Researchers Unveil How Emotions Manifest Through Skin Conductance

A recent study by researchers from NYU Tandon and the Icahn School of Medicine at Mount Sinai has revealed how emotional responses to sensory stimuli can be quantified through skin conductance. Published in PLOS Mental Health, the findings suggest that physiological signals can indicate cognitive arousal—mental alertness and emotional activation—without relying on subjective reports.

The research team, led by Rose Faghih, an Associate Professor of Biomedical Engineering at NYU Tandon, focused on skin conductance as a reliable indicator of autonomic nervous system activity. The skin’s ability to conduct electricity varies with sweat gland stimulation, a process known as electrodermal activity, which has been linked to emotional and cognitive states for some time.

What sets this study apart is its innovative approach, combining physiological modeling with advanced statistical methods to interpret the subtle electrical fluctuations in response to sensory experiences. The investigation originated as a course project for student authors Suzanne Oliver and Jinhan Zhang in Faghih’s Neural and Physiological Signal Processing course, with mentoring from research scientist Vidya Raju and support from James W. Murrough, a Professor of Psychiatry and Neuroscience.

Analyzing Responses to Stimuli

The researchers examined a dataset in which participants’ skin conductance was continuously recorded while they were exposed to various visual, auditory, and haptic stimuli. Participants also provided self-ratings of arousal using the Self-Assessment Manikin, a pictorial scale designed to quantify emotional states.

Using a physiologically informed computational model, the team was able to distinguish between slow and fast components of the skin’s electrical response. They inferred periods of heightened autonomic nervous system activity, employing Bayesian filtering and a marked point process algorithm to derive a continuous measure of cognitive arousal over time.

The analysis uncovered significant insights: the nervous system showed the strongest responses within two seconds of new stimuli, with haptic sensations generating the most immediate reactions. Interestingly, though physiological signals indicated heightened arousal, participants self-reported higher arousal levels associated with auditory stimuli, particularly sounds and music. This discrepancy suggests that while the brain’s perception of stimulation and bodily responses are interconnected, they do not always align.

When the researchers modeled arousal levels, they found consistency with participants’ self-assessments, particularly regarding auditory stimuli. The model effectively tracked transitions in arousal as participants moved from low to high-intensity stimuli, achieving accuracy that exceeded random chance. Furthermore, when participants were analyzed based on their sensitivity to visual or tactile stimuli, the model highlighted distinct responses reflected in their self-reports.

Implications for Health and Technology

The implications of this research extend beyond academic inquiry. In clinical contexts, self-reported measures are often deemed the gold standard for assessing mental states such as anxiety or stress. However, these measures can be subjective and unreliable. The objective metrics derived from skin conductance could provide clinicians with a refined understanding of patients’ emotional dynamics in real time.

Such advancements may prove beneficial in monitoring recovery from conditions like depression, anxiety, or post-traumatic stress disorder, where physiological arousal frequently correlates with fluctuating symptoms. Additionally, the study suggests potential applications in virtual reality and human-computer interaction. By quantifying user reactions to various sensory elements, systems could dynamically adapt to enhance immersion, focus, or relaxation based on user needs.

Despite the promising findings, the authors caution against oversimplifying the link between physiological signals and emotional understanding. Factors such as the duration of stimuli, individual variability, and personal experiences complicate interpretation. The overall correlation between computed arousal and self-reported ratings was modest, reflecting the nuanced nature of emotional experiences. Nonetheless, the model’s ability to consistently identify moments of increased engagement underscores its potential as a complementary measure of internal states.

In summary, this study bridges the gap between physiology and perception, illuminating how emotions manifest through the body’s electrical rhythms. The research invites a more continuous, data-driven perspective on human emotional experience, which may ultimately inform both mental health care and the development of emotionally intelligent technologies. For further details, refer to the work by Suzanne Oliver et al, titled “Cognitive arousal-based measures quantify insights from self-ratings in response to sensory stimuli,” published in PLOS Mental Health in 2025.