Local Parity and Systematic Peterson Counterexamples in the Motivic Hit Problem

Abstract

The motivic hit problem asks for a minimal set of module generators of $H^{*,*}(BV_n;\mathbb F_2)$ over the mod~$2$ motivic Steenrod algebra. Kameko proved that the motivic Peterson-type analogue of Wood's theorem fails by constructing monomials $z_k$ which are not hit even when the corresponding topological degree may satisfy $β(d)>n$. His proof passes to $N_n=M_n/(τ)$ and analyzes, in degree $d=k+2d_1$ with $d_1=(n-1)(2^k-1)$, a distinguished summand whose basis consists of the monotone translates of $z_k$. In this work, we isolate the local content of this summand before quotienting by hit elements. More precisely, we construct a linear projection \[ \vartheta:N_n^{d,*}\longrightarrow V, \] where $V$ is the $M_1$--summand spanned by the images of the monomials $σ(z_k)$, and define a parity functional $ε:V\to\mathbb F_2$ by summing the coefficients of these basis vectors. We prove that the local image of the hit subspace is exactly the parity-zero hyperplane: \[ \vartheta\bigl(A^\sharp_+(N_n)\cap N_n^{d,*}\bigr)=\ker(\varepsilon). \] Consequently, every element whose local $M_1$--component has odd parity is non-hit, and every odd-parity linear combination of the monotone translates of $z_k$ determines a nonzero class in the motivic hit quotient. We also obtain a systematic arithmetic family. For every integer $m\ge 3$, set $n=2^r+1$ and $k=n-m$. If \[ r\ge m+α(m-3), \] then the degree $d=(n-1)(2^{k+1}-2)+k$ satisfies $β(d)>n$. Hence, for every fixed $m\ge 3$, these classes give infinitely many motivic Peterson-type counterexamples with $k=n-m$. The local parity theorem holds over every algebraically closed field of characteristic different from $2$, and naturality under extension of the base field carries its non-hit consequences to every field of characteristic different from $2$.