NEUR1020 TUTORIAL Six Principles of Scientific Thinking 1- Extraordinary Claims: When a study makes an extraordinary claim that defies everything we know, we would expect some pretty convincing data to back it up. If someone claimed to have developed a drug that completely cures advanced stages of dementia, we would not settle with the anecdote of one patient who performed better in a cognitive test. We would need large evidence-based data that would support this claim, a large number of participants in the study, and a very thorough and controlled way of measuring what is being tested. In the same way, when we read about "all or nothing" claims such as "it has been proven that drinking coffee fights Alzheimer", we would need supporting evidence to back this up. Not only this, but if the claim is made by the National Colombian Coffee Federation, we may be sceptical that the people making the claim aren't exactly neutral and have high stakes in the issue. We therefore need strong evidence by independent parties. The more extraordinary the claim, the more evidence we need to be able to question the existing and already tested theories. 2- Testing Predictions: When we make a prediction we generally base it on previous research or theories. By reading previous literature we get an idea of what we would probably expect to happen, but we need to test it. Sometimes, however, the results can be in the complete opposite direction! And this is okay too, because it means something. It may mean that there are different factors that we are not taking into consideration, or that people actually behave in a completely different way that we had initially thought of in a given situation. It is therefore very important to test our predictions, as our intuition can often mislead us. 3- Ockham's razor: This is also known as parsimony to be the best one". Sometimes we may overcomplicate things and there can be perfectly reasonable, more simple explanations. This does not mean, however, that it necessarily is the right explanation, but it would be the best place to start from to test our hypotheses. 4- Replicability: You will hear about the replicability crisis numerous times throughout your studies. And this refers to the issue that many of the articles published in the top psychology journals have failed to be replicated. What does it say about a theory or a finding if you cannot trust its results because every time someone runs the experiment the results are different? It is important for the results to be consistent for us to be able to trust a theory, and this also means the hypothesis should be tested by different experimenters/ entities that are impartial to it or don't have any stakes in it. Page 1 | 2
NEUR1020: TUTORIAL 5- Ruling Out Alternative Hypotheses: If your findings can be equally explained by different hypotheses, how do you know which one is the correct one? Have you even considered that there