Scientific thinking

by Marcus Loane


"Education is the process of moving from cocksure ignorance to thoughtful uncertainty."
Utvich's Observation


When you have a set of facts (observations) that need an explanation, think of a possible explanation (hypothesis) and then think of a way to test your hypothesis. You should have a test that is capable of showing your hypothesis is wrong as well as tests to show it is probably correct. Note the use of the word "probably". Certainty is a luxury which we can't always have. You should think of more than one possible explanation. Think of all the explanations that could explain your observations and then think of ways to test or eliminate each one of them.

If your hypothesis is repeatedly shown to be correct by experiment, it may be promoted to theory. Theory is as good as it gets. If a theory is short and simple (like an equation) it is sometimes called a law.

The formalism of the scientific process can be adapted for everyday problems.

A good rule of thumb is that if two theories explain the facts equally well, choose the simplest one, the one that makes the least assumptions. That is the principle of parsimony. This is also good practice in everyday thinking when we don't have the inclination or time to perform tests.

For example if someone claims to have seen an alien spaceship, what are the possible explanations?

The first explanation makes an assumption that aliens exist and are visiting us. The other 3 explanations do not make any assumptions that are not already known to be plausible (weather balloons exist, people sometimes lie or hallucinate). If we wanted to investigate further we could learn more about the witness. Have they lied or hallucinated in the past? We could check with the meteorological office about the positions of weather balloons. It is very important to think of as many explanations as possible instead of latching unto the first one that comes to mind.

Here is an example of latching on to a theory too soon. Jury members often come to a conclusion about whether someone is guilty or not based only on their initial appearance. Do they look shifty? They then listen only to the evidence that confirms their foregone conclusion and ignore the rest. What they should do is juggle multiple working hypotheses in their head and adjust the probabilities of each as the evidence unfolds. This way of thinking does not come naturally. It has to be learned and practised.

Marcus Loane