Exploring Consciousness Levels through Electrodermal Responses: A Pilot Study with Auditory Stimuli in DoC Patients

Accurate bedside assessment of disorders of consciousness (DoC) is hampered by high misdiagnosis rates and the practical constraints of advanced neuroimaging. Skin-conductance responses (SCR) offer a low- cost, motor-independent window on autonomic reactivity to emotionally salient stimuli. This study investigated whether SCR modulation by pleasant and unpleasant sounds differentiates vegetative state (VS), minimally conscious state (MCS) and emergence from the minimally conscious state (EMCS) from healthy controls. We expected a significant Stimulus × Group interaction such that (i) healthy controls would show a robust SCR increase to the aversive scream relative to all pleasant sounds, and (ii) if affective discrimination was preserved, higher-functioning patients would follow a graded pattern of preservation in differentiating salient from other sounds (VS < MCS < EMCS). Twenty-eight healthy volunteers and three DoC patients (VS, MCS, EMCS) heard five auditory stimuli, four pleasant (harp, carillon, sea waves, birdsong) and one unpleasant (scream) presented 20 times in a pseudo-random order. SCR was recorded with a wearable biosensor (eSense device); the primary outcome (ΔSCR) was the amplitude difference between baseline and the post-stimulus peak. Linear mixed-effects models with an autoregressive model of second order, error structure tested fixed effects of Stimulus, Condition and their interaction. In controls subjects, the aversive scream elicited a robust SCR increase (b = 0.155, p < .001) relative to all pleasant sounds, confirming heightened sensitivity to threat. The EMCS patient showed a significant SCR reduction to sea-wave sounds (b = –0.361, p = .006), suggesting preserved, but altered, affective discrimination. The MCS patient displayed non-significant trends toward larger responses to natural pleasant sounds, whereas the VS patient showed no stimulus-dependent modulation. In the combined model, the main effect of Scream (p < .001) and the interactions Waves × EMCS (p = .043) and Birds × MCS (p = .007) supported the hypothesized gradient. The pattern of autonomic reactivity across patients and controls conformed to our a priori hypothesis, highlighting SCR as an accessible adjunct for bedside evaluation of residual awareness. However, the very small patient sample limits the generalizability of these preliminary findings. Replication in larger cohorts and integration with neuroimaging will be essential to establish diagnostic reliability.