Background: The human prefrontal cortex plays a critical role in cognitive control and behavior, and its dysfunction has been linked to numerous psychiatric and neurological disorders. However, noninvasive measurement of prefrontal activity remains challenging, limiting our understanding of how to optimize prefrontal treatments. Input-output relationships reveal how neural circuits respond to different inputs and are essential for determining optimal treatment parameters and understanding individual variability in treatment response, yet systematic investigation of prefrontal input-output relationships has been lacking. Objective: To characterize human prefrontal excitability with input-output (I/O) curves. Methods: We employed transcranial magnetic stimulation (TMS) with electroencephalography in a randomized mixed-block design with 28 healthy participants receiving single-pulse TMS to left dorsolateral prefrontal cortex (dlPFC) across 12 stimulation intensities (60-140% of resting motor threshold). We quantified prefrontal excitability using early local TMS-evoked potentials (EL-TEPs), local cortical responses measured locally 20-60 ms post-stimulus. Results: We observed a strong effect of TMS intensity on prefrontal EL-TEPs. Sigmoidal EL-TEP I/O curves were observed in 57% of participants, with the sigmoidality partially explained by the signal quality of the EL-TEP. Correlations were observed between EL-TEP and motor-evoked potential curve parameters, but intensity parameterization approaches did not significantly differ in explaining inter-individual EL-TEP response variability. Reliable EL-TEPs could be obtained using fewer TMS pulses at higher intensities, and test-retest assessments revealed robust I/O curve profiles. Conclusions: These findings provide a systematic noninvasive characterization of prefrontal input-output physiology in humans, establishing a validated framework for estimating prefrontal excitability. The comparison of various intensity parameterizations motivates the need for enhanced models and individualized measurement of stimulation responses.
Chen et al. (Sun,) studied this question.