It is known that H^1(Sp(2n,Z),V) = 0 for all n>1, and H^2(Sp(2n,Z),V)
= 0 for n>7.
The first result follows from
Raghunathan, M. S.
Cohomology of arithmetic subgroups of algebraic groups. I, II.
Ann. of Math. (2) 86 (1967), 409-424; ibid. (2) 87 1967 279--304.
And the second from
Borel, Armand
Stable real cohomology of arithmetic groups. II. Manifolds and Lie
groups (Notre Dame, Ind., 1980), pp. 21--55,
Progr. Math., 14, Birkh�user, Boston, Mass., 1981.
.
Are similar results known about the *homology* groups: H_1(Sp(2n,Z),V)
and H_2(Sp(2n,Z),V)? (The action on module V is nontrivial so
universal coefficients do not apply.)
Answering my own question:
There is an isomorphism: H^i(G,V') --> H_i(G,V)', where V' is the dual
module V'=Hom(V,k) and k is any field.
The proof is in Brown's book "Cohomology of groups", ch.VI,
Proposition 7.1.