Digital Phenotyping of Pain Modulation and Associations Among Personality, Attachment, and Behavioral Signatures: Cross-Sectional Study

TL;DR

This study shows that pain perception can be objectively measured through digital behavioral signatures, revealing how personality and attachment traits influence pain responses.

Contribution

The paper introduces a framework for 'digital phenotyping' of pain using real-time behavioral data, moving beyond traditional self-report methods.

Findings

Pain intensity ratings varied significantly based on prior thermal exposure, showing cognitive contrast effects.

Personality traits like vulnerability and self-discipline were linked to specific behavioral signatures of pain modulation.

Secure attachment mediated the effect of self-discipline on pain evaluation instability.

Abstract

The transition from acute to chronic pain often reflects a persistent dissociation between physical tissue damage and subjective reports. In alignment with the 2020 International Association for the Study of Pain definition, pain is a personal experience filtered through a latent “susceptibility architecture.” While clinical assessment currently relies on static, text-based questionnaires, these are often confounded by linguistic interpretation bias and cognitive literacy. We hypothesized that an individual’s internal psychological substrate—traditionally captured via text—can be characterized through real-time behavioral signatures during physical challenge.

This study aimed to demonstrate that the “pain-prone” phenotype can be identified through high-frequency digital assessment of pain ratings. By correlating established psychometric traits with dynamic behavioral signatures, we sought to establish a foundation for “digital phenotyping” that moves beyond the limitations of linguistic self-reports.

A cohort of 534 healthy volunteers (mean age 38.62, SD 22.35 years; n=336, 62.9% male and n=198, 37.1% female) underwent a controlled thermal stimulation protocol (36 °C, 44 °C, 46 °C, and 48 °C). Continuous pain intensity was recorded via a high-frequency (1000 Hz) digital visual analog scale (VAS). To establish a psychological baseline, participants were profiled using the Revised NEO Personality Inventory (NEO PI-R) and the Relationship Questionnaire. Two behavioral indexes were then derived from the digital VAS: the temporal augmentation index (TAI), reflecting within-stimulus physiological sensitization, and the cognitive contrast effect (evaluative instability). Statistical significance was adjusted using the false discovery rate.

Repeated-measure multivariate ANOVA confirmed a highly significant main effect of time for all noxious conditions (P<.001; 46 °C: t533=27.69). Perceived intensity at 46 °C was significantly lower following 48 °C (mean VAS 12.31; SD 15.55) than following 36 °C (mean VAS 30.45; SD 22.38; t533=−25.76; P<.001). Crucially, vulnerability (facet N6 of the NEO PI-R) was significantly associated with contrast magnitude (q=.03) and showed a trend for the TAI (q=.09), whereas self-discipline (facet C5) showed a significant negative association with the TAI (q=.048) and a trend for contrast magnitude (q=.09). Mediation analysis identified 2 distinct pathways: (1) a “stabilization path” where secure attachment fully mediated the inhibitory effect of facet C5 on evaluative instability (direct effect c′=−0.25; P=.11) and (2) an “instability path” where facet N6 exerted a direct amplifying effect on instability (c′=0.34; P=.03).

Subjective pain evaluation is governed by a stable internal psychological substrate. By shifting the assessment modality from linguistic self-reports to dynamic behavioral signatures, we provide a framework for “digital phenotyping.” These evaluation patterns serve as an objective behavioral marker, enabling the identification of latent susceptibility before chronification and offering a novel foundation for personalized precision pain management.

Figures