An experimental design framework for the personalization of indoor microclimates through feedback loops between responsive thermal systems and occupant biometrics

Abstract

How can building technologies accommodate different and often conflicting user preferences without dissolving the social cohesiveness, intrinsic of every architectural intervention? Individual thermal comfort has often been considered a negligible sensorial experience by modern heating and cooling technologies, and is often influenced by large-group norms. Alternatively, we propose that buildings are repositories of indoor microclimates that can be realized to provide personalized comfort, to create healthier environments, and to enhance the attributes of architectural interventions into haptic dimensions. In response, the goal of this study is to characterize an experimental framework that integrates responsive thermal systems with occupants’ direct and indirect experience, which includes stress response and biometric data. A computational model was used up to inform and analyze thermal perception of subjects, and later tested in a responsive physical installation. While results show that thermal comfort assessment is affected by individual differences including cognitive functions and biometrics, further computational efforts are needed to validate biometric indicators. Finally, the implications of personalized built environments are discussed with respect to future technology developments and possibilities of design driven by biometric data.

Keywords

Personalized thermal comfort interactive building technologies bio-feedback loops indoor microclimates

Introduction

Existing building technologies are often focused on meeting large-group comfort standards while overlooking the tensions between individuals in the built environment. Providing thermal comfort has been a fundamental prerogative of architecture, not just as a functional requirement, but rather as a deep bodily experience.¹ However, the prevailing use of contemporary air-conditioning technologies has compounded this issue, transforming buildings’ thermal environments into a prescriptive application of universally valid comfort assumptions. In response to the mechanistic approach for managing and delivering thermal comfort in buildings, diversified thermal environments favor the notion that technological advancement and architectural elements are an inevitable part of a complex integration of personal experiences, cultural identities, and available energy resources.²

How do perception and cognition inform and shape buildings’ thermal identities for personalized comfort? Here, we propose an iterative experimental design framework that integrates distributed thermal technologies with occupant’s conscious and unconscious responses to diversified environmental conditions. An interactive installation was built up by the Atmospheric Delight Cluster during the 2016 Smartgeometry Workshop in Gothenburg, Sweden, in order to evaluate the opportunities and challenges related to localized thermal environments for a bio-responsive thermal system.

Significance of the work

This research addresses the role of thermal environments in individual and collective indoor experiences, and how creating variations in occupant thermal conditions is significant for providing a source of delight and satisfaction. Advancements in distributed building technologies provide temporal and spatial environmental diversity, which has been shown to be important for cognitive processes, occupants’ health and well-being, and cultural preference.^3,4 Additionally, localized thermal conditioning would significantly reduce energy consumption in buildings due to the fact that entire spaces would no longer need to be heated or cooled.⁵ Instead, less energy could be used to condition the small volume of space around occupants, while increasing the quality of indoor environments.

Design of such thermo-responsive building systems will have the broader impact of integrating human metabolisms with building thermodynamics. It will make valuable contributions to the architectural community by offering new performance criteria and metrics that allow for novel design strategies that empower occupant experiences (Figure 1). At the same time, this research will inform the general opinion about the value of the thermal environment to human health relative to energy expenditure, topics in which there is a paucity of information.

Figure 1.

Conceptual representation of heat-transfer contributions of building occupants. In the mechanistic approach (left), the thermal environment is homogeneous, based on averaged heat-transfer contributions. The proposed approach (right) assumes a heterogeneous and transient thermal environment, which may accommodate individual needs.

Toward the personalization of the built thermal environment

Challenges and opportunities with thermal comfort assessment

Between the act of design and the experience of architecture lies a rich, and yet relatively unexplored, territory that entwines building performance with human well-being. Providing thermal comfort has been a fundamental prerogative of architecture, not just as a functional requirement, but rather as a profound bodily experience. In traditional northern Chinese houses, the kang, or dikang, was a raised radiant floor-bed that localized heat in one part of the room.⁶ Similarly, in Japanese tradition, the fireplace served as a heating source of the house as well as the center for social activities of the entire family,² a design which has been incorporated by F.L. Wright. The thermal environment has also been described regarding sacredness, pleasantness, and delight, resulting from the contrast between cold–warm experiences, rather than its uniformity.¹ In recent years, much work has been carried out by Philippe Rahm⁷ Architects, Diller and Scofidio+Renfro, Transsolar, and Massachusetts Institute of Technology (MIT) to express the diverse qualities of microclimates relative to people’s experiences, impact on physiological responses, and energy conservation measures.⁸

While architects have utilized heterogeneous indoor thermal conditions to highlight diversified programmatic functions and increase occupants’ well-being, engineering standards have adopted universal thermal comfort assumptions which allow for first-principle quantification of heat-transfer contributions but neglect intra-personal differences in comfort perception. According to ASHRAE Standard 55: Thermal Environmental Conditions for Human Occupancy, large variations in occupants’ preferences are inherently difficult to meet. In response, the normative application of standards aims to minimize the percentage of people dissatisfied and the heterogeneity of thermal conditions.⁹ Recent studies on thermal comfort at the Center for the Built Environment at University of California (UC) Berkley underscored partial- and whole-body differences in sensation under transient and heterogeneous conditions, which may indicate new frontiers for thermal comfort assessment in buildings with varying thermal environments.^10,11

In the present work, the design and management of the indoor thermal environment is taken as an opportunity to reduce energy consumption and improve occupants’ thermal comfort, by integrating the human experience with research that aims to minimize heating and cooling expenditures.^12,13 As urban residents spend prolonged time indoors, up to 87% in the United States,¹⁴ the role of technologies for heating and cooling has become critical for healthier indoor environments.¹⁵ With a broader scope of informing novel performance requirements for air-conditioning technologies, in this article, we focus on the significance of bio-feedback for the built domain and how user experience is taken as decision-making tool for design strategies.

Accounting for individuals: the role of stress

Differences in thermoregulatory processes, given by age and gender, suggest that environmental conditions should diverge from the paradigm embraced by conventional mechanical technologies.^16,17 Furthermore, physiological differences between individuals, combined with psychological and behavioral contributions, should motivate the development of more bio-responsive building systems to address the different natures of individual comfort preferences. Indeed, the underlying assumption is that the more heterogeneous the occupants (e.g. based upon age, gender, health status, etc.), the greater the divide between the thermal environment delivered and the comfort level of occupants.

Physiological or autonomic (involuntary) changes in the body are a manifestation of responses to external stimuli, such as those produced by built environments, as described here. “Stress” has been described most generally as the response of the organism to challenges to homeostasis (balance).^18–20 This response is now recognized to have three interacting components: autonomic (e.g. changes in heart rate and blood pressure not under conscious control), behavioral/psychological (e.g. self-reported mood/emotion ratings), and hormonal (e.g. cortisol levels in saliva). It is currently recognized that stress response can be initiated not only from challenges, but from many common contextual variables as well. For example, an experimental setup showed that heart rate variation (HRV) (a physiological aspect of the stress response) and electroencephalogram (EEG) recordings could be used as valid indices of thermal comfort.^21,22 These findings support previous studies that emotions have direct effects on physiological human processes and, therefore, can be quantified using indirect metrics.

Some studies have focused on the role of interoception as a complex process that affects balance of physiological systems (homeostasis), behaviors, and cognition, relevant for thermal comfort.^23,24 Craig proposes to separate perception of pain, temperature, and other bodily feelings from touch. Emotions produced by thermal sensations are proposed to be significant for decision making, particularly for discriminating between reward and punishment (which is considered a basic decision-making strategy across mammal species). In light of these findings, it appears reasonable that representation of thermal emotions overlaps with pleasantness of other senses, such as smell, taste, audition, and vision to inform higher processes (e.g. recognition and attractiveness of odor, flavor, and faces, or auditory and visual aesthetics). Therefore, thermal environments engage a significant relationship in modulating representation of emotions, particularly related with affective value of thermal stimuli (not just the perception of its intensity, that is, warm or cold), while helping to satisfy the ancestral needs of safety and protection.

Objectives: an experimental design framework

The proposed approach assumes that buildings are a repository of indoor microclimates, defined as thermal conditions that intentionally link local indoor environments (room scale) with occupants’ conscious and unconscious comfort response (human scale) (Figure 2). The diversification of the traditionally homogeneous indoor environment for heterogeneous and personalized thermal environments is dependent on three notions (Figure 3):

The development on next-generation technologies for heating and cooling is predicated on a decentralized approach toward distributed low-energy-consumption modular systems that can find different embodiments.

The proposed design framework aims to expand the assessment of thermal comfort toward direct and indirect biometric indicators, allowing to register bodily variation in thermal sensation and to anticipate changes of comfort perception.

The successful integration of distributed technologies for personalized comfort with the perceived experience of building occupants requires the characterization of a multi-disciplinary computational framework that is able to register environmental and user-perceived conditions, while providing recursive mechanisms to manipulate the environmental conditions locally.

Figure 2.

Inter-scalar relationship between environmental factors, thermal comfort, and physio-cognitive response.

Figure 3.

Mutual relationship network between responsive thermal systems, personalized comfort assessment, and digital computation.

The primary objective of this research is to determine the fundamental criteria driving the execution of these three points, and to identify the challenges and opportunities related to personalized technologies in buildings. In this study, however, we will only focus on the effects of a radiant heating field on thermal comfort. This represents an important case to investigate as thermal asymmetry plays a critical role in causing potential discomfort.⁹

Methodology

The research method is based on an experimental interdisciplinary collaboration between architecture and cognitive science for the generation and evaluation of spatial choices and its thermal effects on occupants. The research method hinges on a build-to-test approach through an Installation Feedback Loop (IFL), which functions as an engine to correlate simulations and biometric data acquisition with actuating mechanisms of the thermal modules (Figure 4). This approach is effective in addressing the complex relationship between perceived thermal experiences, physiological changes in different environmental conditions and morphological conditions.

Figure 4.

The research methodology pursued during the Smartgeometry workshop focused on the proposal of a spatial cognitive hypothesis and a sequential evaluation and recursive alteration.

A responsive framework: physical interactive installation

To explore the challenges and opportunities related to the creation and control of indoor microclimates for personalized comfort, an interactive installation was set up to probe the effects of radiant asymmetries on thermal comfort and stress indicators. A suspended modular radiant ceiling was assembled during the 2016 Smartgeometry Workshop in Gothenburg, Sweden, by the Atmospheric Delight workshop participants (Figure 5). Since the workshop was conducted during a dominant cold weather condition, testing protocol examined only radiant heating. Additionally, local thermal discomfort caused by radiant asymmetry is greatest for warm ceiling condition when compared to other heated or cooled surface orientations.^25–27

Figure 5.

Suspended canopy populated with radiant units built during the 2016 Smartgeometry workshop by the Atmospheric Delight cluster. Image credit: Bianca Toth.

The modular 10″ × 10″ radiant panels, used to populate the suspended ceiling structure, were composed of 23% perforated aluminum and were coated with a high-emissivity white paint on the downward facing surface. From Figure 6, two electric 12 Volts Direct Current (VDC) heat pads were attached to the backside of the aluminum plate to heat the module. Additionally, the back face of the radiant modules was insulated with 1″ rigid foam to limit unwanted upward heat loss. The modules also included light emitting diodes (LEDs) with the initial intent to investigate the relationship between color perception and thermal comfort; however, these questions were not further studied in this investigation and could be addressed in future experiments.

Figure 6.

Detail diagram of the radiant unit with the key components: the heating pad, LED emitters, aluminum spreader, and thermal insulation.

An 8′ × 10′ foot print was chosen for the installation based on the minimum habitable space allowable in New York City building regulation. This control volume was further delimited by a suspended interlocking grid structure containing 80 active node or attachment points for the module pendant wire. This allowed for multiple configurations of the 40 heating modules with the 80-point grid.

User experience: data acquisition and system control

As a way to recursively control the relationships between user experience and spatial morphology, a digital control tool, which informed the position and orientation of the suspended heating modules, was created in Grasshopper to suggest a real-time control of the surface temperature and color that could be altered through input of data collected by bio-sensors worn by participants. The BITalino device is a low-cost sensor kit developed by PLUX wireless biosignals S.A., the Atmospheric Delight cluster industrial partners for the workshop. The BITalino sensing kit allows users to acquire several types of biometric data, such as pulse, sink conductance, heat rate variation, and lux. The experimental setup considered the virtual boundaries defined by the suspended ceiling structure as spatial boundaries and the articulation of the thermal modules as thermal boundaries (Figure 7). An infrared camera was utilized to detect human presence and activation of thermal modules with the future goal of integrating motion detection with indirect measurement of environmental conditions such as surface temperature. Three different strategies to control interaction were deployed during the workshop: a continuous feedback loop processing biometric data, namely, electrodermal activity (EDA), and regulating activation based on a set threshold; a video processing based on motion detection for predicting position and number of users within a space; and a manual override for baseline self-assessed comfort (Figure 8).

Figure 7.

Smartgeometry installation setup showing the suspended ceiling structure and the control mechanisms, including the infrared camera.

Figure 8.

Interactive protocols established during the Smartgeometry workshop that streams live data directly in the Grasshopper environment enabling real-time control of the modules.

During the Smartgeometry workshop, a real-time data acquisition protocol was developed to establish a communication protocol between biometric sensor data and Grasshopper. BITalino communicates over Bluetooth using Python application programming interface (API). Once the sensor information is in Grasshopper, the data can then inform intentional states of the ceiling installation. The final step of the data flow sequence is to send data to the module controller via serial communication, which was supported by Firefly.

Results of the interactive system

The work conducted during the Smartgeometry workshop examined the different mechanisms that can shape interactive behaviors for personalized comfort in the built environment and identify the driving performance criteria in the development of distributed heating and cooling technologies. The subject population comprised 12 adults (18–36 years old); however, only 8 subjects were tested for stress and comfort response. The population included architecture students, professional designers, and professors.

Analysis of the design morphologies as function of cognitive objectives

Multiple ceiling configurations were proposed over the course of the workshop by the participants with the intention of evaluating the cognitive effects to user experience. The participants pre-tested a functional thermal module to understand how it affects biometric and comfort responses and how the data generated inform spatial modification. Participants had the opportunity to speculate on the effects of certain spatial configurations of the suspended thermal modules and the anticipated effects on user experience and preferred ergonomics. A representational map synthesizes the anticipated effects of design morphologies relative to socialization versus privacy, and to diffuse versus concentrate thermal stimuli (Figure 9). The chosen configuration, characterized by a dome-type spatial orientation of the modules, aimed to create a space to rest and socialize within the Smartgeometry venue. Participants also assumed that a thermally homogeneous environment, provided by the modules pointing toward the lower center of the virtual space, could augment the sensation of being in a protective environment in opposition to the openness of the venue.

Figure 9.

Spatial morphologies explored during the workshop. Chosen configuration to install aimed to promote social gathering.

Qualitative and quantitative examination of the thermal installation

Thermal imaging was used to evaluate the performance of the installation, and it was determined that the surface temperature of the modular units was around 50°C, a value for which ASHRAE⁹ Standards 55 indicates thermal discomfort induced by radiant asymmetry (Figure 10). While thermal imaging aided in evaluating the energy flows between the thermal system and users, it also helped visualize an invisible dimension for cluster participants and visitors. Despite that our initial intentions were to test the effects of color on thermal perception, the use of red LED light was limited to indicate when the heating modules were active (Figure 11).

Figure 10.

Thermal imaging of the installation was used to determine the temperature achieved by thermal modules and the qualitative relationship between active module and participants.

Figure 11.

Image of the installation with functioning thermal modules and other cluster participants.

Thermal comfort perception

To assess the thermal comfort effects of the installation, participants were asked to sit under the installation while it was not in heating mode, and rate their current level of thermal comfort relative to the ASHRAE thermal sensation scale. Once the baseline responses were recorded, the heating modules were turned on for 20 min, a time above which occupants might be considered adapted to the environmental change.⁹ Afterwards, the participants completed an evaluation where they were asked to re-rate their thermal comfort perception when the modules were activated.

The work developed during the workshop (Figure 12) showed self-reported increase in thermal comfort and a reduction of stress (based upon the responses to the question, “Overall, how stressed do you feel at this very moment?”). For the stress measure, four of the eight occupants surveyed responded that they had less stress after the 20-min period underneath the activated modules compared to the baseline test. Two participants reported a reduction from slightly high stress to neutral or slightly low stress; two others had a reduction from low stress or neutral to no stress or low stress, respectively. There were no changes in the other four occupants who reported low or no stress during both tests. For the thermal comfort measure, participants reporting high stress showed an increase in the thermal comfort sensation scale from “neutral” to “slightly warm” when the modules were activated compared to the baseline test when they were off. Instead, participants reporting high stress did not indicate significant effects of heating modules on thermal comfort perception. Only one participant reported being warm during baseline, and then neutral with the active heating modules. Additionally, only one participant reported “neutral” at both tests. Based on the majority of the responses, participants did not report dissatisfaction with the thermal environment related to asymmetric conditions or discomfort.

Figure 12.

Stress and thermal comfort assessment during baseline and with active modules.

Measures of autonomic stress responding were collected for all participants using a wrist cuff to measure systolic blood pressure, diastolic blood pressure, and heart rate. Immediately before completing the questionnaires described above, when the modules were turned off and on, participants took their own blood pressure and heart rate and recorded this on the web-based forms. These results do not show clear differences between participants on these autonomic measures (Figure 13). This pattern of results for systolic blood pressure, diastolic blood pressure, and heart rate is not surprising given that participants’ self-reported stress ratings were low.

Figure 13.

Blood pressure and heart rate assessment during baseline and with active modules.

Although it was not possible to use the BITalino to measure other autonomic functions in all participants, a qualitative examination of two subjects who differed for their subjective response (Figure 14) examined the challenges related to using biometric indicators in the interactive setup, and the promise of using this approach in future studies. A significant amount of work was conducted to enable real-time communication between the device and the Grasshopper platform, which resulted in additional challenges given by raw data processing and noise compensation. At the time of utilization, BITalino was able to stream in real time only raw, unprocessed voltage readings of the EDA, a parameter to test stress in participants.²⁸ The conversion from voltage inputs into EDA was determined through analytical correlation taken from the sensor data sheet; however, it was not possible to mitigate the noise and calibrate the sensor.

Figure 14.

Electrodermal activity (EDA) plot for two females sampled for the first 3 min under radiant ceiling structure with active modules.

Conclusion

We have described an approach to affect and test thermal comfort in human occupants led to incorporate bio-indicators and control mechanisms for personalized comfort. This research offers a sophisticated vision of the thermal environment as a repository of diversified environmental conditions, and negotiates the individual and collective comfort demands. By exploring the valence of physiological and cognitive processes in an iterative experimental setup, this study expands traditional design objectives related to thermal comfort and user experience in space inhabitation. While there is abundant literature on thermal comfort models, very little research focused on the role of distributed thermal technologies and the benefits of personalized comfort. Current medical practice, which has a greater understanding of individual differences as they relate to disease risk and effective treatment, has seen great advances using a “personalized medicine” approach, a paradigm shift in comparison to the use of large-group norms that previously predominated.²⁹ Similarly, the implications of this research approach are significant as thermal environments might have impact on several measures relating to productivity in workplaces,³⁰ indoor environmental quality,¹⁵ and general health, such as obesity^31,32 and emotions.³³ If so, there is a need for a balance between energy-efficient buildings, occupant well-being, and satisfaction; the present describes how this need was addressed in this pilot study.

The acquired data from the metrics described in the methodology section permitted the tracking of potential impacts of localized heating and transient thermal conditions to increase thermal comfort perception and alter stress responses. Results suggested an improvement in thermal comfort score and reduced self-reported stress ratings in subjects under the activated modules, compared to when the modules were off. These patterns of responses occurred concomitant with the modules producing heat, which had greater effects on subjects reporting low stress. At the time of the workshop and due to time restriction, no testing was preformed to examine the relationship between thermal perception, color, and thermal imaging. It is the intention of the authors to incorporate and test those findings further in future studies.

The present results have to be interpreted with caution for a few reasons. There was a small number of subjects that provided self-reports of their experience. There were no clear differences in blood pressure or heart rate among subjects; this could be due to the fact that subjects reported low stress ratings. Importantly, these results show that the experimental setup did not produce stress, despite that radiant asymmetry from heated ceiling has the greatest potential to cause discomfort. While blood pressure and heart rate did not vary significantly during the testing sessions, further studies are required to establish correlations between stress response, discomfort from asymmetric thermal conditions (ASHRAE Standards 55), and measures of autonomic functions. Other biometrics—such as electrocardiogram (ECG), EEG, HRV, and EDA—and real-time sensing will be calibrated and tested on a full-scale prototype in the near future, although non-calibrated tests were conducted to examine the opportunities and challenges (Figure 15). These results, as preliminary as they may be, allowed to explore the challenges related to use and applications of biometric indicators in the daily user experience, which may diverge from controlled environments, used by referenced studies.

Figure 15.

Completed installation during the workshop exhibition controlled using infrared camera for motion detection, with one of the participants testing a batched biometric indicator.

Future studies are planned, based upon these results, to characterize the significance of personalized thermal comfort in buildings and distributed thermo-responsive technologies. The overarching aim of this research is to address two main questions:

Question 1: How do different thermal morphologies address occupant preferences relative to their conscious and unconscious responses? Conventional heating, ventilation, and air-conditioning (HVAC) combines ventilation with heating and cooling, resulting in uniform thermal distribution in space. The work developed by the Atmospheric Delight cluster sets up the basis to assess participatory response in a dynamic environment. A near-term goal is a larger study that assesses the effects of localized heating and cooling on occupants’ thermal comfort and stress response in both residential and commercial environments.

Question 2: How do thermo-responsive architectural systems generate new design solutions while creating personalized thermal identities? The diversified nature of thermoregulatory processes in young persons, adults, and elderly people suggests that environmental conditions diverge from the homogeneous paradigm of current thermal technologies. For example, personalized thermal identities of building spaces might propose cycles of heating and cooling depending on individual differences in metabolic activity or, for example, stress response. A broader question, related to the service and nature offered by biometric data for data driven design, points toward conflicting demands and the extent to which architectural technologies may or may not provide individualized comfort.

This study intends to offer new design solutions and methodologies for creating building thermal identities. The ultimate goal is to characterize the performance criteria and significance of interactive building systems that can leverage comfort metrics tailored to user preferences with different space morphologies, shifting away from the notion of buildings at fixed set-point temperatures. The proposed framework suggests a direction in which biometric indicators may drive participatory design and control strategies, whereby buildings actively respond to the modes of occupations and user experience. Arguably, as the architectural heritage has suggested through the Roman Baths, the Nordic Sauna, the Fireplace, and other exemplars, the role and authority of the designers are in providing choices to occupants and catalyzing value from environmental diversity, rather than assuming a universally valid preference of the masses.

Footnotes

Acknowledgements

The authors would like to thank Dr Hugo Silva for his technical support of BITalino. Also, the authors would like to thank the Research Investigators at the Center for Architecture Science and Ecology (CASE) during the 2016 Spring Semester: Elijah Coley, Amaory B. Portorreal, Allison Turner, and Shi Zhang. Finally, the authors would like to thank the Smartgeometry Directors and the Atmospheric Delight participants: Aaron Zeligs, Alessio Lombardi, Ankita Diwan, Annie-Locke Scherer, Bianca Toth, Diogo Henriques, Dongwook Hwang, Giovanni Campusano, Fredrik Domhagen, Jyoti Kapur, Markus Gustafsson, Nina Åström, Sanna Englund, and Stig Nielsen.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Rensselaer Polytechnic Institute, the Center for Architecture Science and Ecology (CASE), and Smartgeometry.

References

Heschong

Thermal delight in architecture. Cambridge, MA: MIT Press, 1979.

Fernández-Galiano

Fire and memory on architecture and energy. Cambridge, MA: MIT Press, 2000.

Seppanen

Fisk

Faulkner

. Control of temperature for health and productivity in offices. Lawrence Berkeley National Laboratory, 2004.

Stoops

. A possible connection between thermal comfort and health. Lawrence Berkeley National Laboratory, 2004.

Bauman

Zhang

Arens

. Advanced integrated systems technology development: Personal comfort systems and radiant slab systems. 2015.

Bean

Olesen

Kim

KW.

History of radiant heating & cooling systems—Part 2. ASHRAE Journal 2010; 52: 40–47.

Rahm

Meteorological architecture. Archit Design 2009; 79: 30–41.

Ratti

Leigh

Danzmayr

. Local warming. 2013. (accessed 10 February 2017).

ASHRAE Standard 55-2004:2004. Thermal environmental conditions for human occupancy.

10.

Arens

Zhang

Huizenga

Partial-and whole-body thermal sensation and comfort—Part II: non-uniform environmental conditions. J Therm Biol 2006; 31: 60–66.

11.

Arens

Zhang

Huizenga

Partial-and whole-body thermal sensation and comfort—Part I: uniform environmental conditions. J Therm Biol 2006; 31: 53–59.

12.

Hoyt

Arens

Zhang

Extending air temperature setpoints: simulated energy savings and design considerations for new and retrofit buildings. Build Environ 2015; 88: 89–96.

13.

Hoyt

Lee

Zhang

. Energy savings from extended air temperature setpoints and reductions in room air mixing. In: International conference on environmental ergonomics 2009, Boston, MA, 2–7 August 2009.

14.

Klepeis

Nelson

Ott

. The National Human Activity Pattern Survey (NHAPS): a resource for assessing exposure to environmental pollutants. J Expo Anal Env Epid 2001; 11: 231–252.

15.

Bluyssen

PM.

The indoor environment handbook: how to make buildings healthy and comfortable. Abingdon: Routledge, 2013.

16.

Van Someren

. Thermoregulation and aging. Am J Physiol: Reg I 2007; 292: R99–R102.

17.

Kingma

van Marken Lichtenbelt

Energy consumption in buildings and female thermal demand. Nat Clim Change 2015; 5: 1054–1056.

18.

Radley

Kabbaj

Jacobson

. Stress risk factors and stress-related pathology: neuroplasticity, epigenetics and endophenotypes. Stress 2011; 14: 481–497.

19.

Sapolsky

RM.

Why zebras don’t get ulcers: the acclaimed guide to stress, stress-related diseases, and coping-now revised and updated. San Francisco, CA: Macmillan, 2004.

20.

Selye

A syndrome produced by diverse nocuous agents. Nature 1936; 138: 32.

21.

Yao

Lian

Liu

. Heart rate variation and electroencephalograph—the potential physiological factors for thermal comfort study. Indoor Air 2009; 19: 93–101.

22.

Yao

Lian

Liu

. Experimental study on physiological responses and thermal comfort under various ambient temperatures. Physiol Behav 2008; 93: 310–321.

23.

Craig

AD.

How do you feel? Interoception: the sense of the physiological condition of the body. Nat Rev Neurosci 2002; 3: 655–666.

24.

Craig

Chen

Bandy

. Thermosensory activation of insular cortex. Nat Neurosci 2000; 3: 184–190.

25.

American Society of Heating R and Air-Conditioning E. 2009 ASHRAE handbook fundamentals. Atlanta, GA: ASHRAE, 2009.

26.

Fanger

Ipsen

Langkilde

. Comfort limits for asymmetric thermal radiation. Energ Buildings 1985; 8: 225–236.

27.

Halawa

van Hoof

Soebarto

The impacts of the thermal radiation field on thermal comfort, energy consumption and control—a critical overview. Renew Sust Energ Rev 2014; 37: 907–918.

28.

Uğursal

Thermal comfort under transient metabolic and dynamic localized airflow conditions combined with neutral and warm ambient temperatures. Doctoral Dissertation, Texas A&M University, College station, 2010.

29.

Aspinall

Hamermesh

RG.

Realizing the promise of personalized medicine. Harvard Bus Rev 2007; 85: 108–117.

30.

Seppänen

Fisk

Faulkner

Control of temperature for health and productivity in offices. Berkeley, CA: Lawrence Berkeley National Laboratory, 2004.

31.

Johnson

Mavrogianni

Ucci

. Could increased time spent in a thermal comfort zone contribute to population increases in obesity? Obes Rev 2011; 12: 543–551.

32.

Celi

FS.

Brown adipose tissue—when it pays to be inefficient. New Engl J Med 2009; 360: 1553–1556.

33.

Nummenmaa

Glerean

Hari

. Bodily maps of emotions. P Natl A Sci 2013; 111: 646–651.