I disagree with the basic premise, that scientists do not share their research ideas. I, like most scientists that I know, share most research ideas quite freely, for the following reasons:
- I always have more research ideas than I have time to pursue them.
- Ideas are easy, implementation is hard.
- Sharing ideas is a good way to build collaborations.
- The more people who are interested in the ideas I'm interested in, the better it is for us all.
The only type of "ideas" that I do not typically share are the details of proposals that I am currently writing. The general ideas are probably shared with others, but the particular pitch to a particular call for proposals is obviously competition-sensitive information.
Now, there are some scientists who I know who are much more cagey about their ideas, often because they are insecure about their ability to have ideas and undervalue implementation and execution, and I tend to find that these people are less successful than those who share more freely.