Understanding individual heat exposure through interdisciplinary research on thermoception

The study addresses how older adults in European cities experience and adapt to urban heat, asking how individual thermoception and behavior interact with measured thermal environments to shape heat exposure. Motivated by rising extreme heat events and their health risks—especially for people aged 65+—the authors note a gap in conventional heat exposure research that emphasizes environmental measurements over subjective experiences and behaviors. They propose thermoception as embodied knowledge linking sensation, perception, and adaptive action, and argue that integrating qualitative ethnographic insights with sensor-based measurements can better capture individual diversity in heat exposure. The research focuses on Warsaw and Madrid (summers 2021–2022) with small samples to enable deep qualitative engagement.

Heat exposure research has largely been dominated by environmental and biomedical sciences focusing on health risks, mortality, and thermal comfort indices; architecture and geography also contribute through design and perception studies. Thermal indices (e.g., UTCI) often fail to capture subjective perceptions, and prior work documents substantial individual differences influenced by location, season, prior residence, mood, and sensory context. Vulnerability is shaped by socio-economic status, gender, disability, and life dissatisfaction. Age-related changes in thermoregulation increase older adults’ risk and reduce their sensitivity to strain. Prior mixed-methods studies often privilege quantitative surveys or sensor data, limiting insight into individual experiences. The concept of thermoception—as embodied, socio-material and learned 'temperature work'—highlights active engagement with environments and devices. The authors argue for combining qualitative and quantitative approaches to co-produce knowledge that accounts for materiality, routines, emotions, and agency in shaping thermal experience.

• Design: Interdisciplinary mixed-methods combining ethnographic research with indoor temperature sensing.
• Sites: Warsaw (humid continental climate) and Madrid (continental Mediterranean climate), both affected by urban heat island (UHI) effects.
• Sample: 20 participants total; 10 per city; all aged 65+, diverse by gender, socio-economic status, and neighborhood location. Pseudonyms used.
• Ethnography: Conducted over two consecutive summers (2021–2022); bi-weekly meetings with recorded interviews, informal conversations, participant observation, and accompaniment on daily activities. Focus on life histories, health, routines, heat impacts, and adaptation practices.
• Sensors: Kestrel D2 sensors deployed in participants’ homes in 2022 (Warsaw: 10 May–29 Aug; Madrid: 30 May–29 Aug). Air temperature recorded every 10 minutes; data downloaded and outliers removed (values >3 standard deviations from mean). Wearable/embedded sensors provide ecological validity in real-life settings.
• Meteorological references: Official outdoor air temperature from Okęcie (Warsaw) and Barajas (Madrid) stations at 20-minute resolution; hourly averages computed for comparison.
• Thermo-diaries: Participants recorded hourly perceived temperature (very cold–cold–neutral–hot–very hot) and adaptation strategies. Diaries digitized and juxtaposed with sensor data to compare subjective and measured conditions.
• Analysis: Juxtaposition of qualitative (ethnography, diaries) and quantitative (sensors, stations) data to examine convergences/divergences; spatial and temporal visualization (e.g., 24-h averaged daily heat maps; day/night indoor means) and boxplots of perceived comfort vs measured temperature. Emphasis on micro-scale, non-generalizable insights; approach is replicable.

• Indoor vs outdoor temperatures: Participants’ homes were generally cooler than reference stations during daytime but considerably warmer at night. Many indoor environments were hotter overall than meteorological stations suggest, underscoring discrepancies between outdoor-based alerts and indoor exposure.
• Spatial and building effects: Notable spatial variability within cities. Top-floor apartments without canopy or with poor insulation (e.g., Antonio, Anna) showed higher indoor temperatures; shaded windows, thick walls, and night ventilation reduced heat (e.g., Javier, Martín). New buildings with large glazing could still overheat (e.g., Wiesia) despite recent thermal regulations.
• Noise and ventilation constraints: Nighttime natural ventilation was a key practice but was often limited by street or festival noise, impeding effective cooling.
• Device use and socio-economics: A/C availability did not consistently translate into lower mean indoor temperatures in sensor data. Usage patterns were shaped by electricity costs, health beliefs (e.g., COVID-19 concerns), and personal resilience (e.g., Antonio minimized A/C; Sandra used it selectively for evening cooling). Fans, ceiling fans, and night cross-ventilation were common.
• Thermoception variability: Thermo-diaries revealed overlapping temperature ranges for different sensations (neutral/hot/too hot) and wide inter-individual differences.
  • Example (Warsaw, Maria): 'Cold' at 21.5–25.5°C and 'too hot' at 23.3–27.2°C, with overlap; perceived heat depended on duration, health, and context (city vs trips).
  • Thresholds: Andrzej and Bogna indicated 'too hot' only at relatively high points (≈25.8°C and 26.2°C, respectively), whereas Bożena reported 'too hot' starting at ≈21.5°C.
• Temporal acclimatization and mood: Perception changed over the season; some adapted to higher temperatures over time (e.g., Alicia, Anna, Sława, Antonio), while others became more sensitive with prolonged heat (e.g., Maria). Emotional states and social contexts modulated experience (e.g., Elias enjoyed early heatwave during neighborhood festivities, then later reported discomfort).
• Agency and adaptation practices: Participants employed socio-material strategies (e.g., shutters/blinds, night ventilation, spraying water, cool showers, wet towels, strategic clothing) and socio-temporal adjustments (daytime seclusion, night walks). These often improved perceived comfort without necessarily lowering mean indoor temperatures captured by sensors.
• Policy-relevant discrepancies: Periods of heightened perceived heat were not always recognized by official warning systems (e.g., Warsaw, June 7–12, 2022; Madrid’s early May heatwave felt more intense than alert level indicated).

The findings show individual heat exposure results from the interplay between measured thermal environments and embodied, socio-material, and temporal adjustments. By integrating ethnography with sensor data, the study uncovers divergences between objective measures and subjective experiences: similar temperatures can feel different across individuals and time due to acclimatization, mood, health, cultural norms, and agency in adapting. While cooling devices and building characteristics shape indoor temperatures, the perceived relief often stems from active 'temperature work'—ventilation routines, water-based cooling, and clothing—rather than substantial changes in mean temperature. This highlights the central role of thermoception and agency in understanding vulnerability. The results suggest city-wide warning systems based solely on outdoor station data may miss key indoor exposures and temporal windows when vulnerable groups feel greater stress. Recognizing diversity in thermoception and the lived context of older adults can improve risk communication, adaptation support, and targeted interventions.

The study demonstrates the added value of combining qualitative ethnography with quantitative temperature measurements to understand how older adults experience and adapt to urban heat in Warsaw and Madrid. Individual heat exposure emerges from interdependent bodily and environmental factors: materiality of dwellings and neighborhoods, access to cooling, routines, socio-economic constraints, and embodied knowledge. Participants who actively engaged in adaptation practices perceived tangible relief and exhibited nuanced thermoception, even when mean indoor temperatures changed little. The authors argue for integrating thermoception into research and policy, co-producing knowledge that reflects diverse experiences, and improving heat warning systems to account for indoor conditions, temporal dynamics, and subjective vulnerability. Future work should explore upscaling these mixed methods and incorporate biological/physiological dimensions to deepen interdisciplinary understanding.

• Small, non-representative sample (n=10 per city) designed for deep qualitative engagement; findings are not statistically generalizable.
• Sensor analysis cannot confirm statistical significance; outlier removal applied; potential measurement and placement variability inherent in real-world settings.
• Thermo-diaries use a subjective scale with possible intra-individual shifts over time; hourly self-reports may be incomplete or context-dependent.
• Cross-city comparisons are not attempted due to climatic differences; results are context-specific to Madrid and Warsaw.
• A/C and adaptation practice effects may not be captured in mean temperature metrics; microclimatic and transient changes could be underrepresented.
• Potential confounders (health status, medications, physiology) not systematically measured; biological data absent.