A personal anecdote on why the facts don’t really matter
Recently, I wrote three posts on why human reasoning has not evolved to pursue truth. Rather our minds were designed to argue a ‘side’ irrespective of the facts. In a sense, the facts don’t really matter.
- Do The Facts Really Matter?
- Confirmation bias as a feature, not a bug: The facts don’t really matter
- How Political Bias Alters Economic Understanding
I came face-to-face with this phenomenon earlier today. My wife, a trained Montessori teacher and school director, was trying to show me some results from studies about video games and anti-social behaviour. She often claims that studies demonstrate that increased video game usage is highly correlated with perceived anti-social behaviour in boys. I am always very sceptical of these studies because I remember playing a
decent number lot of video games and I know my friends did as well and I never considered any of us to be anti-social (well almost any of us).
So, she read me this from a presentation handout from the recent American Montessori Society (AMS) conference in Chicago called “Boys Will Be Boys”. The handout reads:
“Every investigator who has correlated the amount of time that a child or adolescent or young adult spends playing video games with that student’s academic performance has found a negative correlation.” – Sax
“The strength of the evidence linking video games to antisocial behavior is every bit as strong as the evidence linking second-hand smoke to lung cancer, or lead paint poisoning in infancy to lower IQ scores.” – Sax
My reaction: Bollocks! Not true, total rubbish. At least that is what I was telling myself. She knows how I react to these findings so she expected as much. But I know full well she’s right:
Confirmation bias, where we look for data points to bolster our argument, is a feature not a bug. It helps us to win arguments irrespective of the facts on the ground. In essence, the facts don’t really matter.
So I admitted to her that I was biased because of my own life experiences and that I was willing to cede the point. I did make some mitigating qualifiers in defense of video games, of course; it’s hard to completely capitulate. But after a few weak arguments, I conceded the point.
My take on the matter: confirmation bias is a heavy, heavy behind the scenes influencer of everything we do and every values-related opinion we have. If someone argues against my values (i.e. the innate brilliance of boys who played Donkey Kong, Dig Dug, Pac Man or Tempest), they are likely to get a BS response based on arguments (like the huge hand-eye coordination benefits of gaming) rooted in confirmation bias. I am no different than anyone else.
So I was thinking about this when I came across this chart from an old post of mine on the correlation between healthcare spending and life expectancy. Take a look:
The US is a huge outlier here. How do you explain this chart? And don’t give me the BS response based on your pre-conceived value judgments. I’m on to that.
Healthcare spending and life expectancy: a comparison of graphs – Andrew Gelman, Columbia University