Can your brain come up with truly creative ideas that overpower your built-in confirmation bias? Or does your experience with “good” solutions in the past negate your ability to come up with “great” ideas?
Scientific American released a special edition magazine this year focused on “Mysteries of the Mind.” Beginning on page 29, there was a fascinating article that closely aligned to my keynote on MetaCognition. Back in 1942, an American psychologist named Abraham Luchins conducted an experiment where he asked volunteers to do a math problem in their heads. He “gave” them three jars of water of different capacities: one held 21 units, one 127 units and one 3 units. The problem was to measure out exactly 100 units, but each jar had to be filled to capacity in order to achieve the desired amount. Can you do this math in your head? Try it!
The solution is to fill the 127-unit jar full, then pour water from that jar into the 21-unit jar, then fill the 3-unit jar twice. The remaining liquid in the 127-unit jar is now 100 units (127-21-3-3=100). In his experiments, Luchin gave his volunteers multiple examples, all which required the three-step process. Then he gave them a problem that could be solved in the three-sequence manner, but that also had a two-sequence option. Many of Luchins subjects ignored the easier solution, following the now-ingrained 3-step solution. When Luchin gave them a scenario that could only be solved in a 2-step process, some of the volunteers gave up, saying there was no solution to the problem!
This experiment highlights the tendency of the brain to stick to a tried-and-true solution versus seeking out a more efficient one. This is known as the Einstellung Effect. When we are seeking a solution, we tend to have a known answer come to our minds; often we are quick to seize the known solution instead of ideating for more efficient or eloquent ones. In my keynote on MetaCognition, I encourage attendees to “think about how they think.” Specifically, I ask them to recognize when they are making decisions on auto-pilot, to plan for and execute ThinkTime and to craft thoughtful responses instead of reactions to situations, particularly in a crisis.
One of the critical elements of ThinkTime is using all of the available time to come up with creative solutions to vexing problems. Suppose you have an issue that requires creative thought. You allocate an hour to think creatively on a solution; hopefully, you have a space where you can get away from all other distractions and think. After settling your mind and putting aside all of the distracting thoughts (“that report Jim needs is due this afternoon …”), you begin to really think, and you come up with an idea. Chances are, this idea will be something you have dealt with before. It may be a slight iteration on a tried-and-true solution that has worked in the past. Our natural inclination is to seize on an idea and immediately move towards implementing it as a potential solution. However, this is likely a mistake.
You see, our thoughts are jammed full of unavoidable biases. And confirmation bias, the idea that we tend to look for and assimilate information that affirms our previously ingrained belief about something, is one of the strongest. You must assume any ideas you have, at least initially, are suspect to being influenced by confirmation bias. After recognizing this, the solution is to continue to think, iterating on your idea and moving past your initial bias to come up with other ideas that build in the initial one. It’s possible your first idea is the ultimate solution, but if you have allocated an hour to think, and you get a good idea in the first 20 minutes, what is the harm in continuing to think and ideate for the remaining 40 minutes? You’re likely to come up with a stronger, more eloquent idea if you push past the desire to instantly implement the initial creative thought.
In the Scientific American article, a study by researchers Heather Sheridan—now at the University of Albany—and Eyal Reingold of the University of Toronto, was detailed where they engaged chess players, from amateurs to experts, and set up a board with a specific solution. Most of the experienced chess players and all of the experts recognized the “smothered mate” solution, which called for five moves to achieve checkmate. What was interesting, though, was when the chess players were presented with a board set up with the smothered mate solution available in addition to a different, three-move-checkmate solution, only a few of the chess masters executed the more efficient solution. It was as if they couldn’t even see it once they had the smothered mate solution in mind. The researchers placed sensors to study where the chess players were looking on the board and, in fact, their eyes did not gaze to see the easier solution. They were blind to it.
How does this play out in a business scenario? Years ago when I was CEO of a payments company called Goldleaf, we experienced a situation where the web-based system allowing businesses to enter and submit ACH debits and credits would arbitrarily begin slowing down. It was a death spiral: the system would just get slower and slower. Though never technically “down,” it was so slow it made the users think it was down. This caused users to shut down sessions and log back in, further exacerbating the problem. It occurred on no particular schedule, but seemed to happen in the mid-to-late afternoon. It might happen on a Tuesday afternoon, be fine Wednesday and Thursday, then strike again on Friday afternoon. Some of the events occurred on days when most users were attempting to process transactions for payroll events. As you can imagine, this caused much grief for our banks since their customers were upset about the periodic outages.
We turned our code inside out looking for the bug. We attempted to analyze all of the environmental elements to see if something about the Internet connections were the cause. As we analyzed, we would find something, run tests that showed an improvement and put in the fix, only to see the system death spiral two days later. I was regularly on the phone with bank CEOs who demanded to know what we were doing to resolve the problem.
After many weeks of not achieving success, I hired a database consultant to come in and look at our code. After we described the event and he looked at all of the diagnostic data, he outlined everything possible that could create what we were experiencing. One of the conditions he outlined was a memory overwrite, something we had already crossed off as a potential suspect. But because this consultant knew nothing about payments or how our system worked, we had to give a detailed explanation of how transactions are organized into batches, and then how those batches are collected and processed into files. He asked if an individual user could create enough batches to overwrite the memory. Our engineers said it was possible, but it would require many thousands of batches to do so, and our customers were not that large.
But we had never looked to see how many batches our customers actually had. When we ran a query, we found a single customer at a single financial institution who had mistakenly created a batch for every transaction. So instead of thousands of transactions in a dozen batches, they had thousands of batches! Cross-referencing user activity, we instantly could see the system slowdown occurred whenever that particular user logged in. The fix took five minutes. Our technicians were so focused on seeking a software bug they were blind to the fact that there were other potential causes. We were victims of the Einstellung Effect.
Think about your business and–perhaps more importantly–your thinking. Is it possible you have allowed a confirmation bias to blind you to other possibilities? Especially in a time of crisis, it is critically important you examine all of the potential issues that could cause the problem you are experiencing. You need to be hyperaware of the Einstellung Effect, be vigilant to press through the desire to latch onto the first good idea you conceive and (assuming there is adequate time) continue ideating to get all possible causes on the table. Only then can you prioritize the list from highest to lowest probability and work through eliminating issues until you reach a solution.