Predicting thermal pleasure experienced in dynamic environments from simulated cutaneous thermoreceptor activity

Area of Science:

• Human thermal perception
• Neurophysiology
• Sensory science

Background:

• Traditional research on indoor thermal perception emphasizes steady-state neutrality.
• The hedonic dimension of thermal perception, known as thermal alliesthesia, has been under-explored.
• Understanding the link between physiological responses and subjective thermal pleasure is crucial for improving comfort.

Purpose of the Study:

• To demonstrate that sensory neuron activity can predict thermal pleasure under dynamic thermal exposures.
• To establish an empirical link between the neurophysiology of cutaneous thermoreceptors and the hedonic perception of thermal stimuli.
• To develop a predictive model for thermal pleasure based on simulated thermoreceptor impulses.

Main Methods:

• A numerical model of cutaneous thermoreceptors was developed and applied to skin temperature data from 12 human subjects.
• A random forest machine learning model was trained on simulated thermoreceptor impulses to classify pleasure responses.
• Model performance was evaluated using F1 scores and false positive/negative rates, with validation on an independent dataset.

Main Results:

• The random forest model successfully classified pleasure responses with a 67% F1 score and 4% false positives/negatives.
• Model accuracy improved to 83% when excluding extreme pleasure or displeasure responses.
• Independent dataset validation confirmed the reliability of the predictive model.

Conclusions:

• This study provides the first empirical evidence directly linking thermoreceptor activity to the perception of thermal pleasure.
• The findings highlight the potential of using neurophysiological data to predict subjective thermal experiences.
• These insights can inform the design of built environments to promote sensory engagement and enhance occupant well-being beyond mere thermal neutrality.