Bench Talk

Flying home from my internship in Washington D.C., I began to think about airplane crashes and what it would be like to be in a plane crash. 

 I wondered what I would do if I got in a crash not far from D.C. and survived. The logical me would go to the nearest city and take a plane from there. It would be safer. My emotions, however, would make it difficult to get back on a plane. 

 This goes to the core of human irrational thinking. We often are terrible at looking at the data and behaving logically, whether this be risk assessment or behavioral response. 

 This is where big data comes into play. Big data allows us to gather enough data on specific tasks to make accurate evaluations on cause and effect. This enables us to weed out the outliers, to see the long term trends and therefore understand the cause and effect of one event in context of the whole picture. 

 One example is found in Thinking, Fast and Slow by Daniel Kahneman. People are more prone to improve when given positive reinforcement rather than negative criticism. This goes against my personal intuition. To me, it always seems a really good grade is followed by a mediocre or bad grade. 

 Seen in context of the bigger picture we get accurate information on the slope from the point before the positive reinforcement; and the point after it is steeper, faster improvement, than the point before and after the negative criticism. 

Humans are like averaging functions whose averaging window is really small. We only look at the downward slope between the positive reinforcement and the next practice, rather than the upward trend across three practices. We give too much weight to “the what” has just happened and too little to the past. 

 Because we can’t fathom big numbers, we react in a manner based on how we feel. It is why we are more afraid of skydiving than drinking alcohol near water, unless a friend recently drowned. 

 Fortunately, when it comes to the forward progression of humanity, we no longer have to rely on how things feel and our intuition. We can actually collate millions of particles, gather data across airplane wings to simulate stress points, survey the entire sky, know what advertising works and so much more. It’s up to us as a collective body of researchers and curious people, the wonderers in us all, to adapt to this added perspective on the world and wisely use it for the betterment of our future. 

 There are caveats to this. Once a person becomes exceptionally skilled it is more helpful to point out the shortcoming. There are full articles and books written about the difference. They are very interesting, but I wanted to talk about human perception, not best leadership practices.