Grip-force modulation during action word reading: An RSVP paradigm with Bayesian temporal analysis.

Grip force provides a behavioral index of motor engagement during action-language processing, but prior studies have relied on auditory presentation and fixed analysis windows. We adapted grip force methodology to Rapid Serial Visual Presentation (RSVP) and used a Time-Resolved Bayesian Evidence Display (TBED) to characterize language-induced motor resonance as a temporally evolving signal rather than as an effect confined to a predefined window. To evaluate this pipeline, we used two theoretically motivated test cases: grammatical aspect (present vs. inceptive constructions) and conscious imagery ability (typical imagery vs. aphantasia). In participants with typical imagery, present-tense action sentences produced greater grip-force modulation than non-action sentences in the conventional 600-900 ms window, showing a directional pattern consistent with previous findings. TBED showed that this effect was temporally restricted and modest in magnitude. Inceptive constructions showed no reliable action-related modulation and yielded Bayesian evidence favoring the absence of an Action/Non-Action difference. Participants with aphantasia showed no reliable motor-resonance pattern, but these results remain difficult to interpret because of sample size and signal variability. Overall, the study demonstrates that RSVP can be combined with grip-force recording, while also highlighting the increased variability that may accompany visual serial presentation. TBED provides a statistically grounded way to move beyond exclusive reliance on conventional window-based analyses by making visible when evidence favors a condition difference, when it favors no difference, and when it remains inconclusive. We argue that time-resolved evidence displays should become a standard component of grip-force analyses, especially when the timing, duration, or absence of motor-resonance effects is theoretically relevant.