Chapter 20 Testing Your Ideas
This chapter discusses the "M" step in the PIG In MuD innovation processes.
1.Phrase a question based on your interest, observation, and knowledge;
2.Identify the frames and the alternatives;
3.Generate all possible solutions/ideas;
4.Incubate;
5.Meld your single best idea and validate whether it is true;
6.Disseminate your innovation.
After generating multiple plausible solutions/ideas/frameworks, it is necessary for scientists to have a synthesis of the solutions/ideas/frameworks to get the best one, construct a story/hypothesis/model/theory, and prove it.
It's the main body of scientific works. "To do this you will need to spend far more time than you did generating the idea. You will need to get meticulous about design, obsessive about every experimental detail and its documentation, and consumed with avoiding the pitfalls around which your results could be picked apart. In short, like all other scientists, you will have to embark on a rigorous life’s work."
"Proving that your best idea is valid, critically assessing, “Am I sure I have hit on something with the desired effect or that is the true cause?” is the core of science. It is a long and arduous road. Nonetheless, without the dedication that science demands, an idea is just a fantasy."
One more “miserable” thing is that scientists can never prove something is true, because a single observation against the hypothesis/theory will irrefutably invalidate it! Moreover, no one can enumerate all the working conditions of the nature. So what scientists can do, with certainty, is to falsify something.
The rountines of scientific investigations go like this: "science starts with a hypothesis, from which we expect certain observations. If those are the observations we see, then we infer that our hypothesis was correct. But our hypothesis is only correct insofar as it continues to provide a reasonable explanation for how the world works."
So scientists are always on the way to observe, to propose a hypothesis, to design experiments to verify it, to update the hypothesis, repeating this loop all their life.
Then how can scientists tell what they found is true in practice (notice that in theory it is almost impossible)? For biological sciences, a widely used criteria were put forth by Bradford Hill at follows:
• Strength of association (exposure or intervention and outcome are strongly associated)： it's the level of correlations
• Consistency (observations from different studies generally agree)： the reproducibility
• Temporality (the cause must precede the effect)
• Biological gradient (the greater the exposure in duration or dose, the larger the effect)
• Biological plausibility and coherence (known biology supports the idea)
• Experimental evidence (human experiments—that is, clinical trials, if ethical, concur that the intervention or exposure triggers the outcome)
Notice that the above criteria are only for pactical purpose, because findings obey all of them can also proved to be wrong.
So the ultimate pursuit of scientists is to develop a theory to account for as many observations as possible, to make predictions as accurate as possible, and to approximate the truth as closer as possible.