Confirmation bias

What do JFK, Liberace, Aldous Huxley and C.S. Lewis have in common? I’ll tell you in a moment, but first I want to talk about confirmation bias.

President John F. Kennedy speaks at the Cornerstone ceremony for the Rayburn House Office Building on May 24, 1962. More: go.usa.gov/pvH Credit: Architect of the Capitol

Life appears to be a series of random events, and yet the mind does everything it can to make sense of them. We create patterns where apparently none exist. Psychologists call this ‘confirmation bias’. I first came across it in David McRaney’s fascinating guide to psychology, You are not so Smart.

McRaney describes a situation in which you have a conversation about some old movie, in his example The Golden Child starring Eddie Murphy. Once someone mentions this movie that you have long forgotten, you suddenly start seeing it everywhere. It pops up in a newspaper article. You see a photo of it online. It’s in a pile of DVDs in the shop. Naturally, you assume everything is connected, that this is more than coincidence. McRaney denies that this is true. He describes this situation thus:

The thing is that you disregarded all the other information, all the stuff unrelated to The Golden Child. Out of all the chaos, all the morsels of data, you noticed only the bits that called back to something sitting on top of your brain … Confirmation bias is seeing the world through a filter.

The thing is, of all the brilliant observations in McRaney’s book, that’s one that I don’t believe in. It’s a very convincing explanation and it makes logical sense. However, coincidence clearly does exist. It can’t be explained away as tuning into one particular thing, whilst tuning out all of the noise that otherwise obscures our mental processes.

Take last weekend. One date in particular kept cropping up for me. I was reading up on Doctor Who’s fiftieth anniversary  and I was reminded that the programme was first broadcast on 23 November 1963, one day after the assassination of John F Kennedy.

That same Saturday, I started reading Brave New World by Aldous Huxley. Looking over the biography page, I was astonished to see that Huxley died on 22 November 1963 as well. The obituary pages were rather crowded that day for SF and fantasy fans. With a click of the mouse I soon discovered that C.S. Lewis, creator of Narnia, died the exact same day.

The only person in my opening question who didn’t die on 22 November 1963 was Liberace. However, with the new biopic about his life, I was reading up on the flamboyant pianist on Wikipedia, just moments after putting down Brave New World. What do you know? On 22 November 1963, Liberace collapsed in his dressing room. Terrifyingly, he was suffering acute kidney failure caused by dry-cleaning fumes used in his outrageous costumes. Although he was extremely ill, Liberace did survive, but he too was a person who had faced a crisis event in his life on that particular day, almost fifty years ago.

Call it confirmation bias if you like, but that’s just one too many coincidences for me to encounter in less than twenty-four hours. Unlike the Golden Child example above, I wasn’t looking specifically for that historical date and I certainly didn’t encounter any others over that Saturday. Yet it seemed that everything I was looking at pointed to that same day: 22 November 1963, when Aldous Huxley, C.S. Lewis and John F Kennedy all died.

Advertisements

4 responses to “Confirmation bias

  1. An interesting blog. On 11 December 1963 I left school and on the following monday started work aged 15 as a junior clerk at a timber importers (J.S.Towell) in Boston – only a few days after the day in your blog.

    Dad

  2. ‘coincidence clearly does exist. It can’t be explained away as tuning into one particular thing, whilst tuning out all of the noise that otherwise obscures our mental processes.’

    I’m guessing that McRaney provides evidence for his assertions in the form of Random Control Trials that are peer-reviewed. Have you any evidence against his evidence to support the assertion you make above? All you seem to have done is listed some coincidences. This is a pity, because it would be interesting to know:-
    1) why coincidences ‘can’t’ be explained by confirmation bias?
    2) what your alternative view – ‘too many coincidences’ – means. i.e. too many for what?
    3) how many coincidences can an explanation of coincidence be limited to, and why?

    • Hi Jeff. Interesting questions! I’m not an expert in any way here, although as you can see I have been thinking about this a lot.
      To answer your first point, it’s not that I feel coincidences can’t be explained in any way by confirmation bias. It’s rather that I feel confirmation bias is too neat, and too straightforward to serve as a theory to explain the huge amount of coincidences that take place in our lives. Why are we using science to explain something that may have no logical cause? It’s illogical to use logic to explain an illogical world.
      In effect, I deny that psychology has the same scientific underpinnings as physics or chemistry or any other pure science.
      I don’t believe that coincidence is simply an illusion created by a predisposition to seek out or focus on particular information. Coincidence is not just a filtering out of noise.
      In the same way, I don’t feel that coincidence is the work of a higher power or some sort of design. However, it does have an uncanny impact on us when it occurs.
      My own opinion is that there are limits to what we can explain through trials and other tools of professional psychologists. My question would be: why are people trying to explain this in the first place?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s