Mary, this is superb question, but really you never know things are going to work, that’s why we must all experiment! However, experience does count for a lot. If you watch and listen carefully to others, sometimes you can start to get a hunch for things before they’ve even happened. In my case, lots of other scientists were beginning to wonder and experiment on the powers of the immune system and there were several other experiments that other people published just before I carried out my Nobel-winning work, that made me realise that there must be some way for the immune system could ‘learn’. Big discoveries are almost always the same – taking lots of little bits on information and bringing them together in such a way that shows how the link and why they might be much more important than we’ve realised before.
This is an interesting question, which as Peter says goes to the heart of research. When you are doing cutting-edge research, you can never really know that something is going to work: it’s the fact that you don’t know that makes it worthwhile doing the work in the first place. What you try to avoid is having something fail for a reason you could have done something about. I worked on understanding the structure of biological molecules, so one key factor was to make sure that your sample was a good sample: it contained what it was supposed to contain, and it hadn’t been mistreated (e.g. left out on a hot windowsill), so that the structures would not be damaged. Then you have to make sure that the equipment is set up correctly: you do not want to ruin your X-ray image by under- or over-exposing. I was good at this: my photos were better than the competition! So, I think the answer is: you take the most care you can to ensure that the experiment is set up properly, that it is measuring what you intended to measure, and that the sample you are studying is as good as you can make it. Then you just have to hope that the results tell you what you want to know – or at least give you enough clues to design a better experiment.
Comments
Rosalind Franklin commented on :
This is an interesting question, which as Peter says goes to the heart of research. When you are doing cutting-edge research, you can never really know that something is going to work: it’s the fact that you don’t know that makes it worthwhile doing the work in the first place. What you try to avoid is having something fail for a reason you could have done something about. I worked on understanding the structure of biological molecules, so one key factor was to make sure that your sample was a good sample: it contained what it was supposed to contain, and it hadn’t been mistreated (e.g. left out on a hot windowsill), so that the structures would not be damaged. Then you have to make sure that the equipment is set up correctly: you do not want to ruin your X-ray image by under- or over-exposing. I was good at this: my photos were better than the competition! So, I think the answer is: you take the most care you can to ensure that the experiment is set up properly, that it is measuring what you intended to measure, and that the sample you are studying is as good as you can make it. Then you just have to hope that the results tell you what you want to know – or at least give you enough clues to design a better experiment.