribution isn't 80/20—it's more like 99/1.

This means that R0 for an idea depends enormously on whether it reaches the super-spreaders. The same idea with the same intrinsic properties can have R0 of 0.3 or R0 of 30 depending on the initial conditions of who encounters it first.

Public health learned to target super-spreaders for intervention—find them, contain them, vaccinate them. Information hygiene might learn the same lesson: the leverage points for controlling idea-spread are the amplification nodes, not the general population.

Threshold Models Revisited

The simple/complex contagion distinction (from earlier in this series) maps onto R0 in an interesting way.

For simple contagions, R0 calculation is straightforward: transmissibility × contacts × duration. If R0 > 1, spread happens.

For complex contagions—those requiring reinforcement—the R0 calculation changes. It's not enough for an idea to reach new people. It has to reach them multiple times, from multiple sources, to trigger adoption.

This means complex contagions can have R0 > 1 in dense clusters while having R0 < 1 across the network as a whole. The local epidemic sustains; the global epidemic sputters.

Behavior-change ideas often exhibit this pattern. Within communities that adopt them, they spread readily—R0 > 1. But they fail to cross community boundaries—effective R0 < 1 for bridging transmission. The idea survives in pockets but never becomes universal.

This is why some ideas seem persistently "almost viral"—they achieve high local penetration but never break through to the mainstream. Their R0 is geometry-dependent.

Engineering R0

If you understand what determines R0, you can engineer ideas for spread.

Maximize transmissibility: Make the idea simple, emotional, identity-affirming, socially valuable to share. Strip out nuance. Add emotional hooks. Make it flattering to repeat.

Maximize contacts: Target high-connectivity nodes, time release for maximum attention, use platforms that amplify. The same idea launched at 3 AM or 3 PM will have vastly different R0 based on who's online.

Maximize duration: Tie the idea to ongoing concerns, make it identity-relevant, design for repeated exposure through community reinforcement. Ideas tied to group membership persist because the group keeps circulating them.

This is, essentially, what viral marketers and propagandists do. They're engineering R0.

Consider the structure of successful propaganda. It's not complex argument. It's simple, repeated, emotionally charged, identity-affirming. "Lock her up." "Yes we can." "Make America great again." These aren't accidentally effective—they're engineered for high R0. Simple enough to transmit perfectly. Emotional enough to motivate sharing. Identity-constituting enough to persist.

The most successful disinformation campaigns don't rely on sophisticated deception. They rely on high-R0 structure: claims simple enough that they transmit without degradation, emotional enough that people want to share them, identity-confirming enough that repetition feels good rather than tedious.

The ethical implications are obvious. The same techniques that spread useful public health information can spread disinformation. The same understanding that helps good ideas flourish helps bad ideas flourish. R0 is morally neutral. It amplifies whatever gets fed into it.

This creates an uncomfortable asymmetry. Truth is often complex, nuanced, unsexy. Lies can be engineered for maximum spread. In an R0 competition, the epistemically honest are at a structural disadvantage.

The Takeaway

R0 provides a framework for thinking about idea spread: transmissibility × contacts × duration determines whether an idea grows or dies.

But ideas aren't pathogens. They mutate meaningfully. They compete for attention. They require reinforcement for behavior change. The model is useful but incomplete.

What survives translation: the basic insight that spread is exponential when R0 > 1, and small changes in parameters can flip ideas from dying to thriving. The threshold matters. The super-spreaders matter. The network structure matters.

Your ideas are in competition for survival. The ones that persist are the ones with R0 > 1 in their environment. That's not the same as the true ones, or the good ones, or the ones you'd choose if you were choosing consciously.

Further Reading

- Berger, J. (2013). Contagious: Why Things Catch On. Simon & Schuster. - Centola, D. (2018). How Behavior Spreads. Princeton University Press. - Goffman, W., & Newill, V. A. (1964). "Generalization of epidemic theory: An application to the transmission of ideas." Nature.

This is Part 7 of the Network Contagion series. Next: "Synthesis: Engineering What Spreads"