# Optimizing Stubborn Tendencies on the GMAT

- May 2, 22:12 PM Comments [1]

Nate Burke is a Content Developer at Knewton, specializing in GMAT prep.

Are you stubborn? There are many situations in which stubbornness would help a human being. A stubborn nature can be extremely useful when trying to accomplish a complicated goal over an extended period of time. Building a house, traveling a long distance, hunting for food, and courting someone are all activities central to human history that require, at least to some degree, a knock-down, drag-out, brick-headed resolve to get the damn thing done NO MATTER WHAT.

It is natural, therefore, that you would want to spend 5 minutes on the first question in the GMAT quant section. It’s okay. Natural and cultural forces have optimized our problem-solving heuristics in a certain way; recognizing which ones actually are optimal in certain situations is the key to good performance. In any situation in life OTHER than the GMAT quant section, thinking really hard and creatively about a problem until a solution is found (even if it is for an extended period of time) will usually be of value. Not on the GMAT.

We have a saying here at Knewton–answer when you are 90% sure, and move on. This is easy to say and extraordinarily difficult to do–precisely for the reasons outlined above. The reason why it is absolutely necessary, though, has to do with optimization. When you take a computer-adaptive test, you are given a powerful tool–the test itself. The test can work in your favor if you allow it to. The idea is that, over time, the test will be able to calculate your ability and give you questions that reflect it. If you find that you are given a particularly tricky question, DO NOT TAKE IT PERSONALLY. Do not assume that you are stupid for not having any idea how to approach it. Do not interpret it as a challenge that must be overcome by sheer intellectual willpower. The question was generated by the computer as the next step in its program to determine where you fall on the curve. The computer generated the question with the expectation that you’d spend roughly 2 minutes on it and move onto the next question. If your approach creates a situation that in any way deviates from these basic assumptions, the algorithm is designed to output a score that is a poor approximation of your abilities–and it will always err on the side of portraying you as stupid.

But on test day, let it go. Just answer the question and move on.