Influence Mapping & Information Cascades

Master the science of viral spread. Learn how information propagates like disease, why false stories spread faster than truth, and the ethical responsibility of network influence.

⏸️ Reflection Pause

You Are What You Consume

Information spreads like disease. Who you're connected to shapes what you believe. Your network inputs shape your worldview like water shapes stone—slowly, invisibly, inevitably.

Questions to Consider:

List the 5 accounts you engage with most. Are they making you wiser or angrier?
If false news spreads 6× faster than truth, what does that say about what we value as a society?
You have influence in your network. What are you amplifying?

In epidemiology, we study how diseases spread through contact networks. In network science, we use the same mathematical models to study how ideas spread. A lie from a trusted friend spreads faster than truth from a stranger.

If you discovered you're a high-influence node in your network, what ethical responsibility would you have? With great reach comes great accountability.

🎮 Game 1: Predict Viral Spread

SCENARIO

Three influencers post the same meme. Each has different network structures. All have 100K followers.

Which post will spread FARTHEST (reach the most total people)?

🎮 Game 2: Identify Seeding Nodes

SCENARIO

You're launching a public health campaign about vaccines. Budget allows you to pay 3 influencers to share your message. Goal: Maximum reach.

Which combination of seed nodes maximizes information spread?

🎮 Game 3: True vs False Cascade Patterns

SCENARIO

MIT researchers analyzed 126,000 news cascades on Twitter (2006-2017). They compared true stories vs false stories.

                    Research Finding (Vosoughi et al., 2018):

                    False news reached 1,500 people 6× faster than true news. Top 1% of false cascades reached 1,000-100,000 people; true news rarely exceeded 1,000.

Why do false stories spread FASTER and FARTHER than true stories?

🎮 Game 4: Stop Misinformation Cascade

SCENARIO

A false rumor ("Vaccine contains microchips") is spreading on Facebook. You're the platform's Trust & Safety lead. Early intervention is critical.

Patient Zero (posted 2 hours ago) ↓ 5 friends shared (1 hour ago) ↓ 47 people shared (30 min ago) ↓ ~500 people currently sharing ↓ Projected: 50,000 in next 6 hours

What's the MOST effective intervention to stop this cascade?

🎮 Game 5: Ethical Influence Campaign

SCENARIO

You work for WHO (World Health Organization). A new Ebola outbreak started in West Africa. You need to spread accurate safety information FAST to save lives.

Which strategy is BOTH effective AND ethical?

📚 Information Diffusion Models

Network scientists use mathematical models from epidemiology to predict how information spreads:

                    1. SIR Model (Susceptible-Infected-Recovered)

                    Susceptible: Haven't seen the information yet

                    Infected: Exposed and actively sharing

                    Recovered: Seen it but no longer sharing

                    Used for: Meme lifecycles, viral content prediction

                    2. IC Model (Independent Cascade)

                    Each infected node gets ONE chance to infect each neighbor.

                    Probability of infection = edge weight (trust/influence).

                    Used for: Social media shares, word-of-mouth marketing

                    3. LT Model (Linear Threshold)

                    Node gets infected when enough neighbors are infected.

                    Threshold = what % of friends must share before you share?

                    Used for: Trend adoption, herd behavior

Key Insight: These models reveal that network structure matters MORE than content quality for predicting spread. A mediocre idea in a well-connected network beats a brilliant idea in an isolated one.

📚 Influence Maximization & Cascade Prediction

The Influence Maximization Problem: Given a network and budget for k seed nodes, which nodes maximize total spread?

This is an NP-hard problem (computationally expensive), but greedy algorithms work well:

Degree Centrality: Pick nodes with most connections (fast but suboptimal)
Betweenness Centrality: Pick bridge nodes (good for reaching different communities)
PageRank: Pick nodes connected to other influential nodes (Google's original algorithm)
Greedy Algorithm: Simulate cascade, pick best node, repeat (slow but optimal)

                    Why Misinformation Spreads Faster (MIT Study, 2018)

                    1. Novelty: False news is more novel → triggers sharing instinct

                    2. Emotion: False news evokes fear, disgust, surprise → higher engagement

                    3. Bots? No! Study found humans spread false news faster than bots

                    4. Verification gap: Fact-checks arrive too late (after cascade peaks)

Ethical Paradox: The same techniques that spread health information can spread conspiracy theories. Your network knowledge is morally neutral—your choices make it good or evil.

🧠 CommDAAF Critical Thinking Checkpoint

Before you continue, reflect deeply on what you've learned. Write thoughtful responses (minimum 20 characters each).

1. How is information spread similar to disease spread (SIR model)? Where does the analogy break down?

0 characters

2. Why do false stories often spread faster than true stories on social media? Reference the MIT 2018 study findings.

0 characters

3. If you discovered you were a high-influence node in your network, what ethical responsibility would you have? How would you use that power?

0 characters

🏆

Lesson Complete!

You've mastered information cascade dynamics. You can now predict viral spread, identify optimal seed nodes, and understand why misinformation spreads faster than truth.

⚡ 0 XP Earned