Gyration stability for projective planes (Q6979955)

Let \(M\) be a closed simply-connected \(n\)-manifold. Take an integer \(k\geq 2\) and a class \(\tau\in\pi_{k-1}(SO(n))\), the \(k\)-gyration of \(M\) by \(\tau\) is defined as\N\[\N\mathcal{G}^k_\tau=((M-\mathrm{Int}(D^n))\times S^{k-1})\cup_t(S^{n-1}\times D^k),\N\]\Nwhere \(D^n\) is an embedded \(n\)-disk in \(M\) and the map \(t: S^{n-1}\times S^{k-1}\to S^{n-1}\times S^{k-1}\) is specified by \(t(a,x)=(\tau(x)\cdot a,x)\). In other words, the \(k\)-gyration of \(M\) by \(\tau\) is obtained by a surgery on \(D^n\times S^{k-1}\subset M\times S^{k-1}\) twisted by \(\tau\).\N\NFor ``gyration stability'', the authors consider the following two questions:\N\N\textbf{Question (GSI).} For a given \(k\geq 2\) and \(n\)-manifold \(M\), do we have \(\mathcal{G}^k_\tau(M)\simeq \mathcal{G}^k_\omega(M)\) for all twistings \(\tau,\omega\in\pi_{k-1}(SO(n))\)?\N\N\textbf{Question (GSII).} For a given \(k\geq 2\) and \(n\)-manifold \(M\), how many different homotopy types can \(\mathcal{G}^k_\tau(M)\) have as the homotopy class \(\tau\) is varied?\N\NIn this paper, the authors study these questions for \(\mathbb{C}P^2, \mathbb{H}P^2\), and \(\mathbb{O}P^2\). Their results are summarized in the table located at the bottom of Page 3.