Predictability Is Not Provenance: Rethinking Next-Word Prediction in LLM-Based Neural Encoding

Large language model (LLM)-based semantic encoding studies often interpret pre-onset performance in target-aligned lag–$r$ curves as evidence that the brain predicts upcoming words. We show that this inference has two limitations. First, contextual representations of successive words are statistically dependent, while responses elicited by those words overlap in the measured signal. Conventional target-word encoding can therefore conflate the target contribution with covariance-weighted contributions from historical sources, making predictive success ambiguous with respect to neural provenance. Second, held-out correlation quantifies aggregate predictive adequacy but does not identify which word generated predictable activity or how its contribution evolves from onset. We formalize this ambiguity with a constrained-availability source-time finite impulse response model (CAST-FIR) and derive Target Isolation from it as a source-adjustment procedure. A strictly causal simulation and a passive acoustic benchmark produce prospective-looking performance without anticipatory computation. Across English ECoG and Chinese SEEG, Target Isolation contracts the $r>0.075$ cross-onset temporal-generalization area by $98.1$% and $100.0$%, respectively. Fitted source-relative kernels peak at $75$ ms in English and $225$ ms in Chinese after source onset. These results do not rule out neural prediction. They show that target-aligned predictability alone does not establish target-word provenance: performance should establish predictive adequacy, whereas temporal interpretation requires explicit source-time structure.