I'm designing a system (a computer system actually) and (a whole while ago) I spoke with a friend about it and he had this opinion:
If you combine DNA of more parents (than 2) it's better for the offspring.
It's a bit of a theoretical question but is this true?
I'm using genetic algorithms where you can choose the number of parents, if this were to be true I'd take 4 parents, if not I'd stick to 2.
I saw here a question about the occurrence in nature but it remains unanswered.
(2 parents are easier - in this system - so maybe that's the reason it is not observed in nature)
About the system: it needs to produce 'fit' offspring (as fit as possible and as fast as possible), where fitness is measurable (this part is already covered...) and the most fit individuals in the population are selected to pair.
The only limitation set on this kind of systems is that the offspring has to have a chance to be fitter than the parents (so children are allowed to be less fit than parents (being it 4 or 2).
(Of course this lowers the importance of the question but still: does offspring of more than 2 parents have a higher chance to be fit? (my friend mentioned: the more diversity in the DNA the better...))
If you need more information about this bit I'll try to answer comments (if I do know the answer or can give more information about it).