Personal comfort systems for adults with intellectual disabilities

2.1 THERMAL COMFORT QUESTIONNAIRE

This study adopted an inclusive research approach, valuing the first-person perspectives of individuals with ID. While the reliability of subjective data collection by people with demographics such as elderly, children and people with disabilities has been questioned in previous thermal comfort research (Song & Calautit 2024), this study addresses those concerns through the adaptation of research instruments and the provision of pre-study training. These measures were designed to ensure meaningful and informed participation by all individuals involved.

The thermal comfort questionnaire was adapted to support the participation of individuals with ID, drawing on three instruments: the American Society of Heating, Refrigerating and Air-Conditioning Engineers’ (ASHRAE) point-in-time survey (ANSI/ASHRAE 2023), Brik et al.’s (2021) tool for residential ID facilities, and a thermal comfort questionnaire for Chilean children (Trebilcock et al. 2017). Based on these, a multi-component questionnaire was created to cover thermal sensation (seven-point), preference (three-point), clothing (five-point), activity (five-point) and active PCS use (see Appendix A in the supplemental data online). It was delivered digitally on tablets via Google Forms, with illustrated pictograms to support cognitive accessibility. The instrument was validated through a two-stage process: (1) five trained adults with ID (‘experts by experience’) reviewed and tested the tool; and (2) five academic and professional experts provided feedback. The iterative development process was informed by using inclusive research guidelines established by Puyalto Rovira (2016) and Álvarez-Aguado et al. (2021).

2.2 PARTICIPANTS

This study involved adults with moderate to severe ID. Selection was based on two criteria: (1) living independently; and (2) being in an energy poverty context (as defined in Section 2.5). For individuals with ID, independent living means not requiring daily carer assistance, though it does not preclude support (Puyalto Rovira 2016; Vega 2011). The eight participants resided in two social housing units (four per unit) as part of a ‘Transition to Independent Living’ program, with daily support from the Fundación Coanil. The group included five women and three men, aged 25–57 years (median of 51.5 years), with seven diagnosed with severe ID and one with moderate ID (verified through the national disability certificate). Five participants held paid jobs (mainly cleaning or service roles), while three engaged in unpaid domestic tasks. Several reported chronic conditions (e.g. diabetes, hypothyroidism, cardiovascular or mental health issues), managed with medication. An accessible consent form was provided which was individually accepted and signed by all eight participants. Appendix B in the supplemental data online presents detailed information about each participant.

2.3 DWELLINGS

The study was conducted in two identical semi-detached social-housing units built in 2012 under the thermal regulations in force at the time of construction. The roof was constructed with Zincalum metal sheeting. The external walls comprised reinforced brick masonry on the first floor and timber framing on the second floor, with interior gypsum board lining and exterior Smart Panel cladding. The ground floor consisted of a concrete foundation with ceramic floor finish; the second floor was a reinforced concrete slab with a floating floor finish. The windows were standard sliding aluminum frames with single glazing. Figure 2 shows a plan view of the dwellings.

2.4 EXPERIMENTAL PROTOCOL

In-home field study campaigns were conducted in winter (June–August 2023) and summer (December 2023–March 2024) to examine the relationship between PCS use and self-reported thermal sensation and preference among adults with ID. The winter campaign lasted 10 weeks and the summer campaign 14 weeks. Data collection integrated: (1) a daily point-in-time questionnaire (see Appendix A in the supplemental data online); (2) continuous indoor dry-bulb air temperature monitoring at 15-min intervals using HOBO MX data loggers (Onset, Bourne, MA, US); and (3) individual interviews conducted pre-season and post-season (see Appendix C online). Sensors were calibrated prior to each seasonal campaign. Because data collection was aligned with participants’ daily routines and willingness to respond, questionnaire completion did not follow a fixed time-of-day schedule or measurement sessions, and individual response density was uneven (see the Limitations section). Importantly, this was not a laboratory-style experiment with prescribed ‘with/without PCS’ sessions. PCS use was entirely voluntary; therefore, ‘Using PCS’ and ‘No PCS’ reflect self-selected situations rather than controlled conditions. Outdoor dry-bulb temperature data were retrieved from the Visual Crossing platform using hourly records for each study period (Figure 1).

Figure 1

Measures used in the field study were as follows:

• Self-reported outcomes (see Appendix A in the supplemental data online)

  Thermal sensation was reported daily using the seven-point ASHRAE scale (very cold to very hot), and thermal preference using the three-point ASHRAE scale (cooler to warmer). Thermal preference ‘No change’ was later used as a proxy for thermal comfort. PCS use was self-reported as ‘currently active’ at the time of questionnaire completion; the study did not systematically log duration of use, device settings (such as fan speed), or concurrent adaptive actions. Interviews provided complementary accounts of typical use (such as overnight blanket use), but these patterns should be interpreted qualitatively (see Appendix C online).

• Environmental exposure

  Indoor dry-bulb air temperature (°C) was recorded every 15 min; outdoor dry-bulb temperature was obtained hourly from Visual Crossing (visualcrossing.com).

• Energy context

  Household energy use was characterized by using electricity bills and gas cylinder consumption for space heating, complemented by public data from Energia Abierta (http://energiaabierta.cl/) to contextualize seasonal consumption relative to local per capita medians.

• Scope of variables

  Although additional parameters (such as CO₂ and relative humidity—RH) were recorded as part of the broader doctoral dataset, they are not analyzed here to maintain focus on the association between PCS use and thermal sensation/preference. The implications of excluding these parameters are acknowledged as a limitation (see the Limitations section).

2.5 PCS USED

Prior to the study, residents already used PCS to improve comfort, as is common in homes outside thermal comfort zones (Pérez-Fargallo et al. 2025; Porras-Salazar et al. 2020). In winter, each house had one natural gas stove, and one also had an electrical resistance forced-air heater (RAF 2000W) (Figure 2, a–b). In summer, both houses used medium-sized pedestal fans.

Figure 2

During the study, additional PCS were introduced considering the socio-economic context of the case study and prioritizing devices that were commercially available and economically accessible in the Chilean market, while offering high potential relevance and effectiveness for users. Accordingly, one personal heating technology (personal electric blankets; Pekatherm 0120D) for winter and two personal cooling technologies (small personal USB fans; Spacezat, 148 mm diameter, 4 W; and large pedestal fans; VP35-DC, 415 mm, 30 W) for summer were introduced. Electric blankets and USB fans were provided on a basis of one device per participant (Figure 2, c–d). In addition, a large pedestal fan replaced the older fan in each dwelling (one per house), offering oscillation and eco/night modes to better meet warm-season ventilation needs (Figure 2, e).

Participants were free to use (or not use) the PCS as they wished within each season; device adoption was not enforced. Prior to deployment, each participant received an individual introduction to the new devices, including their functions and possible uses, to support familiarization. Adaptation strategies such as window/door opening were not systematically recorded in this study and are therefore considered a limitation for interpreting warm-season outcomes (see the Limitations section).

2.6 ENERGY POVERTY MEASUREMENT

Researchers have defined energy poverty through multiple approaches, ranging from objective criteria, such as energy consumption, energy costs, and household income, to subjective dimensions, including personal assessments of thermal comfort and perceptions of housing conditions (Waddams Price et al. 2012; RedPE 2019). This study prioritized the ability to achieve adequate indoor temperatures, identified by the Chilean Energy Poverty Network (RedPE) as the most relevant dimension for characterizing habitability. Over- and under-consumption indicators, which define thresholds by location and season using public energy data, were also used (Cerda-Fuentes & Pérez-Fargallo 2024).

Colina-specific reference data were obtained from the Open Energy Platform (http://energiaabierta.cl/). These provided seasonal median (M) consumption values. M was adjusted by using the average household size from the latest census (INE 2017) to estimate per capita values. In the case study dwellings, winter and summer electricity and gas consumption for heating were recorded. Gas use was converted to kWh using its lower heating value, per national guidelines (Ministerio de Energía 2018).

2.7 ANALYSIS

Statistical analyses were undertaken in R Studio (v2024.04.1) using dplyr, readr, and stats. To evaluate the association between PCS use and thermal comfort, within-participant differences in thermal preference responses between ‘Using PCS’ and ‘No PCS’ were analyzed. Because responses were repeated within individuals, collected under naturalistic conditions, and not obtained through prescribed ‘with/without PCS’ sessions, inferential analyses were conducted at the participant level. The proportion of thermal preference votes for each season was based on participants’ indications of ‘No change’ (used as a proxy for thermal comfort) separately for ‘Using PCS’ and ‘No PCS.’ The within-participant difference was calculated thus:

A Wilcoxon signed-rank test (two-sided; α = 0.05) assessed whether the median within-participant Δ differed from zero. Indoor temperature distributions were non-normal; therefore, medians, and interquartile ranges (IQRs) are reported.

Visualizations (heatmaps, density plots, box plots, stacked bars) were made with ggplot2 and gridExtra. To aid interpretation, each participant was assigned a composite label with socio-demographic and contextual variables. Labels include gender, age, ID level, and dwelling. For example, a participant labeled ‘F 42 sev 2’ corresponds to a 42-year-old female with a severe ID residing in dwelling number 2.

Thermal comfort metrics have been formalized in standards such as ASHRAE 55 and ISO 7730, and operationalized through tools such as the Center for the Built Environment’s (CBE) Thermal Comfort Tool (ANSI/ASHRAE 2023; ISO 2005; Tartarini et al. 2020). To contextualize monitored indoor temperatures, reference thresholds derived from the ASHRAE 55 comfort zone under representative residential assumptions were used. Although clothing insulation (clo) and activity level (met) were recorded as part of the broader dataset, they were not incorporated into the threshold calculations in this article to maintain the analytic focus on the relationship between PCS use and self-reported thermal sensation/preference. Accordingly, thresholds were derived assuming sedentary activity (1.0 met), still air, and 50% RH, with winter clothing set at 1.0 clo and summer clothing at 0.4 clo, yielding a lower comfort limit of 21.5°C for winter and an upper comfort limit of 28°C for summer (Huizenga et al. 2024).

In this study, thermal agency is defined as the ability of individuals to make choices and act upon those choices to regulate their thermal environment, specifically focusing on behaviors such as operating PCS. It was inferred from seasonal PCS usage frequency and interviews on thermal adaptation behaviors. Interview narratives and usage logs informed assessments of thermal agency.

3.1 INDOOR TEMPERATURE AND ENERGY POVERTY

The winter data (June–August) show consistent temperature patterns, with outdoor minimums ranging from –2.7 to 2.7°C and maximums up to 25.3°C, rarely exceeding 20°C, except between 15.00 and 17.00 hours (Figure 3, a). Indoor maximums ranged from 23.2 to 26.8°C, while minimums stayed above 8.8°C. However, 92.6% of indoor temperatures remained below the lower thermal comfort limit of 21.5°C, especially in the early mornings and evenings, times when homes were most occupied. These findings suggest partial insulation effectiveness. Median outdoor and indoor temperatures were 11.0 and 17.5°C, respectively, indicating a thermal gap maintained by the envelope (Figure 3, c).

Figure 3

During summer (December–March), outdoor minimums ranged from 8.9 to 12.0°C and maximums from 32.2 to 35.4°C. Temperatures under 15°C occurred mostly between 01.00 and 10.00 hours (Figure 3, b). Indoors, maximum temperatures reached 36.6°C, and minimums stayed above 18.3°C. The buildings were hot: 64.1% of the time temperatures exceeded 26°C, 41.3% were over 28°C, and 22.4% surpassed 30.0°C. This highlights the limited capacity of the homes to buffer heat. Median indoor temperature was 27.1°C, compared with 21.8°C outdoors (Figure 3, d). This is relevant, especially considering that high temperatures persisted into the evenings and nights, with high occupancy of homes.

Overall, the homes rarely maintained indoor temperatures within comfort ranges, despite complying with national insulation standards.

Figure 3 indicates that in winter, indoor temperatures remained above 8.8°C, with a median of 17.5°C, showing some insulation from outdoor cold. However, 92.6% of the time the temperature was below the comfort threshold of 21.5°C (marked in green). In summer, indoor temperatures peaked at 36.6°C, with a median of 27.1°C, often exceeding outdoor values and revealing poor heat protection. Indoor temperature in summer exceeded the upper thermal comfort limit of 26ºC 64.1% of the time.

Concerning energy consumption, the city thresholds for per capita varied seasonally, reflecting differences in thermal demand. For winter, the lower threshold was set at 54.92 kWh per person, while the upper limit for adequate provision reached 219.67 kWh. In summer, the thresholds were slightly reduced, with a lower limit of 52.40 kWh and an upper limit of 209.59 kWh.

In winter, both dwellings surpassed the lower threshold: 98.21 and 85.21 kWh per person, respectively, exceeding the minimum by 43.29 and 30.29 kWh, respectively, but still well below the upper limit, suggesting moderate energy use. In contrast, during summer, both dwellings reported consumption levels below the minimum threshold. Dwelling 1 consumed 26.25 kWh, and dwelling 2, 28.50 kWh, representing deficits of 26.15 and 23.90 kWh, respectively. These figures also fall more than 180 kWh short of the upper threshold, suggesting notable underconsumption during warmer months.

These patterns may reflect limited access to cooling or energy-saving behaviors driven by economic constraints. Winter use remained near the lower threshold, suggesting minimized heating, possibly due to costs or lack of efficient systems. It remains unclear whether these are user choices or structural limitations.

3.2 THERMAL SENSATION AND PREFERENCE

A total of 489 thermal comfort responses in winter and 561 in summer were collected, forming the basis for the analysis.

In winter, 66% of participants reported feeling slightly cool, cool, or cold (Figure 4, a), despite using both existing and new PCS. ‘Neutral’ sensation (represented in green) was low (16%). Individual differences stood out: F 52 sev 2, for instance, reported 97% of responses on the cold side. The dominant preference was ‘warmer’ (45%), while ‘No change,’ used as a proxy for comfort, averaged 35% (Figure 4, b).

Figure 4

In summer, 74% of participants reported feeling warm or hot (Figure 4, c), while ‘neutral’ sensation remained low (6%). The most frequent preference was ‘cooler,’ and comfort (‘No change’) dropped to 25% (Figure 4, d), 10 points below winter. These findings show that 65% of the time in winter and 75% in summer, people wanted a different thermal environment than that provided. M 25 mod 1, who, despite frequently reporting cold thermal sensation in winter, expressed a high level of thermal comfort (as described by ‘No change’ in the thermal preference question), while in summer, reported almost no comfort.

To further explore thermal comfort, the ‘No change’ preference was analyzed in greater detail, as it reflects participants’ willingness to prefer the current indoor temperature. Figure 5 displays all ‘No change’ votes from both winter and summer in relation to recorded indoor temperatures. In winter, the median temperature in which participants reported ‘No change’ was 17.8°C (IQR = 16.3–19.6°C). In summer, it was higher and broader: median = 27.3°C (IQR = 24.2–30.5°C). Notably, the range of temperatures associated with comfort was broad in both seasons, encompassing nearly the full spectrum of recorded indoor temperatures. In other words, across almost every temperature registered indoors there was at least one instance in which a participant evaluated it as comfortable.

Figure 5

3.3 IMPACT OF PCS

The recorded use of PCS by each participant shows widespread use and significant variation in individual preferences regarding each technology (Figure 6, a). Notably, the only participant with moderate ID (participant 8) demonstrated lower use of shared PCS, such as the gas stove and pedestal fan, whereas individuals with severe ID, identified as more vulnerable (participants F 52 sev 2, F v53 sev 1, and F 54 sev 1), appeared to be the primary beneficiaries of shared-use PCS.

Figure 6

Interviews offered complementary insights. In winter, most participants reported cold indoor conditions, especially in the mornings and evenings. Strategies included wearing layers, drinking hot beverages, staying in bed longer on weekends, and using PCS such as gas stoves, electric heaters, and electric blankets. The latter was valued for its quick warming, though some reported not using it due to forgetfulness or lack of electric outlets (F 42 sev 2, F 57 sev 2). Several accounts explicitly linked thermal sensation/preference to action, illustrating how participants acted upon their thermal needs through PCS use (‘I feel cold, so I turn it [the electric blanket] on’). Interview accounts also indicated that electric blankets were used primarily at night, before and during sleeping time, which may help explain lower questionnaire reports of blanket use, as questionnaire responses often captured point-in-time conditions during daytime at-home routines.

In summer, participants widely valued the small desk fan for its ease and portability, often used at night before sleeping time, or during shared leisure time. Participants also described perceived control and learned operation of cooling devices (‘I like it because it blows air and I can move it. I learned how to use it’). The pedestal fan, despite being more powerful, posed usability barriers. Four participants (M 51 sev 2, F 52 sev 2, F 53 sev 1, F 54 sev 1) struggled with its digital controls, relying on others to operate it. This reduced their autonomy and highlighted challenges in thermal agency, understood here as the ability to express thermal preferences and enact actions to regulate one’s immediate thermal environment. In a smaller number of accounts, participants emphasized ownership and routine use of personal devices (‘This one is mine, I use it at night in the bedroom’), reinforcing the role of individually operated PCS in supporting day-to-day self-regulation. This reduced their autonomy and highlighted challenges in user agency for thermal technologies.

Results indicate that PCS use was associated with higher thermal comfort, operationalized as thermal preference ‘No change,’ particularly in summer. Descriptively, pooled responses show a 14% increase in winter and a 16% increase in summer (Figure 6, b–c). Overall, the proportion of ‘No change’ votes, both with and without PCS, was slightly higher in winter than in summer.

When analyzing the impact of PCS across different temperature ranges, the greatest improvements in thermal comfort were observed in the 15.0–19.9ºC range (Figure 6, d) and the 25.0–29.9ºC range (Figure 6, f), meaning that PCS are most useful when the environment is considered thermally discomfortable. It is worth noting that indoor temperatures below 15.0ºC and above 29.9ºC did not have enough ‘No PCS’ votes to allow for meaningful comparison. The data also show that PCS had less impact in the 20.0–24.9ºC range (Figure 6, e), which is consistent with this being the most thermally comfortable zone, where participants reported lower usage of PCS.

However, despite these improvements, overall levels of thermal comfort remained relatively low in both seasons. Because responses were repeated within participants and PCS use was voluntary (not prescribed ‘with/without’ sessions), the within-participant differences were evaluated with a Wilcoxon signed-rank test on participant-level:

In winter, Δ was small and not systematic across participants (median Δ = –0.014, IQR = –0.083 to 0.069; V [the Wilcoxon signed-rank test statistic] = 17, p = 0.944). In summer, PCS use was associated with a consistent increase in ‘No change’ votes (median Δ = 0.126, IQR = 0.022–0.147; V = 36, p = 0.014), reinforcing the observed warm-season benefit of PCS under real-world conditions.

4.1 LIVING OUTSIDE THE COMFORT ZONE

Thermal comfort, while not a direct indicator of health status, reflects a physiological condition in which the body is not under thermal stress, and thus is unlikely to contribute to heat- or cold-related health risks. In this study, indoor temperatures remained predominantly outside the thermal comfort zone (92.6% below 21.5ºC and 64.1% above 26ºC). While winter registered a higher percentage of time below this threshold (Figure 4, c–d), summer conditions were more concerning due to the extreme indoor temperatures, occasionally exceeding 35°C, and the prolonged periods of exposure experienced by residents, particularly overnight. Some recent studies have proposed upper healthy indoor temperature limits, suggesting thresholds with still air of 28°C (Huizenga et al. 2024) and 26°C (Meade et al. 2024). This issue is particularly critical given the health risks associated with heat exposure (Tham et al. 2020), especially for populations, such as the participants in this study, who meet multiple criteria for vulnerability to thermal extremes.

Despite colder indoor temperatures and longer exposure outside the comfort zone in winter, participants showed higher levels of thermal comfort in winter than in summer (Figure 6). This suggests that individuals may be better able to adapt to low temperatures through behavioral strategies. Interviews indicate that households relied on multiple cold-season strategies such as adding layers of clothing, drinking hot beverages, or using low-tech comfort measures not fully captured in this study. Economic constraints may significantly shape thermal comfort practices in Latin America. In contexts where heating is limited or expensive, individuals often adopt adaptive strategies during colder months. This behavior reflects a response to a broader disposition toward energy-saving practices aimed at reducing household expenses (Pérez-Fargallo et al. 2025; Simões et al. 2025).

In contrast, summer conditions involved frequent and sometimes extreme indoor overheating (Figures 3 and 4). While PCS were reported as in use, the available cooling PCS in this study (fans) were unlikely to fully offset heat exposure; rather than demonstrating elimination of overheating, the evidence suggests that PCS supported partial relief and improved perceived acceptability under heat stress while substantial exposure remained (Figure 6). This seasonal asymmetry highlights a gap in housing performance priorities, as winter performance has historically been emphasized, whereas summer overheating is increasingly relevant for equity and public health under climate change.

Thermal comfort data and the distribution of comfortable temperatures (Figures 4 and 5, respectively) indicate seasonal differences in thermal preference that are consistent with thermal comfort theories, where seasonality plays a key role in shaping thermal expectations. However, the wide range of temperatures rated as comfortable may also reflect the influence of ID, which could be affecting both thermal perception and preference. Previous studies on thermal comfort among individuals with physical disabilities have anticipated this phenomenon. Hill et al. (2000) find that while people with disabilities tend to have similar average comfort responses compared with those without disabilities, their responses show greater variation and less internal agreement. Additionally, it is important to consider how temporal and contextual factors might influence thermal perception. For instance, some participants work outside the home several days a week, leading to prolonged exposure to different thermal conditions. This may influence their perception and evaluation of indoor temperature upon returning home, as well as shape their thermal expectations.

4.2 PERSONAL COMFORT AND THERMAL AGENCY

PCS in this study functioned less as an ‘intervention’ and more as a set of everyday adaptive resources deployed under real-life constraints, where use was voluntary and shaped by accessibility. personal routines, and preferences. This framing matters because it shifts interpretation from ‘PCS effectiveness under controlled exposure’ to ‘PCS contribution to perceived acceptability and agency within constrained domestic environments.’

The seasonal contrast in the paired analysis provides an interpretable signal. In winter, PCS use was not associated with a systematic within-participant increase in ‘No change’ votes (median Δ ≈ –0.014; p = 0.944), whereas in summer PCS use showed a consistent increase (median Δ ≈ 0.126; p = 0.014). Rather than suggesting that winter PCS were ineffective, this pattern is consistent with the idea that, under moderate-to-cold indoor conditions, comfort outcomes may be dominated by behavioral strategies, shared heating practices, and time-of-day use (night-time blanket use), which are only partially captured by point-in-time reports. Conversely, under severe summer overheating, even small local cooling strategies (fans) may become more salient to perceived acceptability, producing a clearer within-participant contrast despite the persistence of high indoor temperatures.

Situating these results in the context of ID adds a further layer: thermal agency is about not only comfort optimization but also supporting autonomy, dignity, and self-determination in everyday life (Puyalto Rovira 2016). When PCS are cognitively accessible and reliably usable, they can operate as ‘participation infrastructure’ for self-regulation, enabling people to express needs, enact preferences, and reduce reliance on others for immediate comfort actions. Conversely, when systems require assistance or are controlled by others, PCS can reproduce dependence and frustration, even if they provide thermal relief.

Additionally, this study draws attention to a broader issue in housing and energy policy: most thermal retrofitting strategies and comfort interventions in low-income housing in Chile and Latin America remain focused on winter conditions, neglecting the increasing impact of summer overheating (Trebilcock-Kelly et al. 2023). As climate change increases the frequency and intensity of heatwaves, promoting summer thermal resilience becomes a technical challenge and a matter of social equity and public health, particularly for populations facing compounded vulnerabilities.

4.3 LIMITATIONS

This study presents several limitations that must be acknowledged. Most notably, the small sample size of eight participants limits the generalizability of the findings to the broader population of individuals with ID. Rather than aiming for representativeness, the research is framed as an exploratory case study that offers an initial approach to inclusive thermal comfort research with adults with moderate and severe ID. One key contribution of this work is to demonstrate that, with appropriate adaptations and support, individuals with ID can meaningfully participate in empirical studies of this nature. However, the reliability of self-reported data is inherently dependent on participant engagement, which, while a common challenge in field research, may be further nuanced when working with cognitively diverse populations.

Additional methodological constraints include the lack of standardized response times for the thermal comfort questionnaire, as data collection was adjusted to participants’ daily routines and willingness to respond, resulting in an uneven distribution of individual responses. The study focused on thermal sensation and preference in relation to indoor temperature, excluding other relevant physiological and environmental parameters, such as clothing insulation (clo), metabolic rate (met), CO₂ levels, air movement, and RH, and other contextual factors (noise, natural ventilation behaviors such as window/door opening, and household ventilation practices), which may influence thermal perceptions in residential settings. The decision to exclude them was based on the specific focus of this article: understanding the relationship between thermal sensation, thermal preference, and the use of PCS among adults with ID. The extended dataset, which includes these variables, is part of a broader doctoral research project and will be analyzed in future studies.

In addition, some PCS were newly introduced for the study; although participants received individual familiarization, short-term novelty, or learning effects that may have influenced early perceptions, and reported use patterns cannot be ruled out.

While energy poverty is recognized as the contextual framework for this research, the study did not conduct a detailed assessment of energy consumption patterns or the specific energy impacts of PCS use, an area that merits further investigation. Finally, individual expectations and lived experiences, which may influence perceptions of thermal comfort, were not systematically explored and represent an important avenue for future research.