Physiological Versus Perceived Foot Temperature, and Perceived Comfort, during Treadmill Running in Shoes and Socks of Various Constructions

Abstract

The purpose of this investigation was to determine whether people could accurately perceive physiological foot temperature during brief bouts of treadmill running in different combinations of shoe and sock models, and also how perception of comfort was influenced. Sixteen young adult males (21.3  0.8 years, 181.8  1 cm, 74.6  1.5 kg) participated in two separate studies where they alternated running and resting for 10 min each with temperature probes attached at two sites on the lateral dorsal aspect of the right foot. Subjects reported perceptions of foot comfort and temperature after each run using 10 cm visual analogue scales. In the first experiment, different sock models were tested with the same shoe model; in the second experiment, different shoe models were tested with the same sock model. Foot temperature did not differ statistically as a function of shoe or sock model in either experiment. Subjects did not perceive any difference in foot temperature in the shoe experiment, but perceived their foot as being cooler when wearing either a polyester sock or a calf compression sleeve and more comfortable when wearing shoes with less mass. Taken together, the results suggest that subjects’ perceptions of foot temperature may not coincide with physiological foot temperature and are more strongly influenced by sock characteristics than shoe characteristics. Further, shoe mass (but not sock fiber weave or composition) may impact comfort perception by subjects.

Full text

AMERICAN JOURNAL OF UNDERGRADUATE RESEARCH VOL. 10, NO. 3 (2011)
7
Physiological Versus Perceived Foot
Temperature, and Perceived Comfort, during
Treadmill Running in Shoes and Socks of Various
Constructions
Rachel M. Barkley, Mike R. Bumgarner, and Erin M. Poss
Biochemistry, Cell and Molecular Biology Program
Drake University
Des Moines, Iowa 50311 USA
David S. Senchina*
Biology Department
Drake University
Des Moines, Iowa 50311 USA
Received: September 11, 2011 Accepted: November 1, 2011
ABSTRACT
The purpose of this investigation was to determine whether people could accurately perceive
physiological foot temperature during brief bouts of treadmill running in different combinations of
shoe and sock models, and also how perception of comfort was influenced. Sixteen young adult
males (21.3  0.8 years, 181.8  1 cm, 74.6  1.5 kg) participated in two separate studies where
they alternated running and resting for 10 min each with temperature probes attached at two sites
on the lateral dorsal aspect of the right foot. Subjects reported perceptions of foot comfort and
temperature after each run using 10 cm visual analogue scales. In the first experiment, different
sock models were tested with the same shoe model; in the second experiment, different shoe
models were tested with the same sock model. Foot temperature did not differ statistically as a
function of shoe or sock model in either experiment. Subjects did not perceive any difference in
foot temperature in the shoe experiment, but perceived their foot as being cooler when wearing
either a polyester sock or a calf compression sleeve and more comfortable when wearing shoes
with less mass. Taken together, the results suggest that subjects’ perceptions of foot
temperature may not coincide with physiological foot temperature and are more strongly
influenced by sock characteristics than shoe characteristics. Further, shoe mass (but not sock
fiber weave or composition) may impact comfort perception by subjects.
I. INTRODUCTION
When shopping for footwear,
runners face hundreds of options that differ
structurally, functionally, and cosmetically.
Runners should select shoes that are
designed for their foot architecture, gait
mechanics, and intended training patterns
because such selections will minimize injury
risk and maximize performance potential [1].
One factor especially important for distance
runners is how well a sock or shoe
dissipates heat, primarily through sweat
convection. Heat dissipation is important
not only for maintaining appropriate heat
dynamics during running but also from a
comfort standpoint.
The sock is critical for maintaining
foot climate because it wicks sweat from the
foot to the shoe upper for evaporation.
* Author for correspondence (DSS):
2507 University Ave., Olin Hall Room 415
Drake University, Des Moines, IA 50311
Tel 1 515 271 2956
Fax 1 515 271 3702
dssenchina@drake.edu

AMERICAN JOURNAL OF UNDERGRADUATE RESEARCH VOL. 10, NO. 3 (2011)
8
Synthetic fibers with better wicking
properties such as acrylic and polyester
have replaced cotton fibers in most athletic
sock models [2]. Cotton fibers are less
conducive to moisture transport than
synthetic fibers, have longer drying times [3]
and, on average, swell 9 more than
synthetic fibers [4]. If the foot is wet,
evaporation of perspiration is reduced and
the cooling effect from sweating is
hampered [5]. Data on the effects of sock
construction on foot temperature are
available but conflicting. A study of five
different commercially-available socks
showed that both sock material and weave
impacted sweat accumulation and foot
temperature during running [6]. In a
separate study, experimenters asked
subjects to run on a treadmill for thirty
minutes in either a standard sock or an
ergonomic fitted sock. There were no
differences in physiological foot temperature
as a result of sock model, though subjects
perceived the ergonomic sock as being
more comfortable and their feet as being
more cool compared to the standard sock
[7]. This suggests that runners’ perceptions
of foot temperature may differ from
physiological foot temperature. One author
team opined that temperature perception is
often confused with perception of sweat
accumulation [5]. Factors such as sock
softness and dryness have also been
correlated with perceptions of comfort [8].
In shoes, the upper is the most
important component when it comes to heat
transfer because it is the component that
wraps around the foot surface. Uppers can
be made of cotton, leather, polyurethane
(PU), polyvinyl chloride (PVC), or other
synthetic fibers [5], and may be thin, thick, or
perforated to allow for air flow. Across a
range of relative humidity, materials such as
PU- or PVC-coated woven fabric have much
better thermal conductivity than materials
such as microporous PU or PU-coated non-
woven fabrics, with various types of leather
exhibiting intermediate properties [5]. Water
vapor permeability is higher for materials like
leather and lower for PVC-coated polyester
or fabric (Boulanger et al. 1976 in [5]). In
terms of comfort perception, the tensile
properties of the upper material [5] and foot
contact perception mediated by the
midsole/insole [9] are also important factors.
Despite this knowledge, we are unaware of
any published studies that have scientifically
characterized the effects of shoe material on
actual and perceived foot temperature
during running.
The purpose of this study was to
test if people could perceive differences in
foot temperature as a consequence of
running in shoes and socks of varying
construction. In the first experiment, three
different sock models and a calf
compression sleeve were tested while
runners wore the same shoe. In the second
experiment, four different shoe models were
tested while runners wore the same sock.
Three hypotheses were tested. Regarding
physiological foot temperature, we
hypothesized that there would be
differences in physiological foot temperature
based on shoe/sock materials such that the
sock with the lowest percentage cotton and
the shoes with the highest amounts of mesh
would be associated with lower physiological
foot temperature (Hypothesis A). Regarding
perceived foot temperature, we
hypothesized that there would be
differences in perceived foot temperature
based on shoe/sock materials such that the
sock with the lowest percentage cotton, the
compression sleeve, and the shoes with the
highest amounts of mesh would all be
associated with lower percevied foot
temperature (Hypothesis B). Finally, we
hypothesized that subjects’ perceptions of
comfort would be inversely proportional to
those of foot temperature as a hotter foot is
perceived as less comfortable (Hypothesis
C).
II. METHODS
The Drake University Institutional
Review Board gave approval for the study
(ID 2009-10088). Inclusion criteria were that
potential subjects had to be male, capable of
exercising safely in a men’s size 11.5 shoe,
and able to run for at least 30 min
continuously. Exclusion criteria were any
injuries or disabilities that precluded running
for that length of time, or that predisposed
one to run with an abnormal gait. Sixteen
young adult males (21.3  0.8 y, 181.8  1
cm, 74.6  1.5 kg) who self-reported as
regular exercisers participated in each of
two separate studies (total n=32). Eight
subjects participated in both the shoe and
sock experiment; thus, anthropometric

AMERICAN JOURNAL OF UNDERGRADUATE RESEARCH VOL. 10, NO. 3 (2011)
9
Figure 1. Experimental technique. Flexible thermometers were attached midway on the lateral
dorsal surface of the right foot. One thermometer was placed directly against the skin (left), and
one thermometer was placed just medial to the same site against the sock surface (middle).
Thermometers were held in place with electrical tape at the measurement site, above the ankle,
and just below the knee. A shoe was then placed over the foot (right).
characteristics did not vary by group (sock
experiment = 21.3  1.4 y, 181.9  1.6 cm,
73.8  2.4 and shoe experiment = 21.3  1.4
y, 181.6  1.9 cm, 75.5  2.6).
Subjects reported to the lab wearing
a T-shirt and shorts. After completing a
medical history questionnaire to ensure it
was safe for them to participate, subjects
were fitted with a heart rate monitor (Polar
Electro Oy) and two flexible termperature
probes or thermistors (YSI Instruments).
One thermometer was placed directly
against the skin surface midway on the
dorsal lateral aspect of the right foot (Fig 1,
left) and held put with electrical tape. We
chose this site from all possible choices
because 70% of sweat secretion from the
foot occurs on the upper surface [10]. A
second thermometer was placed just medial
to the first over the sock (Fig 1, middle) in
similar fashion, and then both probes were
taped at the ankle and just below the knee
before being threaded up through the
waistline. The test shoe was then placed
over the sock (Fig 1, right). Thus, in the
shoe experiment the probes were placed
once initially and never adjusted, whereas in
the sock experiment the second probe had
to be reattached each trial because the sock
changed. We only measured temperature at
one site because a previous study that
examined foot temperature during treadmill
running at multiple sites reported that all
sites showed similar patterns [7] and the
lateral dorsal aspect was deemed least
noticeable by subjects during pilot trials.
The sock experiment tested three
different sock models (all ankle length) and
a calf compression sleeve. The three sock
models varied by fiber composition: one
sock was 100% cotton (“cotton”); one sock
was 53% polyester, 37% cotton, 8% olefin,
and 1% each natural latex and spandex
(“blend”); and one sock was 98% polyester
and 2% spandex (“polyester”). The
graduated calf compression sleeve was
made of 70% polypropylene and 30%
spandex and had a compression level of 26
mmHg. Each subject’s calf circumference
was measured prior to being fitted with
either a medium-sized (12-13.5 in
circumference) or large-sized (14.5-16 in
circumference) sleeve which was worn at
the same time as the blend sock. Subjects
wore the same shoes (Asics GT-2110 size
11.5; Fig. 1, right) for all four sock trials.
The Asics GT-2110 was used only in the
sock trials and never in the shoe trials.
The shoe experiment tested four
different shoe models all men’s size 11.5.
Two different brands were used. For each
brand, we selected one shoe model whose
upper was made of mostly mesh (“mesh”)
and a second shoe model whose upper was
made of leather and/or vinyl with ventilatory
grommets. These shoes will be referred to
as “Brand X Mesh”, “Brand X Solid”, “Brand
Y Mesh,” “Brand Y Solid” in this article.
Subjects wore the same socks (blend) for all
four shoe trials.
In the shoe experiment, all 16
subjects completed a 5 min warm-up at
approximately 70% of the treadmill speed
they wanted to use for the actual trials
followed by a 5 min rest. In the sock
experiment, half of the subjects (n = 8)

AMERICAN JOURNAL OF UNDERGRADUATE RESEARCH VOL. 10, NO. 3 (2011)
10
Cotton
Blend
Blend+Sleeve
Polyester
Warm-Up
Skin Site
+2.06  0.33
+2.23  0.33
+2.68  0.32
+2.14  0.32
Sock Site
+1.95  0.39
+2.35  0.38
+3.15  0.38
+2.84  0.38
Cold-Start
Skin Site
+2.19  0.3
+2.53  0.29
+2.44  0.3
+1.91  0.27
Sock Site
+2.25  0.38
+2.88  0.37
+3.09  0.39
+2.35  0.35
Table 1. Change in physiological foot temperature in the sock experiment after treadmill running
for 10 min. Values are average in C  standard error. There were no statistically significant
differences by site or shoe.
Cotton
Blend
Blend+Sleeve
Polyester
Warm-Up
Comfort
4.0  1.4
4.2  1.3
5.0  1.3
5.7  1.3
Temp.
5.9  0.7*,†
5.8  0.6**,‡
3.7  0.6*,**
4.0  0.6†,‡
Cold-Start
Comfort
6.2  0.9
4.7  0.8
5.4  0.9
5.7  0.8
Temp.
5.9  0.6
5.9  0.6
4.3  0.6
3.8  0.6
Table 2. Perceived foot comfort and temperature in the sock experiment after treadmill running
for 10 min as assessed on a 10 cm visual analogue scale. A “0” indicates most
uncomfortable/hot imaginable whereas a “10” indicates most comfortable/least hot imaginable.
Values are average measurements  standard error. The single asterisk and double asterisks
represent statistically significant differences between the blend+sleeve compared to either the
cotton or blend alone, respectively. The single dagger and double daggers represent statistical
trends towards a difference between the polyester compared to either the cotton or blend alone,
respectively.
completed a similar warm-up and half
started cold (n=8). We did this so that we
could investigate whether a warm-up period
influenced our results, and the data from the
two shoe experiment subpopulations were
analyzed separately. No data was collected
during warm-up.
Trial order in both the shoe and
sock experiments was counterbalanced.
After being fitted with the sock and shoe for
their first trial, subjects ran for 10 min at a
constant speed on the treadmill with foot
temperatures from both the skin and sock
sites recorded pre- and post-run. Subjects
ran at the same constant speed for each
trial. Heart rate was also recorded pre- and
post-run. Subjects were allowed water ad
libitum. At the end of 10 min, subjects
stopped and straddled the treadmill belt so
they could immediately complete two 10 cm
visual analogue scales assessing foot
comfort and foot temperature. The foot
comfort scale was anchored with the phrase
“least comfortable imaginable” on the left
and “most comfortable imaginable” on the
right. The foot temperature scale was
anchored with the phrase “most warm
imaginable” on the left and “least warm
imaginable” on the right. Subjects marked
their perception by drawing a vertical line on
the continuum. Subjects then sat on a
physical therapy table for 10 min to rest.
After the 10 min rest period, subjects
received their next shoe/sock combination
and repeated the cycle until all four trials
were completed.
ANOVA was used to analyze the
results, with “trial order” and either “sock
model” or “shoe model” as included factors.
ANOVA was performed separately for the
sock and skin thermometer sites, heart rate,
comfort perception, and temperature
perception. An alpha level of 0.05 was used
for significance, with statistical trends
defined as p-values between 0.05 and 0.1.
III. RESULTS
a. Sock Experiment
Temperature recordings for all four
trials in the sock experiment are shown in
Table 1 for both the warm-up (n=8) and
cold-start (n=8) groups. Changes in foot
temperature pre- to post-running were not
statistically different for any sock model in
either group (all p>0.504). There were no
trial order effects for foot temperature at
either the skin or sock site in the warm-up
group, nor for the sock site in the cold-start

AMERICAN JOURNAL OF UNDERGRADUATE RESEARCH VOL. 10, NO. 3 (2011)
11
Brand X Mesh
Brand X Solid
Brand Y Mesh
Brand Y Solid
Skin Site
+2.00  0.35
+1.88  0.36
+2.18  0.35
+1.82  0.35
Sock Site
+1.80  0.43
+1.77  0.44
+2.07  0.43
+2.16  0.43
Table 3. Change in physiological foot temperature in the shoe experiment after treadmill running
for 10 min. Values are average in C  standard error. There were no statistically significant
differences by site or shoe.
group; however, there was a significant trial
order effect in the cold-start group at the
skin site (p=0.033) such that later trials had
a smaller increase in foot temperature, as
expected given the lack of a warm-up.
There were no significant differences in
heart rate change seen in either the warm-
up or the cold-start group.
Perceptions of foot comfort and
temperature during the sock experiment are
shown in Table 2 for both the warm-up and
cold-start groups. In the warm-up group,
there were significant differences in foot
temperature perception by sock. Subjects
perceived their feet as being cooler when
wearing the blend + sleeve as compared to
when wearing the blend alone (p=0.034) or
the cotton sock (p=0.026). There were also
statistical trends such that subjects
perceived their feet as being cooler when
wearing the polyester sock as compared to
blend alone (p=0.076) or the cotton sock
(p=0.058). Similar effects, though not
significant, were seen in the cold-start
group. There were no differences in comfort
ratings for either the warm-up group
(p=0.789) or the cold-start group (p=0.669).
b. Shoe Experiment
All subjects in the shoe experiment
completed a 5-minute warm-up prior to data
collection as described in the Methods.
Despite the warm-up, there was a significant
main effect of trial order at the skin site
(p=0.012), and a trend for a main effect of
trial order at the sock site (p=0.086), such
that in both situations the changes in foot
temperature from pre- to post-exercise was
much greater for trial #1 than any
subsequent trial (p0.026). To examine this
further, we removed all data points from trial
#1 and re-ran the ANOVA using only the
data from trials #2-4. There were no trial
order effects in the edited data set.
However, more importantly, all p-values that
were significant in the original analysis
remained significant in the second analysis,
and all p-values that were non-significant in
the original analysis remained non-
significant in the second analysis. This
suggests that the main effect of trial order
did not meaningfully influence the main
effect of shoe condition. Consequently, the
data presented for the shoe experiment
includes all trials.
Temperature recordings for all four
trials in the shoe experiment are shown in
Table 3 (n=16). There were no significant
differences in foot temperature change pre-
to post-running for any shoe model at either
the skin thermometer (p=0.934) or sock
thermometer (p=0.893) sites.
There were no significant
differences in heart rate change and no trial
order effects.
Perceptions of foot comfort and
temperature during the shoe experiment are
shown in Table 4. There were significant
differences in subjects’ perceptions of shoe
comfort (p<0.001) such that subjects
perceived “Brand X Mesh” as significantly
more comfortable than “Brand X Solid” or
“Brand Y Mesh” (both p<0.001), and
demonstrated a trend towards perceiving it
as more comfortable than “Brand Y Solid”
(p=0.087). Subjects perceived “Brand Y
Solid” as being significantly more
comfortable than either “Brand X Solid”
(p=0.031) or “Brand Y Mesh” (p=0.011).
There were no significant differences in
subjects’ perception of foot temperature
across shoes (p=0.123).
IV. Discussion
The main finding of this study was
that even though there were no differences
in physiological foot temperature across the
different shoe and sock trials, subjects
perceived differences in foot temperature in

AMERICAN JOURNAL OF UNDERGRADUATE RESEARCH VOL. 10, NO. 3 (2011)
12
Brand X Mesh
Brand X Solid
Brand Y Mesh
Brand Y Solid
Comfort
6.8  0.5*,**,•
4.0  0.5*,†
3.7  0.5**,‡
5.6  0.5•,†,‡
Temperature
3.5  0.6
5.2  0.6
5.4  0.6
4.6  0.6
Table 4. Perceived foot comfort and temperature in the shoe experiment after treadmill running
for 10 min as assessed on a 10 cm visual analogue scale. A “0” indicates most
uncomfortable/hot imaginable whereas a “10” indicates most comfortable/least hot imaginable.
Values are average measurements  standard error. The asterisk indicates a statistically
significant difference between Brand X Mesh and Brand X Solid whereas the double asterisks
indicates a statistically significant difference between Brand X Mesh and Brand Y Mesh. The dot
indicates a trend towards a statistically significant difference between Brand X Mesh and Brand Y
Solid. The dagger indicates a statistically significant difference between Brand X Solid and Brand
Y Solid whereas the double dagger represents a statistically significant difference between Brand
Y Mesh and Brand Y Solid.
the sock study and in foot comfort in the
shoe study.
In Hypothesis A, we speculated
there would be differences in physiological
foot temperature based on shoe/sock
materials such that the sock with the lowest
percentage cotton and the shoes with the
highest amounts of mesh would be
associated with lower physiological foot
temperature. Neither the sock data (Table
1) nor the shoe data (Table 3) support this
hypothesis as there were no differences
across trials within either experiment. We
initially thought that a shoe upper made from
mostly mesh would allow for better air flow
compared to one that had a solid upper with
ventilation grommets, so these findings were
somewhat surprising (Table 3). Regarding
the sock data (Table 1), other teams have
also shown that sock sweat accumulation
and foot temperature are unrelated to sock
cotton content [6; 7]. For instance, in the
study mentioned earlier involving treadmill
running, the standard sock was 76% cotton
and the ergonomic sock was 44% cotton
and 42% polypropylene, yet no difference in
physiological foot temperature was found
post-run between the two socks. Consistent
with those findings, another team
investigating two sock models of equal
weave but different fiber content (100%
acrylic vs. 100% cotton) wrote that there
were no differences in runners’ perceptions
of foot temperature or comfort between the
two sock models [11].
For Hypothesis B, we predicted that
there would be differences in perceived foot
temperature based on shoe/sock materials
such that the sock with the lowest
percentage cotton, the compression sleeve,
and the shoes with the highest amounts of
mesh would all be associated with lower
perceived foot temperature. Our sock trial
findings partially support this hypothesis
(Table 2) but our shoe trial findings do not
(Table 4). Considering the sock data first
(Table 2), we predicted that the calf
compression sleeve would elicit lower
perceptions of foot temperature. Ratings of
perceived foot temperature were statistically
significantly lower for the blend+sleeve trial
compared to both the cotton sock and blend
sock, but not the polyester sock. Further,
statistical trends were seen for lower ratings
of perceived foot temperature in the
polyester sock compared to both the cotton
sock and blend sock. No differences in
temperature perception were found in the
shoe trials (Table 4).
Our last hypothesis (Hypothesis C)
related to both experiments and was that
subjects’ perceptions of foot comfort would
be inversely proportional to their perceptions
of foot temperature. Stated another way, we
hypothesized that the two measures would
coincide with one another. The data in
Tables 2 and 4 do not support our
hypothesis and suggest that subjects
separated perceptions of foot comfort and
foot temperature in this study. We were
somewhat surprised that there were no
differences in comfort ratings for the sock
experiment (Table 2) given that the socks
were made of different fibers and had
slightly different weaves. It is possible that
the proprioceptive abilities of the foot were
not sensitive enough to detect the
fiber/weave differences, or that these

AMERICAN JOURNAL OF UNDERGRADUATE RESEARCH VOL. 10, NO. 3 (2011)
13
particular fibers/weaves have similar enough
textural properties as to be indistinguishable.
Subjects reported comfort
perception differences in the shoe trials
(Table 4). Whether a shoe was “mesh” or
“solid” did not influence comfort ratings in
this study, as the two shoes with higher
comfort ratings were “Brand X Mesh” and
“Brand Y Solid.” Notably, “Brand X Mesh”
and “Brand Y Solid” had substantially less
mass than “Brand X Solid” and “Brand Y
Mesh” (330.8 and 318.8 g versus 457.9 and
430.9 g, respectively). It is tantalizing to
suggest that mass may have played a role in
comfort perception. To explore this
possibility, we weighed the shoes that our
subjects normally trained in (e.g., their own
shoes). Across the 16 subjects in the shoe
trial the average training shoe mass was
355.4  5.4 g (range 323.2 to 381.6 g).
Therefore, both the “Brand X Mesh” and
“Brand Y Solid” shoes were more similar in
mass to the shoes that subjects normally ran
in and, by comparison, both “Brand X Solid”
and “Brand Y Mesh” had greater mass than
shoes that subjects normally ran in. This
follow-up supports the idea that mass may
have played a role in comfort perception.
Curiously, the two shoes that were
perceived as most comfortable had the two
lowest prices—both “Brand X Mesh” and
“Brand Y Solid” cost $40-45 whereas “Brand
X Solid” cost $125 and “Brand Y Mesh” cost
$100. Superficially our finding that the
lower-priced shoes were rated as more
comfortable than the higher-priced shoes is
congruent with findings reported by another
group. Researchers used low-, mid-, and
high-cost running shoe models from two
different brands to examine plantar pressure
distribution and compared those findings
with subjects’ perceptions of comfort using
visual analogue scales similar to ours [9].
As in our study, they found that the lower-
cost shoes were rated as more comfortable
by the subjects. Shoe mass values were not
reported, and it is unknown whether these
researchers factored shoe mass into their
analysis. If the higher-cost shoes contain
additional materials or parts that are lacking
in the lower-cost shoes (hence “justifying”
the higher price), then the reduced comfort
ratings seen in these studies may be
partially explained by the increased mass of
the higher-cost shoes. To explore this
possibility, we recorded the make and model
of the running shoes that our subjects
normally trained in and retrospectively
determined purchase price using the
Internet based on the most recently-
available model. The average purchase
price was $103  $5.40 (range $60-$140).
This follow-up does not support the idea that
shoe cost and shoe mass are related.
Nevertheless, it is clear that shoe mass is an
important factor to consider when
performing experiments that measure
comfort perception across different footwear.
This study has several limitations.
First, we used only a 10-min running bout.
Most people, especially trained runners, do
not run for only 10 minutes and it is possible
that differences in physiological foot
temperature dynamics are only manifest
after longer running times. (However, one
previously-referenced study discounts this
notion [7]). Second, although a treadmill in
a laboratory environment provided for a
controlled experiment, it does not mimic the
real environment and does not account for
any environmental factors. It is possible that
physiological foot temperature dynamics
would be different under different
environmental conditions. Third, subjects
were able to see the shoes and socks and
may have made inferences based on
preconceptions of or prior experiences with
particular brands or styles that influenced
their perceptions. This problem is difficult to
address; for instance, had we used athletic
tape to mask certain features on the running
shoes, it would have altered ventilation
properties of those shoes, and it would be
extremely difficult to hide dyed or sewn-in
identifying information on the socks. Fourth
and finally, because we wanted to minimize
inter-subject variability our subject pool was
one of convenience and included only males
who could exercise in a men’s size 11.5
shoe. Our results may be different for
females or males of different stature, and we
did not account for training status in our
experiment.
There are several options for future
experiments that build on these findings.
Considering the limitations of the present
study, future experiments could run for
longer times, run in different environments
or terrains, or include a more diverse subject
pool. The role of shoe mass in perceptions
of comfort needs to be better clarified, for
example, it is unclear what threshold of

AMERICAN JOURNAL OF UNDERGRADUATE RESEARCH VOL. 10, NO. 3 (2011)
14
difference is necessary for subjects to
perceive differences in shoe mass, or
whether subjects can judge shoe mass
similarly when wearing shoes versus hefting
them. Wool socks represent a different type
of sock fiber and could be investigated in a
similar model.
ACKNOWLEDGMENTS
We thank all the subjects who participated in
this study for volunteering their time. RMB
was the lead student researcher and peer
mentor on this project, MRB and EMP
helped conduct some sessions, and DSS
was faculty mentor. Students in FYS 29
“Running: Body, Mind, Sole” helped conduct
experimental procedures. Jodi Gullicksrud
performed periodic checks of thermometer
accuracy during the course of the
experiment.
REFERENCES
1. M. Yamashita, Evaluation and selection
of shoe wear and orthoses for the
runner. Physical Medicine and
Rehabilitation Clinics of North America
16 (2005) 801-829.
2. D. Richie, Athletic socks. in: M. Werd,
and E. Knight, (Eds.), Athletic footwear
and orthoses in sports medicine,
Springer, New York, 2010, pp. 69-78.
3. R. Euler, Creating "comfort" socks for
the U.S. consumer. Knitting Times 54
(1985) 47.
4. M. Harris, Handbook of textile fibers,
Harris Research Laboratory, Inc.,
Washington DC, 1954.
5. D. Tollafield, and L. Merriman, Clinical
skills in treating the foot, Elsevier,
London, 1997.
6. E. Hennig, T. Sterzing, T. Brauner, and
J. Kroiher, The influence of sock
construction on foot climate in running
shoes, Footwear Biomechanics
Symposium, Footwear Biomechanics
Group (International Society of
Biomechanics), Cleveland, OH, 2005.
7. A. Purvis, and H. Tunstall, Effects of
sock type on foot skin temperature and
thermal demand during exercise.
Ergonomics 47 (2004) 1657-1668.
8. M. Morris, H. Prato, and N. White,
Relationship of fiber content and fabric
properties to the comfort of socks.
Clothing and Textiles Research Journal
3 (1984) 14-19.
9. R. Clinghan, G. Arnold, T. Drew, L.
Cochrane, and R. Abboud, Do you get
value for money when you buy an
expensive pair of running shoes? British
Journal of Sports Medicine 42 (2008)
189-193.
10. N. Taylor, J. Caldwell, and I. Mekjavic,
The sweating foot: Local differences in
sweat secretion during exercise-induced
hyperthermia. Aviation, Space, and
Environmental Medicine 77 (2006)
1020-1027.
11. K. Herring, and D. Richie, Comparison
of cotton and acrylic socks using a
generic cushion sole design for runners.
Journal of the American Podiatric
Medical Association 83 (1993) 515-522.
www.drake.edu