05 April 2026
Silicon Canals
Research suggests that high intelligence doesn't protect against bad decisions - it makes people better at constructing convincing justifications for the bad decisions they were already going to make
Here’s a finding that stopped me cold: a 2012 study out of Yale found that people with the highest science literacy weren’t the most aligned on contested facts. They were the most polarized. Their superior reasoning didn’t bring them closer to truth. It made them better at constructing arguments for whatever they already believed. So the smarter you are, the better you are at being wrong—convincingly.
The assumption most people carry around is that intelligence protects you from bad thinking. That if you’re smart enough, you’ll see through your own biases, weigh the evidence fairly, and arrive at the right conclusion. It sounds reasonable. It’s also wrong. A landmark 2012 study published in Nature Climate Change by Yale researcher Dan Kahan and colleagues tested whether higher science literacy and technical reasoning skills would help people converge on the facts about contested issues like climate change. What they found was the opposite: people with the highest degrees of science literacy weren’t the most aligned on the evidence. They were the most polarized. Their superior reasoning abilities didn’t bring them closer to truth; they made them better at constructing arguments that supported whatever they already believed.
The motivated numeracy experiment further illustrated this. Participants were given a dataset and asked to interpret the results. When framed as a study about skin cream effectiveness, people with higher numeracy skills performed well. But when the same data was reframed as evidence about gun control, highly numerate participants failed to outperform others when the correct answer contradicted their political beliefs. They used their superior math skills to find answers that aligned with their identity rather than truth.
Psychologists call this motivated reasoning—the process where desires shape how we interpret information. Intelligence amplifies this: it gives people a larger vocabulary of justifications, making their rationalizations more sophisticated and harder to challenge. The implication is that intelligence becomes a tool for justification rather than inquiry.
Kahan’s research extends beyond politics. In domains like business, relationships, and health, intelligence can be weaponized to support preexisting beliefs. The solution? Science curiosity—the genuine desire to learn something new, especially when it challenges existing views. Curiosity, not knowledge or reasoning ability, consistently resists motivated reasoning.
The article concludes that the smartest move isn’t to trust intelligence but to be suspicious of it, especially when it aligns with what we want to hear.
Related Articles
06 April 2026
Yahoo
South Korea says 'credible intelligence' indicates North Korean leader's daughter is successor
South Korea's spy agency now believes North Korean leader Kim Jong Un's teenage daughter has been positioned as his successor, citing credible intelligence coll...
Read More
05 April 2026
CoinDesk
Drift says $270 million exploit was a six-month North Korean intelligence operation
Attackers posed as a trading firm, met Drift contributors in person across multiple countries, deposited $1 million of their own capital, and waited half a year...
Read More
03 April 2026
富途牛牛
Geekplus-W and Ubtech Robotics have successively introduced chief scientists, marking the entry of embodied intelligence into a 'top-tier brain' arms race.
Recently, Ubtech Robotics (09880.HK) announced a recruitment drive for a Chief Embodied Intelligence Scientist, offering salaries up to 124 million yuan, signal...
Read More
03 April 2026
富途牛牛
Recruiting for the Position of 'Head of Human AI Solutions': Artificial Intelligence is Spawning a New Wave of Professions
The development of artificial intelligence (AI) has raised concerns about it becoming a 'job killer,' but this technology is creating 640,000 new AI-related job...
Read More