Functional Fixedness: Real-world examples

You may remember the candle experiment from the recent post “Motivation 2.0: Daniel Pink on the surprising science of motivation“. The whole point of the candle experiment is to demonstrate that overcoming functional fixedness can not be accelerated with carrots and sticks – on the contrary.

Here, I’d like to give three real-world examples for overcoming functional fixedness. Or actually… one example for, two examples against it.

The first example is a story a friend of mine told me. I may or may not remember the details right, but the point stays.

Painting Cars

He was starting his master’s thesis at the time and looking for a place where he could combine earning some money with an interesting topic. He found an automobile manufacturer with the following situation:
After painting the autobody, the bodies were put in a special hall to dry. Then, they’d be moved to the next hall for the next assembly line. For reasons of no concern for this argument, the drying effectively randomized the order of the autobodies, so in order to calibrate the assembly line, a technician would crawl under the autobody and look for an embossed ID. Then he’d enter that ID at a keyboard at the beginning of the assembly line. (For those guessing, yes, it’s already a while ago.) Of course, this process was slow and error-prone. Wouldn’t it be a nice master’s thesis to set up a camera and some OCR to simplify life for all participants?
It was. The thesis was a plain success. Except… the failure rate of the OCR was around .1% (don’t hang me on the actual number) – and mis-assembling one car out of a thousand was no commercially useful option. So when my friend left, technicians were back to crawling under the car.
And when my friend told me that story a decade or so later, I immediately said: Well, why didn’t you put the camera image on the monitor? People might still have to type in the IDs, but at least the crawling is over. My friend immediately agreed, but back then, everybody was so fixated on the link between camera and OCR that nobody realized how simple it could be.

The second story is from another friend.

1,001 web pages

That friend told me that he had financed some of his university life with a web shop. (Again: Back when…) For some reason, they made a deal with one of their professors: prof pays for some of their expenses, and in return they make the institute’s web site. Lesson learned: Ask for the spec of the web site first! – it was an institute for image processing (in a different city than “Painting cars”), and the professor had wanted a dedicated page for each one of about 1,000 images demonstrating his image processing research. CGI, PHP etc. were barely invented and definitely not robust at the time, so the two of them sat down and hand-coded 1,000 pages as they had always done. Everybody was fixated at the way they had always worked.
Except they could have generated the entire site with shell scripts, perl, maybe even awk or some such. They even had the skills. And again, when we discussed this option a decade or so later, this option was immediately confirmed.

Now you may have the impression that my friends are idiots, and I can assure you that they are really bright. Everybody can get caught in such thinking habits. I’m pretty sure “they” get me every once in a while, and even without knowing you, my valued reader, I’m pretty sure you are being caught regularly yourself.

Here comes the third story:

What we can do about it

After the discussion on the “1,001 web pages”, I felt compelled to tell the story of the candle experiment, as I’ve done so many times in recent weeks. The 1,001 web pages discussion had already led us to a joyful mood and we laughed a lot. When I had reached the point of asking my friend: “you get a box of thumbtacks, matches, and a candle, and a pinboard. You have to attach the burning candle to the pinboard so that it doesn’t drip on the floor” he immediately replied “I put the pinboard flat on the ground with the candle on top”.
First, I’d like to note that – based on my description – this is a valid solution to the candle problem. By the way, it overcomes functional fixedness about the pinboard.
Second, when I said “that’s firmly attached to the wall”, we both had another good laugh and he came up with the “right” answer: use the box from the thumbtacks etc.
For me the conclusion is:

When you’re stuck, have a good laugh.

The above two examples and many, many more speak a clear language: Laughing usually breaks the knot.

When was the last time you made your colleagues laugh when they were stuck?

What do you think?