Actually, I disagree.
It seems to me that in general, the researcher cannot choose the effect - in the sense Te, Fo, RDT etc. - but can choose whether it's produced by a formulaic spell, an enchanted device etc. Of course, in some cases the specific nature of the research rules out certain possibilities. For example, if an effect is a Ritual, in most cases enchanted devices are unsuitable.
Let me also quote somewhat more expansively the AM p.8 passage quoted by silveroak:
"anything that uses a Lab Total and creates something may be a legitimate Insight outgrowth. However, spells are by far the most common, followed by item enchantments.
The troupe should decide on the effects that result from particular pieces of Insight, choosing ones that
advance the saga without upsetting play balance. Player characters can vary things as much as they like once they fully integrate the system, so the restrictions at this stage should be accepted with good grace."
a) the fact that spells are more common might simply be an outgrowth of the fact that it's typically cheaper to experiment with spells than with enchanted devices.
b) even if you rule that the researcher has no freedom to choose between spell and device or other "stuff", the effect is by the rules as written
- a troupe decision
- that should advance the saga
- and such that any restrictions only temporarily impede the magus, with the idea that eventually he'll get the breakthrough and the restrictions will be lifted.
Thus, if your seeker cannot use formulaic spells because of some flaws, the troupe might make integration a little more challenging for him (I mean, a Flaw is supposed to be a Flaw), possibly ruling out this or that source of Insight, requiring special vis, etc. but definitely not impossible.