Modeling Haptic Data Transfer Processes through a Thermal
Interface using an Equivalent Electric Circuit Approach
Yosef Y. Shani and Simon Lineykin
a
Mechanical Engineering and Mechatronics, Ariel University, Ramat HaGolan Street 65, Ariel, Israel
Keywords: Data Transfer, Haptic Thermal Interface, Termo-Electric Cooler, Equivalent Circuit Modeling.
Abstract: Many activities and scenarios today require human-computer interactions (HCI), and since traditional
communication channels such as vision and hearing are often overloaded or irrelevant, there is an increasing
interest in haptic interfaces, specifically thermal. Designing and optimizing an effective tactile interface
requires an easy-to-use simulation tool to reduce the time for empirical experiments. An original modeling
tool was developed in this study to support cutting edge research on human response to thermal stimuli. The
human skin tissue model is developed as an equivalent electrical circuit for simultaneous simulation with a
thermal display scheme and its control circuitry. The simulator enables monitoring heat flows and temperature
variations at any location of the system without intervening in the process itself and inside the skin tissue, for
instance, at the depth of the thermoreceptors. The other generic advantage of performing tests with a simulator
is the ability to adjust the parameters according to the variety of skin types, test conditions, or thermo-display
characteristics, and to simulate the response to different generated stimuli. This report presents the
methodology and structure of the model along with an initial empiric validation and suggests directions for
further research and future implementation.
1 INTRODUCTION
The human thermal sensory modality offers a novel
dimension for transferring information, provided the
presented thermal cues are designed in accordance
with the sensory system properties. Like other senses,
the thermal sense is based on the effect a stimulus has
on the sensory system, and its sensitivity to the
change that occurred. Therefore, the effect of
different properties of thermal stimuli on the sensory
system has been studied, and thermal thresholds have
been determined. The results show that humans are
remarkably sensitive to changes in skin temperature,
especially for cooling (Stevens, 1998). For instance,
we can resolve a difference of 0.02–0.07°C in the
amplitudes of two cooling pulses or 0.03–0.09°C for
warming pulses (Smith, 1977), (Kenshalo, 1976). and
detect thermal stimuli when skin temperature rises by
0.2°C or descends by 0.11°C (at rates above
0.1°C/sec) (Stevens, 1998), (Ho, 2018). These
fundamental properties of the thermal sensory
system, and others, provide a framework for defining
a
https://orcid.org/0000-0002-4251-0725
the optimal characteristics of any thermal stimuli
presented to the skin.
Much of the research regarding thermal displays
has been focused on their use to simulate the thermal
properties of objects in VR scenarios or to facilitate
material identification and discrimination based on its
typical thermal signature (Jones, 2008). Recent
studies have started investigating the feasibility of
conveying information via thermal sensation, i.e.,
using a thermal display to present encoded abstract
information, to be sensed and decoded by touch. First
steps have been made towards designing suitable
thermal cues (Wilson, 2012, 2013), (Singhal, 2015,
2018). The thermal cues must be knowledgeably
designed in order to guaranty perceptual distinction
with high reliability. There are many challenges in
using thermal cues for this purpose, due to the human
factor on one hand and technology or experimental
factors on the other. Some of the major challenges are
the limited and small number of sensations evoked by
changes in skin temperature, the multiplicity of
parameters that influence the human thermal
sensitivity, spatial summation causing poor location
Shani, Y. and Lineykin, S.
Modeling Haptic Data Transfer Processes through a Thermal Interface using an Equivalent Electric Circuit Approach.
DOI: 10.5220/0010262101270135
In Proceedings of the 14th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2021) - Volume 3: BIOINFORMATICS, pages 127-135
ISBN: 978-989-758-490-9
Copyright
c
2021 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
127
capability, thermal adaptation, and thermal dynamics
of the tissue affecting the human response and hence
the sensation, creating the stimuli and monitoring the
response. To date, the concept of using a thermal
display as a means of transferring data has been
established, but the basis for designing the actual cues
requires further study regarding the human response
to dynamic thermal stimuli.
A haptic thermal interface is a device that excites
tactile sensations by relevant thermal stimulation
(Nam, 2005). A thermoelectric cooler/heater is very
suitable to serve as a thermal interface. It responds
fast to variations in the electric driving current and
allows the wave-shape forming of thermal haptic
signals. It also allows various methods of the drive,
including different types of feedback control, such as
controlling the temperature of the transmitting
surface, the amount of sinking/sourcing heat, and
more.
Monitoring skin response to thermal stimuli,
throughout the experiments, is necessary for
understanding the sources of inefficient
identification, and consequently allows stimuli fine-
tuning seeking an unambiguous distinction. This
issue was illustrated at MIT (Singhal, 2018),
(Singhal, 2015), as the skin temperature measured at
skin-thermal display interface, while exposed to
various types of thermal inputs, indicated that
different waveforms such as square waves, sinusoids,
and triangular waves resulted in very similar changes
in skin temperature, and therefore were unlikely to be
perceptually distinguishable. Furthermore,
comparing the easiest stimulus to identify (with 100%
correct responses across participants), that involved a
linear decrease in skin temperature, with the pattern
with the lowest percentage of correct responses (64%)
that involved a linear increase in temperature from
cold to warm, lead the authors to suggest conclusions
that can be used as guidelines for designing thermal
cues. However, measuring skin temperature at the
skin-thermal display interface, is a known challenge.
The main difficulties being that the very act of
measuring influences the result as it enhances local
pressure on the skin and interferes with the heat
transfer process, as well as the fact that the layer of
the skin relevant for warmth and cold sensation is
inherently inaccessible. (Ho, 2017) apparently
neglected these influences for her research practical
purposes and placed the thermistor at the skin-display
interface. Another reported example is (Singhal,
2018) that chose to place a thermistor at the periphery
of the contact area of the skin and the display, out of
the actual interface area. Both approaches achieving
their research goals.
The complicity and the time consumption of these
experiments lead us to develop a unique emulator
capable of predicting thermal response to various
stimuli. The main advantages of this model are that it
enables effortless extensive resetting and retesting, as
well as 'monitoring' temperature at any chosen point
throughout the display - skin array, free of
interference in the heat transfer process. Hence, it
shall support further research allowing a better
understanding of the human thermal sensory
processing and enhancing thermal feedback
applicability.
A universal model of averaged skin tissue that
allows co-simulation with different thermal interfaces
can serve as a useful tool for choosing the optimal
thermal interface. This method makes it possible to
make predictions about parameters in the formation
and perception of a tactile signal such as the attack,
amplitude, delay, etc. before experimenting with
participants.
This paper presents the modeling methodology,
the structure of the developed skin-tissue & thermal-
display model and an initial empiric validation.
Finally, directions for further research and future
implementation are suggested.
2 SCIENTIFIC BACKGROUND
AND METHODOLOGY
A simplified mathematical model of the skin close to
the contact zone was developed, in the form of an
equivalent electrical circuit. The advantage of this
kind of model is that it can be used to simulate
stimulus-response process in conjunction with the
electric current driver model using electrical circuits
simulation software. In addition, we have developed
and built a laboratory system for generating thermal
exciting signals (stimuli). The following is a
description of the laboratory system, the proposed
skin model, and examples of simulating the stimulus-
response process in dynamics.
The heat transfer within tissues and the heat
transfer in solids can be modeled using an equivalent
electrical circuit, as shown in the (Holman, 2002). An
equivalent electrical circuit is convenient for joint
simulation of thermal processes in tissues and in a
thermoelectric module, as well as electrical processes
in drivers and control networks. The system of
thermo-electrical analogies, where voltage is
analogous to temperature and current is analogous to
a heat flow, is summarized in the Table 1.
BIOINFORMATICS 2021 - 12th International Conference on Bioinformatics Models, Methods and Algorithms
128
This system of analogies helps to represent a
thermal system with distributed parameters as an
equivalent electrical circuit with lumped parameters
consisting of resistors (equivalent to thermal
Table 1: Thermo-electrical analogies.
Thermal
quantities
Units Analogues
Electrical
Quantities
Units
Heat, q W Current, I A
Temperature, T °C Voltage, V V
Thermal
resistance, Θ
°C/W Resistance, R
Ω
Heat capacity, C J/°C Capacity, C F
Common
temperature
0°C Common ground,
gnd
0V
resistance) and capacitors (equivalent to thermal
masses). The initial and boundary conditions of the
equivalent circuit are specified as voltage sources
(equivalent to a temperature source) and current
sources (equivalent to heat sources). The LTSpice
program was used in the current study as the
environment for simulations. Any other software for
electric circuits simulation can be used for this goal
as well.
2.1 Modeling the Skin Tissue
The surface of human skin is not smooth. The varying
degrees of moisture, oiliness, elasticity, hairline, and
other parameters make even measurement of the skin
surface temperature, nontrivial. Heat distribution
under the skin surface is not measurable in vivo.
There are many types of models for thermal
processes in tissues. We have chosen the equivalent
electrical circuit model type to simulate it together
with the model of the thermal signal generator. The
basic concept of the presented model is a division of
the tissue into submillimeter layers, so that lumped
parameters can be used to simulate each layer. Since
the overall depth of interest is small, the heat transfer
can be considered one-dimensional (Cui, 1990).
Each layer has a thermal resistance (𝛩, °𝐶 ∙ 𝑊

)
and a thermal capacity (𝐶, 𝐽 °𝐶

). The values of
those
parameters can be calculated based on the data
in Table 2 and using the following expressions:
Θ

(1)
Θ

(2)
𝐶
𝑚
∙𝑐
𝜌𝐴

∙ℎ
(3)
𝐶
𝑚
∙𝑐
𝜌𝐴

∙ℎ
(4)
where 𝐴

is the contact area, k and c are thermal
properties tabulated in, subscripts eand drefer to
epidermis and dermis respectively, h is the thickness
of the corresponding layer.
When building the model, we assume the
temperature of the subcutaneous tissue remains
unchanged throughout the experiment (𝑡𝑠𝑡 37°𝐶),
due to its low thermal conductivity and since the
effect on the skin is performed for a short period of
time. Figure 1 shows a schematic visualization of a
skin structure using data from (Ratovoson, 2010), (Xu,
2008), and an equivalent circuit model of the
fragment of skin close to contact zone. The outer
layer of the skin (stratum corneum) is shown at the
top of the scheme. The dermis layer of tissue is
significantly thicker than the epidermis layer. The
temperature of the inner part of the dermis can be
assumed as constant stimuli independent (Cui, 1990).
The coloured semi-transparent frames show the
approximate location of the various receptors
responsible for tactile sensations within the skin
tissue. The tissue is divided into elements that can be
represented as lumped parameters. The mesh
resolution was chosen empirically so that increasing
the grid resolution does not lead to significant
differences in the simulation results. Thus, the
epidermis is presented as one element, and the dermis
as five identical elements. Each element has its
thermal resistance and thermal capacity
corresponding to the values of Table 2 and
expressions (1)-(4). In normal conditions, the skin is
cooled by ambient air. The ambient air temperature is
depicted as an equivalent voltage source in Figure 1
(b). The thermal resistance of the convective heat
exchange between the skin and the air is shown as an
equivalent resistor. As a result, the temperature on the
skin surface is about (32°C), which is lower than the
temperature of subcutaneous tissue. The voltage at
point 𝑡
is equivalent to the temperature at the depth
of the thermal receptors in kelvin.
Table 2: Skin mechanical and thermal properties (Ratovoson, 2010).
Layer Thickness, h
(m)
Specific heat, C
(J.kg
-1
.
o
C
-1
)
Thermal conductivity, k
(W.m
-1
.
o
C
-1)
Specific mass,
(Kg.m
-3
)
E
p
idermis 0.081
m
3590 0.24 1200
Dermis 2m 3300 0.45 1060
Modeling Haptic Data Transfer Processes through a Thermal Interface using an Equivalent Electric Circuit Approach
129
(a)
(b)
Figure 1: Schematic visualization of the skin structure - (a)
Scheme of the fragment of skin based on (Ratovoson,
2010), (Xu, 2008) and (b) Equivalent circuit model of the
skin at thermal contact zone based on data from
(Gowrishankar, 2004), (Xu, 2008).
2.2 Modeling the Thermal Interface
The laboratory setup is shown schematically in Figure
2(a). The thermostatic hot plate is the substructure of
the thermoelectric driver. The temperature of the hot
plate is mounted at a constant value of about 40C
using the PID temperature controller. The current-
driven solid-state thermoelectric cooler (TEC) is
attached with its hot side to the hot plate. Thus, the
cold side of the TEC, which is in contact with the skin
of the experiment participant, tends to be cooled as
the driving current increases. In such a way, the
maximum possible temperature of the contact surface
is less than 40C, and the lowest possible temperature
value is set to about 5°𝐶. This is a safe range for test
subject at short-time thermal cues. The TEC is
thermally insulated laterally using aerogel, and at the
cold side using Teflon sheet, except for the contact
window, as shown in the
Figure 2
(a).
The stimuli generator has been tested in the
laboratory. It works properly according to design. The
primary experiments were carried out with the
participation of the researchers. There were just
preliminary qualitative measurements without
statistical elaboration. The preliminary results are
presented below in the corresponding section and do
not differ dramatically from the data indicated in the
literature.
In a steady-state or quasi-steady-state case, the
temperature of the contact surface of the TEC is
proportional to the amount of heat pumped from the
~
(a)
(b)
Figure 2: An experimental setup (a) includes the
thermoelectric cooler (TEC), thermostatic plate, thermal
insulation and a current driver. The circuit for simulation
depicted in (b). Solid line represents electrical domain and
the dotted line is used for thermal domain. V and I represent
voltage and current correspondently, T is the temperature in
°C, is the thermal resistance of convective heat transfer.
The subscripts ref, hp, h, amb, sa, contact and c refer to
reference, hot plate, hot surface, ambient, skin-to-air,
contact, and cold surface respectively. The model of TEC
is taken from (Lineykin, 2007), the model of skin is shown
in Figure 1.
surface by the TEC. The amount of heat sourced or
sunk from/to the surface at constant temperature is
proportional to the current flowing through the TEC,
so it can be easily controlled. At the same time, during
transients, a significant part of the heat goes to
heating the cooler itself, which has some heat
capacity. In order to study the thermal transients and
in order to create temperature excitation signals of the
desired shape, we used a model of a thermoelectric
cooler, developed in early research and published in
(Lineykin, 2007). This model can be easily simulated
BIOINFORMATICS 2021 - 12th International Conference on Bioinformatics Models, Methods and Algorithms
130
using a circuit simulator such as LTSpice together
with electrical driver circuitry.
Equivalent circuit model of the TEC has two
electrical terminals (“+” and “-“) and two terminals in
thermal domain (“Th” and “Tc”), as shown in
Figure
2
(b). The system of thermo-electrical analogies,
where voltage is analogous to temperature and current
is analogous to a heat flow, is summarized in the
Table 1.
Figure 2
(b) depicts the simulation circuit of the
laboratory setup. The electrical part of the circuit is a
voltage-controlled current source (Trans-admittance
amplifier). Resistors R
1
, R
2
, and R
sh
are selected in
such a way to provide the trans-admittance gain of 0.1
(A/V). In the figure, the thermal interface system is
loaded by human skin model in thermal domain. The
contact thermal resistance is shown as Θ

. The
principle of skin modeling is explained above.
As an example, let us simulate the effect on the
skin of a thermal signal having a relatively standard
pattern with a trapeze-shaped cooling wave, as shown
in Figure 3. This shape refers to the desired curve
describing temperature change of the skin surface vs.
time, in response to a corresponding thermal
excitation signal. We have designed the shape of the
necessary electrical exciting signal, that when applied
to the TEC will produce the corresponding thermal
signal that is predicted to cause the desired skin
temperature change at the contact zone. The result of
the simulation of the process of excitation and
reaction is shown in Figure 3.
Figure 3: Example of simulating the temperature of skin
exposed to a cold pulse thermal excitation. The upper
waveform represents the TEC current vs. time. The lower
curves show the temperature (1 Vcorresponds to 1°C) vs.
time. The green curve shows the skin surface temperature
at the contact zone. The red curve shows the temperature of
the thermo-display (TEC) under the skin.
3 EXPERIMENTAL VALIDATION
To validate the emulating model, and prove its
reliability and precision, we shall refer to previously
explored pattern-based thermal icons. For the sake of
simplicity and clarity, we chose to focus on one of six
basic patterns reported by (Singhal, 2018): a cooling
step, i.e, an 8 seconds sequence initiating at steady
normal skin temperature, a sudden reduction of 6°C
@ 3°C/sec rate of change (ROC), followed by
temperature maintenance, and return to baseline
@3°C/sec ROC.
3.1 Method
The characteristics of the specific thermoelectric
Peltier element used by the authors of (Singhal,
2018), were entered into the emulator, and a similar
thermal pattern was reproduced. The output simulated
thermal cutaneous response, was then observed in
comparison to the reported empirical data. To
complete the validation process the electrical input
required to produce the simulated designated thermal
cue, was retrieved and applied to an experimental
thermal display setup in our laboratory. The resulting
skin and display temperatures were observed in
comparison to the corresponding simulated data.
3.2 Results
The heading of a section title must be 13-point bold
in all-capitals, aligned to the left with a linespace
exactly at 15-point, hanging indent of 0,7-centimeter
and with an additional spacing of 24-point before (not
applicable to the first title section of the paper) and
12-point after.
3.2.1 Comparison to Reported Data
The thermal display used in prof. Jones' laboratory
(Singhal, 2018, 2015) was composed of a
thermoelectric cooler model TE-127-1.0-2.5, TE
Technology, Inc. 30mm in length and width, attached
to a heat sink. Three monitoring thermistors were
mounted on the system; thermistor #1 on the surface
of the thermal display (used for feedback control)
indicating the thermal patterned display temperature
change, thermistor #2 on the skin at the edge of the
contact area with the display indicating the skin
response to the thermal stimulus, and #3 on the skin
distanced from the contact area indicating the
baseline skin temperature. Similarly, the simulated
temperature was sampled at two locations in the
model; by the TEC and at the TEC-skin contact point.
Figure 4
visually shows the correlation between the
simulated skin response thermal and the reported
empiric skin response to a chosen stimulus.
Modeling Haptic Data Transfer Processes through a Thermal Interface using an Equivalent Electric Circuit Approach
131
Figure 4: Comparison between simulated and reported
signals - (a) Temperatures simulated at skin-display
interface (green, solid) and at the display (blue, dashed),
(b) Temperature measured by thermistor#2 on skin at the
edge of the display (green, solid) and by thermistor #1 on
the display (blue, dashed), in response to a cooling step
stimulus.
3.2.2 Empirical Evaluation
The electric stimulus, i, e. the electric signal that
produced the designated thermal pattern, was
retrieved from the emulator and applied to an
experimental thermal display setup in our laboratory.
The outcoming thermal stimulus was presented to the
thenar eminence, as the examinee placed a hand on it
and waited for a 2 minute thermal adaptation period.
The temperature was measured at three points using
three thermistors. Thermistor # 1 was placed directly
on the display with a layer of aerogel thermally
isolating it from the skin, indicating the thermal
display temperature change, thermistor #2 was placed
directly on the display with no isolating material
making direct contact with the display and with the
skin, and indicating skin temperature at interface
zone, and thermistor #3 was isolated from the display
and in contact with the skin, as shown in
Figure 5
. The
temperatures obtained were recorded and compared
with the corresponding simulation data, see
Figure 6
..
Figure 5: Scheme of experimental thermo-display and
monitoring thermistors' layout.
The goal of the experimental measurements at this
stage was to prove the proper operation of the thermal
display and the correlation between laboratory
measurement and simulation results. The experiment
did not involve a group of participants. The
researchers themselves were the test subjects. In this
study, one of many examples is demonstrated that
illustrates the correlation between simulation results
and laboratory measurements.
4 DISCUSSION
The example of a cooling step brought herein,
together with similar results from the evaluation
process conducted for other patterns as well, clearly
indicates that the presented simulating model stood
up to the expectations.
The emulator predicts cutaneous response to a
given evoking thermal stimulus, and enables
sampling the temperature at any desired location
without intervening in the heat transfer process. It
also enables deriving corresponding electrical
stimulus or predicting thermal response to a given
electrical stimulus. Finally, it allows for adjustments
to be made throughout testing, in any parameter of the
thermodisplay-skin system, in a simple and
convenient manner.
Observing the simulated thermal response vs. the
reference reported empiric one (Figure 4) shows that
the correlation between them is not perfect. Whereas
for the reported experiment results, the curve
indicating the display temperature change is a well-
shaped symmetrical step as predetermined, and the
curve indicating the skin temperature is somewhat a
distorted step, for the simulated output, it is vice
versa. The reason the reported data is such is due to
BIOINFORMATICS 2021 - 12th International Conference on Bioinformatics Models, Methods and Algorithms
132
the experiment description defining the display
temperature as the controlling parameter. Another
iteration with the emulator would have produced a
better correlation, but we specifically chose to strive
for the skin temperature curve to be the well-shaped
step, because that is the more relevant parameter for
thermal sensation. The thermal cue is actually not the
displayed temperature change, but rather the change
sensed by the skin.
Figure 6: Comparison between measured and simulated
signals:
(a) Temperatures measured by thermistor #2 at the
skin-display interface (green, solid), by thermistor #1 on the
display with isolation from the skin (blue, dashed) and by
thermistor #3 on the skin with isolation from the display
(brown, doted) all in response to the cooling step stimulus,
(b) Temperatures simulated at skin-display interface (green,
solid), at the display (blue, dashed) and inside the dermis
layer (brown, doted).
The locations of the monitoring thermistors reveal
a difference between the reference reported
experiments and the experiment described herein.
The thermistor monitoring the display temperature
(marked #1) was mounted the same way, i.e. at the
contact zone with isolation from the skin. The
reported thermistor #2 was used to measure the skin
temperature, and was attached to the skin at the edge
of the contact zone. Whereas we chose to place the
second thermistor adjacent to the first one at the
contact zone, only without any isolating material, as
it was meant to monitor the temperature at skin-
display interface. The different approach was mainly
for practical reasons, and may have a slight
quantitative influence on the magnitude of the
temperature change, but should not have any
qualitative impact on the shape of the curve. The
different technical approaches demonstrate part of the
challenges involved in conducting haptic thermal
sensing experiments, many of which may be avoided
by the presented emulator.
The thermal sense is based on the effect a stimulus
has on the sensory system and its sensitivity to the
change that occurred. It is important to emphasize that
the actual effect on the skin is the resultant of the heat
transfer process that occurs between the thermal
display and the skin following a thermal stimulus.
The heat flow depends not only on the stimulus but
also upon the thermal characteristics of the elements
involved in the interaction, namely the display, the
skin, and the thermal resistance of the contact at the
interface between them. The contact resistance is
derived from many parameters other than the
material's thermal resistance, e.g, the contact area,
quality of contact due to surface roughness, pressure
etc. All these parameters are simulated by the
presented model with electrical elements and can
easily be controlled and adjusted, representing
various people, skin conditions, locations on the
body, etc. as well as contact conditions.
When investigating the haptic thermal sense in the
context of data transferring capabilities, the main
issue is the recognizability of the thermal cues.
Although temperature monitoring on the surface of
the skin throughout the heat transfer process, provides
essential data for understanding the process, however,
the ability to distinguish between cues and recognize
them, derives from the temperature changes that the
thermoreceptors are exposed to, combined of course
with the psychophysical aspects that ultimately
determine the thermal sensation. Actually, at the
biochemical level the receptors really respond to the
temporal absolute temperature (by firing action
potentials at frequencies in accordance to the
temperature), however, since the thermo-sensation is
the outcome of the combined warmth and cool
thermoreceptor's transmissions together with the
adaptation phenomena (at temperatures of 30-36°C)
and other neural and cognitive analysis, at the
functional level the sensory system responds to
changes in cutaneous temperature. The temperature
change that occurs inside the dermis layer at the
Modeling Haptic Data Transfer Processes through a Thermal Interface using an Equivalent Electric Circuit Approach
133
depths of the cold and warmth thermoreceptors, is
determined by the rate of heat flow initiated by
stimulus at the skin-display interface. Under normal
steady-state conditions of body core temperature
(approx. 37°C) and skin surface temperature (approx.
32°C), a natural heat flow constantly occurs due to the
temperature gradient, resulting in, and balanced by,
body heat emitting off the skin. Once skin surface is
evoked by the thermal cue, heat flow changes
accordingly. A cooling pulse increases the
temperature drop, hence the rate of heat flow pumped
out of the body through the skin. A warm pulse only
decreases the total heat flow but the local and
temporal response involves a change of direction. To
conclude this remark, the absolute temperature on the
skin surface does not reveal the thermal sensation,
whereas the temperature change at the
thermoreceptors does indicate the sensation. The
presented emulator will allow exploring the
possibility that the heat flow, simulated by electrical
current, can predict the sensation perhaps even with
greater reliability than temperature change.
As mentioned in the description of the
experiment, a third thermistor (marked #3) was
placed at the skin-display contact zone but with
thermal isolation from the display (see Figure 5). This
was in an attempt to capture the skin response without
the display temperature distorting the measurement.
However, in practice, this caused not only the
thermistor to be isolated from the display, but also a
significant area of the skin was isolated from it,
thereby distancing it from the heat flow between the
display and the skin. This distance dramatically
flattened the temperature curve, due to cutaneous
thermal resistance, as shown in Figure 6(a). It is
interesting to notice the resembling behavior to the
simulated temperature curve sampled inside the
dermis layer, indicating the cutaneous response
sensed by the thermoreceptors, as shown in Figure
6(b). In other words, this preliminary finding
insinuates that a distance on the surface of the skin
may perhaps represent the inaccessible depth, for
practical experimental purposes. Another application
could be using the simulated temperature change
sensed by the thermoreceptors, to design thermal
cues. This obviously requires empirically
characterizing criteria for recognizable thermal
changes. In a more general view, this finding is just
an example of the potential contribution the emulator
may have to advance further research.
It shall be noted that the amplitude of the
temperature change by the thermoreceptors, is
approximately 1°C @ rate of change (ROC) of
approximately 0.2°C/sec. This value is substantially
greater than the human threshold, as detailed above
(see introduction), supporting the fact that the thermal
cue was well noticed by experiment participants.
Known design limitation: The temporal
processing of the human thermal sense, has a great
impact on the ability to design thermal displays, but
at this stage is not simulated or in any way reflected
by the presented model. This is especially significant
when cues include both warm and cool stimuli, as
explained hereinafter. The response time to thermal
stimuli was found to differ between cold and warm
stimuli, with a more rapid response to the cold stimuli
(0.3-0.5 sec for ROC greater than 0.1°C/sec) than
warm stimuli (0.5-0.9 sec) (Ho, 2018), (Nam, 2005).
This finding is consistent with the biological structure
and biochemical function of the thermal sensory
system, with warmth thermoreceptors located deeper
in the dermis in comparison to cool thermoreceptors
(0.3 vs. 0.15 mm) hence, heat transfer initiated at the
interface on skin surface is detectable with delay, and
the different types of nerve fibers innervated by the
receptors cause a conduction velocity difference in an
order of magnitude (0.5-2 vs. 3-30 m/sec). Ho et al.
(Ho, 2017) investigated the physical-perceptual
correspondence of dynamic thermal stimuli. They
showed that for warm stimulation, as expected,
subjective sensation always comes after the
corresponding physical event (change in skin
temperature), in respect to onset and to peak. Whereas
for cold stimulation, although the subjective onset
always follows the physical onset (with smaller delay
than the response to warm onset), the sensation of
cooling peak is accelerated to an extent that it can
even precede the physical temperature peak
temperature. They concluded that the sense of cold is
more transient than the sense of warmth, therefore
responds more readily to transient changes (dT/dt). If
necessary, this issue will be addressed in the future.
5 CONCLUSIONS
The presented equivalent circuit-based simulator was
developed in light of scientific knowledge available
to date that is based on decades of research in the
different relevant disciplines. The model simulates
heat transfer processes in thermodisplay-skin
interactions, and provides a generic capability that
can potentially be used to advance R&D in this
unique and promising field of interest. The simulator
has been validated via several tests and experiments.
The main advantages the simulator offers, are the
following: The ability to monitor temperature
changes during heat transfer processes, at any chosen
BIOINFORMATICS 2021 - 12th International Conference on Bioinformatics Models, Methods and Algorithms
134
location without influencing the very process and
hence the result, including subcutaneous layers. The
ability to conduct extensive tests by readjusting the
parameters according to the variety of skin types, test
conditions, or thermo-display characteristics, and
various stimuli, thereby dramatically reducing the
need for actual experiments. In addition, the ability to
conveniently determine the electrical stimulus
required to produce a designated thermal cue, or
alternatively to predict the cutaneous thermal
response to a given stimulus.
Implementation in Future Work: The general
objective of the research is to develop methods to
effectively use the human thermal sense, as a data
transfer medium an alternative or complementary
channel for various scenarios in which conventional
channels, vision, hearing, and tactile sensing are not
applicable or not sufficient (e.g. enhance a
communication capability for the deafblind (Korn,
2018), transfer messages in noisy/silent
environments). Future study will focus on
diversifying the variety of thermal cues suitable to
convey information, with two specific goals:
Create a broad set of recognizable thermal
icons the research will include evaluating a
variety of pattern-based structured thermal
icons designed in light of the latest reported
findings.
Evaluate new approaches for advanced
thermal icons - overcoming inherent human
limitations due to special summation and other
sensual phenomena.
REFERENCES
C. Stevens Kenneth K. Choo, Joseph. (1998). Temperature
sensitivity of the body surface over the life
span. Somatosensory & motor research, 15(1), 13-28.
Cui, Z. F., & Barbenel, J. C. (1990). The influence of model
parameter values on the prediction of skin surface
temperature: I. Resting and surface insulation. Physics
in Medicine & Biology, 35(12), 1683.
Darian-Smith, I., & Johnson, K. O. (1977). Thermal
sensibility and thermoreceptors. Journal of
Investigative Dermatology, 69(1).
Gowrishankar, T. R., Stewart, D. A., Martin, G. T., &
Weaver, J. C. (2004). Transport lattice models of heat
transport in skin with spatially heterogeneous,
temperature-dependent perfusion. BioMedical
Engineering OnLine, 3(1), 42.
Ho, H. N. (2018). Material recognition based on thermal
cues: Mechanisms and applications. Temperature, 5(1),
36-55.
Ho, H. N., Sato, K., Kuroki, S., Watanabe, J., Maeno, T., &
Nishida, S. Y. (2016). Physical-perceptual
correspondence for dynamic thermal stimulation. IEEE
transactions on haptics, 10(1), 84-93.
Holman, J. P. (2002). Heat Transfer-Si Units-Sie. Tata
McGraw-Hill Education.
Jones, L. A., & Ho, H. N. (2008). Warm or cool, large or
small? The challenge of thermal displays. IEEE
Transactions on Haptics, 1(1), 53-70.
Jourdan, F., Ratovoson, D., & Huon, V. A study of heat
distribution in human skin: use of Infrared
Thermography. In EPJ Web of Conferences (Vol. 6).
Kenshalo, D. R. (1976). Correlations of temperature
sensitivity in man and monkey, a first approximation.
In Sensory functions of the skin in primates (pp. 305-
330). Pergamon.
Korn, O., Holt, R., Kontopoulos, E., Kappers, A. M.,
Persson, N. K., & Olson, N. (2018, June). Empowering
persons with deafblindness: Designing an intelligent
assistive wearable in the SUITCEYES project.
In Proceedings of the 11th PErvasive Technologies
Related to Assistive Environments Conference (pp.
545-551).
Lineykin, S., & Ben-Yaakov, S. (2007). Modeling and
analysis of thermoelectric modules. IEEE Transactions
on Industry Applications, 43(2), 505-512.
Nam, C. S., Di, J., Borsodi, L. W., & Mackay, W. (2005).
A haptic thermal interface: Towards effective
multimodal user interface systems. Proceedings of
IASTED-HCI, 13-18.
Ratovoson, D, Jourdan, F. and Huon, V. (2010). A study of
heat distribution in human skin: use of Infrared
Thermography. EPJ Web of Conferences, 6, 1-8.
Singhal, A., & Jones, L. A. (2015, June). Dimensionality of
thermal icons. In 2015 IEEE World Haptics Conference
(WHC) (pp. 469-474). IEEE.
Singhal, A., & Jones, L. A. (2018). Creating thermal
icons—a model-based approach. ACM Transactions on
Applied Perception (TAP), 15(2), 1-22.
Wilson, G., Brewster, S., Halvey, M., & Hughes, S. (2012,
September). Thermal icons: evaluating structured
thermal feedback for mobile interaction.
In Proceedings of the 14th international conference on
Human-computer interaction with mobile devices and
services (pp. 309-312).
Wilson, G., Brewster, S., Halvey, M., & Hughes, S. (2013,
April). Thermal feedback identification in a mobile
environment. In International Workshop on Haptic and
Audio Interaction Design (pp. 10-19). Springer, Berlin,
Heidelberg.
Xu, F., Lu, T. J., & Seffen, K. A. (2008). Bio-thermo-
mechanics of skin tissues. Journal of the Mechanics
and Physics of Solids, 56(5), 1852-1884.
Modeling Haptic Data Transfer Processes through a Thermal Interface using an Equivalent Electric Circuit Approach
135