The problem is only about the indexing. When you do something like sV(k, c), you're actually getting a matrix with all possible permutation betwenn k and c. For you to get only the single combinations you need to convert it to linear indexing, as example the function sub2ind . The following code illustrate in a simplified problem your issue
p = [1:6]';
k = [1:3]';
c = [4:6]';
sV = zeros(6);
% Will not work, since sV(k,c) is a matrix. This is the error that you're getting
%sV(k,c) = [p(k)-p(c)]'
% Will work, since now you get only three index which are the combination
% of both k and c in ther respectively order
ind = sub2ind(size(sV),k,c)
sV(ind) = [p(k)-p(c)]'
sV =
0 0 0 -3 0 0
0 0 0 0 -3 0
0 0 0 0 0 -3
0 0 0 0 0 0
0 0 0 0 0 0
0 0 0 0 0 0
