Young children (eg age 8) say they prefer savannah landscapes over other types of natural landscape (Balling & Falk, 1982). Older children and adults don’t exhibit this preference. The evolutionary psychology interpretation of this is that there is an innate preference for the environment within which modern humans evolved, but that this preference is over-ridden by lifetime development of aesthetic preferences which are influenced by your personality and environment.
Or, put another way, you’re born with a feel for the plains of east Africa, but as you get older you can grow to love the flats of Peckham.
In lots of ways this seems like a typical piece of evolutionary psychology. It could be true – and if it was true it might be interesting – but there’s no reason why it has to be true. Has the experiment been replicated? Has it been replicated cross-culturally? Has it been replicated when controlling for scene complexity and for the adaptive value of the landscape (ie the prospect-refuge affordances). The answers to these questions seem to be either ‘no’ or ‘not alot’ (ie not very well). Obviously i could be wrong and some more delving into the literature might turn up some more references .
It also seems to be crying out for a replication with pre-linguistic infants using a preferential-looking paradigm…
 I think my further reading would begin here:
Appleton, J. 1996. The Experience of Landscape. Revised edition. New York, Wiley.
Orians, G.H. & Heerwagen, J.H. 1992. Evolved responses to landscapes. In Barkow, J.H., Cosmides, L. & Tooby, J. (eds) The Adapted Mind. Oxford, Oxford University Press, pp. 555-579.