Apologies if this belongs on MSE, but none of the tags I wanted existed, so I took it as a hint to post on MO.
Edit: here is my definition of nearby cycles. Suppose $X$ is a complex analytic space with a map $f: X \to \mathbb{A}^1$, and $F$ is a sheaf on $X$. Then $\psi_f(F) := i^*j_*\pi_*\pi^*j^*F$ where: $i$ is the inclusion of the zero fiber, $j$ is the inclusion of its complement $X^*$, and $\pi: \tilde{X^*} \to X^*$, where $\tilde{X^*}$ is the product over $\mathbb{A}^{1*}$ of its universal cover and $X^*$.
For simplicity, let's suppose I have a complex rank one local system $\mathscr{L}_{\lambda}$ on $\mathbb{G}_m$ with monodromy given by multiplication with complex number $\lambda$. The nearby cycles $\psi_{id}(\mathscr{L}_{\lambda})$ for identity map $id: \mathbb{G}_m \to \mathbb{G}_m$ is the vector space $\mathbb{C}$ considered as a stalk at $0 \in \mathbb{A}^1$. It has a monodromy operator $M: \psi_{id}(\mathscr{L}_{\lambda}) \to \psi_{id}(\mathscr{L}_{\lambda})$ like all nearby cycles, and it's common knowledge that this operator is multiplication by $\lambda$. But why?
I'd like an explanation that doesn't appeal to Milnor fibers. Looking purely at the definition of the nearby cycles functor, it seems to me like $\psi_{id}(\mathscr{L}_{\lambda})$ should have trivial monodromy operator because the pullback of such a local system along the universal cover of $\mathbb{G}_m$ is the constant sheaf with stalk $\mathbb{C}$. I am clearly confused about something, but not sure what...