I am reminded of episodes of Star Trek (I'm most familiar with its first incarnation) where it seemed like in every episode our intrepid explorers encountered an entirely new physical phenomenon. Gazing at his instruments and the starship's view screen, Spock says "Fascinating, Captain! A being of pure energy!" A minute later they begin conversing with the being. This is not faithful to human nature. When we encounter a physical phenomenon which we had not imagined before hand, as a rule we don't even notice it. We presume things are exactly the way we imagined them, and the only uncertainties are the ones we've already thought of. The process of noticing something is different than we imagine it to be requires not minutes but centuries - and the main way that it occurs is by us collectively running our heads into a brick wall many many times until our stubborness releases its clutch for a minute and our minds begin to bend around the newly perceived fact. Humans are not good at being surprised.
Long before the advent of physics many people including philosophers spent lifetimes and many manuscripts reasoning about the physical world and how it must be, developing elegant accounts of motion and inertia, heat and cold, germination and death, the stars and planets. These were smart people and after millenia their reasoning about politics, geometry, math, engineering, architecture, art, literature, and logic continues to be a foundation stone for the human race. However their reasoning didn't do so well when applied to the physical world around us, and after the advent of natural philosophy - physics - their ideas about motion and inertia, heat and cold, etc., have been swept away.
One key to natural philosophy's success was, I believe, that natural philosophers started to try to deliberately force their ideas and themselves into collision with nature. It was as if they realized that the only way to get through human thick-headedness was to have head first collisions with facts, and therefore decided to have as many collisions as possible. Natural philosophers deliberately construct situations where they have a very clear and precise idea of what nature will do, and then check that nature actually does what they expect. The human spirit - individually and collectively - wants to be right in these situations, and to obtain confirmations of one's rightness, and has a hard time recognizing evidence to the contrary. This is really the opposite of scientific progress, which is the process of getting one's wrong-ness smashed into one's intellect until one surrenders to the facts of what one is actually observing. Therefore natural philosophers spend a lot of time and effort - whole careers - trying to outwit their own hunger for rightness. This process is called experiments, or the experimental method. The main idea is to force yourself to actually see what is in front of your eyes, to really observe nature. You make a detailed plan of exactly what parts and chemicals etc. will be used, how they will be put together, and all of this gets written very carefully in a lab notebook. You also think out very meticulously what you expect to see, touch, smell, and why you expect it. You make sure to think of and write out as many details as possible - you are not interested in just one number, but in the entire apparatus, everything you will see not just one detail. If you are not attentive to every detail of what you observe then you can not be a good scientist. Then you actually do the experimental procedure, and keep careful records of everything. Later you compare your records of what you actually observed with your records of what you originally expected - you do this comparison in the spirit of an accountant, sweating over every missing penny. There is a theoretical possibility that you the individual researcher may notice something that clearly teaches you something new about the real world, if you're lucky. But much more likely the things she notices seem to to be matters of refining and perfecting her technique, of needing to think things out more clearly the second time, third time, thousandth time. The real learning occurs as many scientists run headfirst at the same wall over and over again, comparing notes as they go, perfecting their technique, and argueing their reasoning out with each other, until finally the group as a whole begins to catch on to a way they might be wrong, and to develop a feel for the real surprise in what they are observing.