The more scientists you assign to a research topic, the more efficient each scientist will be. I extracted this table from the exe.
The number to the left is the max number of scientists the topic allows. The numbers along the top are how many scientists are assigned.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
4: 7 9 11 13
5: 7 9 11 12 13
6: 7 9 10 11 12 13
7: 7 8 9 10 11 12 13
8: 7 8 9 10 11 12 12 13
9: 7 8 9 10 11 11 12 12 13
10: 7 8 9 10 11 11 12 12 12 13
11: 7 8 9 10 10 11 11 12 12 12 13
12: 7 8 9 10 10 11 11 11 12 12 12 13
13: 7 8 9 10 10 11 11 11 12 12 12 12 13
14: 7 8 9 10 10 11 11 11 11 12 12 12 12 13
15: 7 8 9 10 10 10 11 11 11 11 12 12 12 12 13
16: 7 8 9 9 10 10 10 11 11 11 11 12 12 12 12 13
17: 7 8 9 9 10 10 10 10 11 11 11 11 12 12 12 12 13
18: 7 8 9 9 9 10 10 10 10 11 11 11 11 12 12 12 12 13
I should also point out that this is the exact range of value for which the table seems to be valid. If you have a research topic with with more of less max scientists, then you can get odd behavior. The research efficiency lookup will be outside of this table and will result in odd behavior.
*** Make sure the max number of scientists is from * 4 - 18 * (inclusive) when editing the research files! ***
If that's not the case, you'll probably end up with research rates of 0, or ridiculously high values.
Edit: Updated table. It was missing the last entry of each row.