Let $G$ be a compact Lie group and $V$ a finite-dimensional orthogonal $G$-representation. Write $S^V$ for the quotient $D(V)/S(V)$, where $D(V)$ and $S(V)$ are the unit disk and sphere in $V$, respectively. The space $S^V$ is known to be a finite pointed $G$-CW-complex.

Cappell and Shaneson [Ann. of Math. 113 (1981), 315-155] proved that there exists a finite group (plenty of them, in fact) and finite-dimensional orthogonal $G$-representations $V$, $W$ such that $V$ and $W$ are *not* equivalent as $G$-representations, but the $G$-homotopy types of $S^V$ and $S^W$ coincide.

I have found a note in another article stating that this cannot happen for compact and *connected* Lie groups. Can anyone explain to me why this is the case?