$$A=\left[ \begin{array}{cc} a & b \\ c & d \end{array}\right] $$
$$B=\left[ \begin{array}{ccc} a^{2} & ab & b^{2} \\ 2ac & ad+bc & 2bd \\ c^{2} & cd & d^{2} \end{array}\right] $$
Let eigenvalues of $A$ be $a_1,a_2$. Then, eigenvalues of $B$ are $a_1^2$, $a_1a_2$, $a_2^2$. Why does this happen? Is this related to spectral analysis of boolean function or spectral property of Cayley graph?