2007, Piran, Slovenia

Thermal comfort
A NEW METHOD FOR EVALUATION OF BEHAVIORAL
THERMOREGULATION IN HUMANS.
Daniel (Wolowske) Yogev , Igor B . Mekjavic
Department of Automation, Biocybernetics and Robotics, Jozef Stefan Institute
Jamova 39, 1000 Ljubljana, Slovenia.
Contact person: daniel.wolowske@ijs.si
INTRODUCTION
Behavioural responses are the most efficient of thermoregulatory responses, since relatively
simple actions can prevent the need to activate the more metabolically costly autonomic
responses. Dependence on behavioural responses may increase in environmental conditions
where the prevailing non thermal factors (NTFs) attenuate autonomic responses and alter
thermal comfort (Mekjavic and Eiken 2006, Mekjavic et al. 1996, 2003). Thermal
(dis)comfort is considered to be the driving force for the initiation of behavioral
thermoregulatory responses (Weiss and Laties. 1961). Thus, alteration in the perception of
thermal comfort by a NTF might jeopardize normothermia by preventing the initiation of
appropriate behavioural responses (Mekjavic et al. 1996). This was confirmed in animal
studies (Macdonald et al. (1989, Pertwee et al. 1986), but data in humans is lacking. The aim
of the current study was to develop a valid protocol that would reliably evaluate behavioural
thermoregulatory responses in humans.
METHODS
Eleven subjects (6 males and 5 females) participated in the study. The experimental set-up
consisted of a water-perfused suit (WPS) with a manual control unit designed to allow the
experimenter or the subject control over the temperature of water perfusing the WPS.
Following familiarization with the equipment and study protocol subjects donned the WPS
and assumed a supine position on a bed. The study consisted of three trials separated by a
minimum of 48 hours. In each trial the temperature of the water perfusing the WPS varied
from 27°C to 42°C and back to 27°C, at a rate of 1.2°C min -1 . Subjects were requested to
indicate the moment they perceived the temperature changing from a comfortable to a slightly
uncomfortable level. One of the trials (counterbalanced between the subjects) was followed
by a 60-min subjective control trial during which the temperature of the WPS fluctuated in a
sinusoidal manner between 27°C and 42°C at the same rate. Subjects were instructed to
maintain maximal thermal comfort by interfering (i.e pushing a button on a remote controller)
with the cooling or warming of the WPS whenever it reached a slightly uncomfortable level.
In all trials, the subjects were naive regarding the absolute temperature of the water perfusing
the suit. The temperature of the water exiting the WPS (Tout; °C) at the moment subjects
reported feeling slightly uncomfortable (trials 1-3) or at the moment they pushed the remote
controller button (trial 4) were recorded and analyzed.
RESULTS
Subjects identified the moment of cold and warm discomfort in a reproducible manner. There
were no significant differences (P>0.05) between Tout chosen in trials 1 to 3. The average
coefficient of variation (calculated as SD/mean*100 of Tout) between trials 1 to 3 was 2.32%
for cold discomfort and 5.5% for warm discomfort.
No significant changes (P>0.05) were found in Tout at the moment subjects reported feeling
uncomfortably warm or cold and the moment they responded to it behaviourally. The Pearson
Correlation coefficient (P.Corr) between the perceived warm discomfort (Average of trials 1-
3) and the average upper limit chosen during trial 4 was 0.84 and between perceived cold
359