Human Experience in Regulated Offices (HERO) dataset

Synopsis The multimodal Human Experience in Regulated Offices (HERO) dataset captures human physiological and perceptual responses to controlled indoor thermal environments during realistic office-like activities. Twenty-four healthy adults (pseudonymized as PXX; 24 participants with non-contiguous IDs up to P26) each attended up to two laboratory sessions under distinct thermal regimes: a lower-temperature (LT; ~22 °C ± 0.5 °C) and a higher-temperature (HT; ~30 °C ± 0.5 °C) conditions. Each session is identified as ID_Session = PXX_LT or PXX_HT. Experimental protocol and temporal backbone In each session, participants completed a structured nine-phase protocol comprising baseline, reading, writing, group discussion, a simulated phone call, and interleaved washouts. A global timeline file (timeline.csv) specifies the ISO-8601 start and end timestamps of all phases for every session and serves as the temporal backbone for aligning all modalities. Data level and preprocessing scope This Zenodo release provides Level-0 data: device-level streams and contextual measurements restricted to the experimental phases and annotated with phase metadata, but without downstream signal preprocessing (e.g., no EEG re-referencing/ICA, no EDA tonic–phasic decomposition, no artefact removal, no additional filtering beyond device export). This allows researchers to implement and compare their own preprocessing pipelines while benefiting from consistent phase alignment. Included modalities Electroencephalography (EEG, Muse S): Four-channel scalp EEG (TP9, AF7, AF8, TP10) Electrodermal Activity (EDA, EmotiBit): Skin conductance. Peripheral Skin Temperature (ST, EmotiBit): Peripheral temperature (°C). Environmental indoor climate (env): Room-level environmental measurements recorded throughout sessions. Perceptual / survey data (survey): Phase-aligned self-report measures collected at baseline and during/after key phases. Video modality (derived; privacy-preserving): Raw webcam videos are not published. Instead, a publishable manifest and derived numeric features are provided for quality control and analysis support (e.g., motion/brightness/sharpness proxies and face_present_ratio computed at low sampling rate). No frames, audio, or biometric embeddings are released. Intended research use The HERO dataset is designed to advance research on: Neurophysiological dynamics Thermal comfort and thermal stress modeling Affective and cognitive responses under thermal challenge Multimodal signal processing and machine learning Researchers can use the Level-0 streams to implement their own pipelines (e.g., EEG filtering/ICA, EDA decomposition, environmental feature engineering, survey–physiology integration) while relying on the dataset’s phase-accurate temporal alignment. Ethics and privacy All participant identifiers are pseudonymized, and the release contains no directly identifying personal information. Data collection was conducted under appropriate ethical approvals, and participants provided informed consent for anonymized data sharing for research purposes. The dataset is released for non-commercial research and educational use under the license specified in the Zenodo record. For more information, go through README file supplied with the dataset.