Neural dynamics during heat pain threshold assessment using the method of limits

The assessment of the heat pain threshold (HPT) as part of quantitative sensory testing (QST) protocols is a widely applied method which is used both in experimental settings and for the characterization of clinical populations. The neuronal responses that occur during clinically utilized HPT assessments, however, have been scarcely investigated directly. To address this gap, we investigated the peristimulus BOLD time courses in response to ascending thermal ramps employing the "Method of Limits" (MoL) approach of HPT testing in 30 healthy male participants. This showed that several brain regions, such as rostro-dorsal parts of the anterior insula (aINS), exhibit stimulus-response (SR) behavior that approximates a linear pattern corresponding to stimulus intensity. In contrast, other regions, including the anterior midcingulate cortex (aMCC), the primary somatosensory cortex (SI), and ventral sections of the aINS, show a sudden signal increase upon exceeding the HPT. The Neurological Pain Signature (NPS), a well-known biomarker for nociceptive pain, demonstrated good prediction of the HPT at the group level and moderate accuracy at the individual level. Notably, NPS subregions overlapped spatially with brain areas predominantly exhibiting linear SR behavior, indicating that the NPS response may be partly driven by stimulus intensity. Employing a Hidden Markov Model (HMM) to also capture distributional properties of the BOLD responses, along with appropriate transition probabilities, enabled a reliable prediction of the individual Method of Limits-derived HPT and provides probabilistic insights into how the brain responds during the transition from heat to pain.