In Part 1, we examined how external forces – the Firehose of Falsehood and algorithmic amplification – flood our information environment with noise. But there’s a darker truth: even if we could eliminate every bot, every troll farm, every manipulative algorithm, we would still be vulnerable. The most insidious threat to independent thought isn’t external. It’s the architecture of our own minds.
We like to believe we’re rational creatures, weighing evidence and arriving at logical conclusions. This is a comforting fiction. The reality, confirmed by decades of psychological research, is that our brains are riddled with systematic flaws – cognitive biases – that make us predictably irrational. And those who seek to control us know exactly how to exploit them.
The Two Systems: Fast, Wrong, and Confident
Nobel laureate Daniel Kahneman’s groundbreaking work Thinking, Fast and Slow revealed that human cognition operates through two distinct systems:[¹]
System 1 is automatic, intuitive, and effortless, operating below conscious awareness using mental shortcuts. System 2 is deliberate, analytical, and effortful; but also lazy – activating only when necessary.
Here’s the problem: System 1 is always running, making snap judgments before System 2 even wakes up. And because System 2 is cognitively expensive, we default to System 1 most of the time, even when making important decisions. This wouldn’t be catastrophic if System 1 were merely imperfect. But it’s systematically flawed in predictable ways that can be weaponized.
The King of Biases: Confirmation Bias
Of all cognitive biases, confirmation bias reigns supreme. Once we form a belief, our brains become evidence-filtering machines, accepting information that supports our view and rejecting information that challenges it, regardless of quality. Research shows this affects experts across fields, leading to systematic errors even among highly trained individuals.[²]
The Mechanism:
– Selective Exposure: We seek out sources that agree with us
– Selective Perception: We interpret ambiguous evidence as supporting our view
– Selective Recall: We remember hits and forget misses
This creates epistemic closure – a self-sealing worldview where contradictory evidence is automatically rejected. Different groups literally cannot agree on basic facts because they’re operating in parallel realities.
The Supporting Cast: Four More Biases That Amplify the Problem:
Availability Heuristic: If You Can Remember It, It Must Be Common
If you can easily recall instances of something, your brain assumes it’s common. This is why people overestimate plane crash risks (dramatic, memorable) and underestimate heart disease (mundane, forgettable); despite heart disease killing exponentially more people.[³]
The Firehose of Falsehood exploits this directly. By flooding channels with repeated false narratives, propagandists make those narratives easily recalled, which our brains interpret as probably true.
Anchoring Bias: The First Number Wins
We over-rely on the first piece of information encountered (the “anchor”) when making decisions. Research shows that initial values systematically bias estimates toward that anchor, even when arbitrary or irrelevant.[⁴]
Motivated Reasoning: Thinking as a Lawyer, Not a Scientist
Yale Law School professor Dan Kahan’s research revealed a disturbing finding: higher cognitive ability doesn’t make people more objective – it makes them better at rationalizing preexisting beliefs.[⁵]
In one study, participants with strong math skills correctly solved a problem about skin cream effectiveness. But when the same mathematical problem was framed as gun control data, math skills predicted politically motivated wrong answers. Smarter people weren’t more objective – they were better at motivated reasoning.[⁶]
The Backfire Effect: When Corrections Make It Worse
Correcting misinformation can paradoxically strengthen belief in the original falsehood. Recent research suggests this is rarer than initially thought, but when it occurs, it’s devastating.[⁷] Corrections can increase familiarity with misinformation, inadvertently spreading it to new audiences.[⁸]
Case Study: Brexit and the Triumph of Bias Over Evidence
The 2016 Brexit referendum is a limelit example in how confirmation bias shapes political outcomes.
Confirmation Bias in Action: Research revealed that voters exhibited strong confirmation bias, seeking information that reinforced preexisting views.[⁹] A study found that cognitive ability predicted voting patterns, but higher cognitive ability correlated with Remain votes not because smarter people were more objective, but because they were better at constructing sophisticated justifications for their preexisting political identities.[¹⁰]
The £350 Million Lie: The Leave campaigns claim: “We send the EU £350 million a week. Let’s fund our NHS instead” was textbook anchoring bias exploitation. The number was demonstrably false, debunked repeatedly. But the damage was done. The £350 million figure became the anchor that even informed voters couldn’t escape.[¹¹]
Availability Heuristic and Immigration: The Leave campaign flooded media with immigration stories. Paradoxically, areas with the lowest immigration rates voted most strongly for Leave. In the absence of actual experience, voters relied on the availability heuristic: the ease of recalling immigration stories became their proxy for reality.[¹²]
The Ideological Turing Test: The Antidote to Confirmation Bias
Economist Bryan Caplan proposed a brilliant diagnostic for confirmation bias: the Ideological Turing Test.[¹³] The concept is simple but devastating in its implications:
Can you articulate your opponent’s position so accurately that an outside observer couldn’t tell the difference between your summary and their own words?
Most people fail spectacularly. We think we understand opposing views, but what we actually understand is a caricature – a strawman version that’s easy to dismiss. This is confirmation bias in action: we’ve never genuinely engaged with the strongest version of the opposing argument because doing so is cognitively uncomfortable.
The Test in Practice:
-> A progressive should be able to argue the conservative position on immigration so convincingly that a conservative would nod in agreement
-> A conservative should be able to articulate the progressive case for climate action so accurately that a progressive couldn’t identify it as coming from an opponent
The Ideological Turing Test forces System 2 thinking. It requires you to temporarily suspend your identity, step into an opposing worldview, and argue from their values and evidence, not from your strawman version of their position.
Why It Works:
When you can pass the Ideological Turing Test, you’ve proven you understand the opposing view well enough to have genuinely considered it. You may still disagree but your disagreement is now informed rather than reflexive. You’ve broken the confirmation bias loop.
What Could Have Been: Interventions That Might Have Worked
The Brexit outcome wasn’t inevitable. Research suggests several interventions could have mitigated cognitive bias effects:
Pre-bunking Over Debunking: Warning people about manipulation tactics before they encounter them is more effective than fact-checking afterward.[¹⁴] If voters had been educated about anchoring bias and the £350 million tactic before the campaign, they might have been more resistant.
Mandatory Ideological Turing Tests: Imagine if ballot materials required voters to accurately summarize the opposing position before casting their vote. This simple intervention would force System 2 engagement and break confirmation bias patterns.
Diverse Information Diets: Breaking filter bubbles that reinforce confirmation bias through algorithmic transparency requirements.
The Uncomfortable Truth
Here’s what makes cognitive biases so insidious: knowing about them doesn’t make you immune. Cognitive biases aren’t bugs we can patch with education. They’re features of human cognition, evolved over millennia.
The question isn’t whether you have cognitive biases. You do. The question is: Can you pass the Ideological Turing Test for positions you oppose?
If you can’t, you’re not thinking; you’re tribal signaling. Your beliefs aren’t conclusions; they’re identity markers. And that makes you controllable.
The Path Forward: Intellectual Humility as Defense
The Ideological Turing Test is more than a diagnostic; it’s a practice. Regular engagement with steelmanned opposing arguments builds intellectual humility: the recognition that our beliefs might be wrong, our reasoning might be flawed, and our confidence might be misplaced.
Footnotes
[1] Kahneman, D. (2011). *Thinking, Fast and Slow*. Farrar, Straus and Giroux.
[2] Blanco-Donoso, L. M., et al. (2021). “The Impact of Cognitive Biases on Professionals’ Decision-Making: A Review of Four Occupational Areas.” Frontiers in Psychology, 12, 802439. https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2021.802439/full
[3] Hilbert, M. (2012). “Cognitive biases resulting from the representativeness heuristic in operations management: An experimental investigation.” Management Science, 58(7), 1287-1303. https://pmc.ncbi.nlm.nih.gov/articles/PMC6462158/
[4] Bahník, Š., & Strack, F. (2016). “A literature review of a cognitive heuristic: The anchoring effect.” Judgment and Decision Making, 11(4), 311-325. https://www.researchgate.net/publication/370701063_A_literature_review_of_a_cognitive_heuristic_The_anchoring_effect
[5] Kahan, D. M., et al. (2012). “Ideology, Motivated Reasoning, and Cognitive Reflection.” Judgment and Decision Making, 8(4), 407-424. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2182588
[6] Kahan, D. M., et al. (2017). “Motivated numeracy and enlightened self-government.” Behavioural Public Policy, 1(1), 54-86. https://rcgd.isr.umich.edu/wp-content/uploads/2018/07/motivated_numeracy_and_enlightened_selfgovernment.pdf
[7] Swire-Thompson, B., et al. (2022). “The backfire effect after correcting misinformation is strongly associated with reliability.” Journal of Experimental Psychology: General, 151(7), 1655-1665. https://pmc.ncbi.nlm.nih.gov/articles/PMC9283209/
[8] Ecker, U. K. H., et al. (2020). “Can corrections spread misinformation to new audiences? Testing for the elusive familiarity backfire effect.” Cognitive Research: Principles and Implications, 5, 41. https://cognitiveresearchjournal.springeropen.com/articles/10.1186/s41235-020-00241-6
[9] Becker, S. O., et al. (2017). “Who voted for Brexit? Individual and regional data combined.” European Economic Review, 95, 49-67. https://www.sciencedirect.com/science/article/pii/S0176268018301320
[10] Rollwage, M., et al. (2023). “Cognitive ability and voting behaviour in the 2016 UK referendum on EU membership.” PLOS ONE, 18(10), e0289312. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0289312
[11] Hobolt, S. B. (2016). “The Brexit vote: a divided nation, a divided continent.” Journal of European Public Policy, 23(9), 1259-1277. https://www.tandfonline.com/doi/full/10.1080/13501763.2016.1225785
[12] Hobolt, S. B. (2016). “The Brexit vote: a divided nation, a divided continent.” Journal of European Public Policy, 23(9), 1259-1277. https://www.tandfonline.com/doi/full/10.1080/13501763.2016.1225785
[13] Caplan, B. (2011). “The Ideological Turing Test.” EconLog, Library of Economics and Liberty. https://www.econlib.org/archives/2011/06/the_ideological.html
[14] Roozenbeek, J., & van der Linden, S. (2019). “Fake news game confers psychological resistance against online misinformation.” Palgrave Communications, 5, 65. https://www.nature.com/articles/s41599-019-0279-9