For a kid who wants to learn how to ride a bicycle, there are many options (see Figure 2). A normal bike with extra wheels for support (2A), or one with a support handle at the back for the parents to help (2B) or a tricycle (2C). However, it turns out that a very good solution (2D) is to remove almost everything, the pedals, the chain, the brakes and certainly the extra wheels and what is left is a balance bike: by removing all unnecessary items, one can focus on the skill that really matters: balance.

Gabrielle Adams and colleagues tell us (in this video and in this wonderful paper), that the wisdom of solving problems by removing parts is not as commonly appreciated as it should be. In a heroic effort with so many observational and interventional experiments, they demonstrate very clearly that people systematically prefer to look for additive solutions and, in doing so, many people fail to see the often obvious subtractive solution that is hiding in plain sight, right in front of them. Below, we review some of these experiments briefly.

In one experiment, they gave people puzzles like the one you saw at the beginning, in Figure 1. People could click on each square in the gird to flip its colour between white and blue and their goal was to make the design symmetrical from top to bottom and from left to right. The results are depicted in Figure 3 and showed that most people added 9 more blue squares whereas only 20% of people removed 4 to reach a perfectly symmetric design.

In another field study, they examined their university archives from when a new incoming president of the university had asked everyone to send in their suggestions for how to improve things. They found that out of a total of 651 suggestions, 11% i.e., 70 suggestions for change involved removing an existing regulation and 89% proposed new regulations to be added.

In another study, they gave people a LEGO problem to solve (see Figure 4).

The important point here is that they introduced a cost for adding things. This cost could be avoided if you found a solution that involved removing. Every participant in the experiment would receive a bonus of 1$ for solving the problem. However, for every block that they added, their bonus be reduced by 10 cents. The results are shown in figure 5.

These experiments showed that coming up with additive solutions is easier and more spontaneous than subtractive solutions. Adams and colleagues concluded that subtractive solutions are less accessible. However, one great thing about the article and this discovery is that so many other possible explanations come to mind too. In a very insightful commentary, Tom Meyvis and Heeyoung Yoon offer a number of such alternatives. It could be that:

(1) subtractions are less likely to be appreciated by others. In other words, subtractive solutions may have come to people’s mind but then discarded because they did not think much of them or thought others may give us enough credit for them.

(2) subtractions may be received negatively by others. We often hear that previous rules and regulations have been put in place “for a reason” which may not be apparent to us.

(3) people may be affected by sunk-cost fallacy. Once an investment is made (in money, time, effort, …) and a solution is developed, people may be reluctant to accept that the investment has been wasted and abandoning that solution is a good idea. There are many example of people continuing a bad decision because it would be too hard for them to accept that their investment has been lost.