I understand that multicollinearity is different from variable interactions. What I don't understand is whether an interaction of two variables can account for some of the common effects (multicollinearity) that contribute to the high R^2. I did not include variable interactions in the regression that led to the high R^2, but I did wind up with some of the variables having very large common effects, according to commonality analysis, and a variance inflation factor of 9. So, can I rule out variable interactions being responsible for some of the multicollinearity?

If anyone can help, thanks.