Effect of the upward curvature of toe springs on walking biomechanics in humans

Abstract

Although most features of modern footwear have been intensively studied, there has been almost no research on the effects of toe springs. This nearly ubiquitous upward curvature of the sole at the front of the shoe elevates the toe box dorsally above the ground and thereby holds the toes in a constantly dorsiflexed position. While it is generally recognized that toe springs facilitate the forefoot's ability to roll forward at the end of stance, toe springs may also have some effect on natural foot function. This study investigated the effects of toe springs on foot biomechanics in a controlled experiment in which participants walked in specially-designed sandals with varying curvature in the toe region to simulate toe springs ranging from 10 to 40 degrees of curvature. Using inverse dynamics techniques, we found that toe springs alter the joint moments and work at the toes such that greater degrees of toe spring curvature resulted in lower work requirements during walking. Our results help explain why toe springs have been a pervasive feature in shoes for centuries but also suggest that toe springs may contribute to weakening of the foot muscles and possibly to increased susceptibility to common pathological conditions such as plantar fasciitis.

Full text

1
Vol.:(0123456789)
Scientific RepoRtS | (2020) 10:14643 | https://doi.org/10.1038/s41598-020-71247-9
www.nature.com/scientificreports
Eect of the upward curvature
of toe springs on walking
biomechanics in humans
Freddy Sichting1*, Nicholas B. Holowka2, Oliver B. Hansen3 & Daniel E. Lieberman3
Although most features of modern footwear have been intensively studied, there has been almost no
research on the eects of toe springs. This nearly ubiquitous upward curvature of the sole at the front
of the shoe elevates the toe box dorsally above the ground and thereby holds the toes in a constantly
dorsiexed position. While it is generally recognized that toe springs facilitate the forefoot’s ability
to roll forward at the end of stance, toe springs may also have some eect on natural foot function.
This study investigated the eects of toe springs on foot biomechanics in a controlled experiment in
which participants walked in specially-designed sandals with varying curvature in the toe region to
simulate toe springs ranging from 10 to 40 degrees of curvature. Using inverse dynamics techniques,
we found that toe springs alter the joint moments and work at the toes such that greater degrees of
toe spring curvature resulted in lower work requirements during walking. Our results help explain
why toe springs have been a pervasive feature in shoes for centuries but also suggest that toe springs
may contribute to weakening of the foot muscles and possibly to increased susceptibility to common
pathological conditions such as plantar fasciitis.
Most humans today use footwear with numerous features that protect the sole of the foot and increase comfort.
While many features have been intensively studied1, one nearly ubiquitous feature that has been almost entirely
unstudied is the toe spring. is upward curvature of the sole of the shoe below the metatarsal heads orients the
toe box dorsally relative to the rest of the shoe (Fig.1). e toe spring is generally thought to help the forefoot
roll forward during the propulsive phase of walking, between when the heel and the toes leave the ground.
e benets in terms of mechanical work of this rolling motion have already been demonstrated in footwear
with curved, rocker-bottom surfaces2–4. Specically, this rolling motion appears to reduce center of mass work,
although the extent to which the conditions in these studies correspond to the toe springs in conventional shoes
is unclear. To date, no experimental study has examined how the toe spring aects the way the human foot func-
tions during gait, and how it may aect the foot’s vulnerability to injury.
It is well established that the ability to dorsiex the toes relative to the rest of the foot at the metatarsophalan-
geal (MTP) joints is one of the key evolved features that enable humans to walk and run bipedally eectively
and eciently. In addition to having shorter, straighter phalanges, human metatarsal heads are characterized by
more dorsally oriented and mediolaterally broad articular surfaces compared to those of our closest relatives,
the African apes5. During the propulsive phase of walking, the dorsally oriented metatarsal heads in the human
forefoot are thought to increase the range of dorsiexion motion at the MTP joints by providing more dorsal
articular surface area on which the proximal phalangeal base can slide6–10. Although recent research shows that
transverse splaying of the metatarsal heads helps stien the midfoot via the transverse tarsal arch11, it has long
been argued that dorsiexion at the MTP joints also helps stien the foot through a windlass mechanism12. Dur-
ing this action, dorsiexion of the toes tightens the plantar aponeurosis, a broad sheet of highly brous tissue
whose collagen bers span the plantar aspect of the foot from the heel to the toes (for review see13). e increased
tension on the plantar aponeurosis pulls the calcaneus and metatarsal heads towards each other, creating an
upward force that elevates the longitudinal arch, counters compressive forces from above, and stiens the foot as
a whole (Fig.2A). Recent research, however, challenges this traditional perspective of the windlass mechanism.
In a static invivo loading experiment, Welte etal.14 found that raising the longitudinal arch by dorsiexing the
toes actually decreases the longitudinal arch’s stiness. In another static invivo experiment, Farris etal.15 found
open
1Department of Human Locomotion, Chemnitz University of Technology, Chemnitz, Germany. 2Department of
Anthropology, University At Bualo, Bualo, NY, USA. 3Department of Human Evolutionary Biology, Harvard
University, Cambridge, MA, USA. *email: freddy.sichting@hsw.tu-chemnitz.de
Content courtesy of Springer Nature, terms of use apply. Rights reserved

2
Vol:.(1234567890)
Scientific RepoRtS | (2020) 10:14643 | https://doi.org/10.1038/s41598-020-71247-9
www.nature.com/scientificreports/
that the windlass mechanism has little eect on longitudinal arch motion while the arch is experiencing the high
loads associated with push-o. While these ndings are compelling, further verication from dynamic invivo
locomotion is necessary, and the windlass mechanism remains a widely utilized model for understanding the
functional signicance of the longitudinal arch (e.g.,16–18).
Regardless of the extent to which the windlass is a passive stabilizing mechanism, a growing body of research
has shown that the intrinsic foot muscles also play important roles in supporting the longitudinal arch and sta-
bilizing the MTP joints19–21. During propulsive phase, the metatarsal heads and the distal phalanges are the only
points of contact with the ground on the trailing leg and hence become load-bearing. As a result, the ground
reaction force loads applied to the distal phalanges generate a moment that causes the MTP joints to dorsiex
(Fig.2B). Electromyographic studies indicate that the intrinsic muscles of the foot, especially the exor digitorum
brevis and abductor hallucis, are active at the end of stance phase, balancing the dorsiexion moments at the MTP
joints (Fig.2C)19,22. According to these ndings, proper intrinsic foot muscle activity, therefore, acts in concert
with passive mechanisms such as the windlass to maintain foot stability during propulsion.
Because of the role that intrinsic foot muscles play in stabilizing the forefoot, weakness or dysfunction of
these muscles may be associated with a variety of overuse injuries including plantar fasciitis23,24. is pathological
inammation causes pain and immobility in more than 2 million patients each year in the United States, mak-
ing it the most common condition encountered by podiatrists25. Etiologically, plantar fasciitis is recognized as
an injury caused by excessive and repetitive loading of the foot’s longitudinal arch26. Recent evidence suggests
that plantar fasciitis could be related to weak foot muscles that are not strong enough to provide foot stability,
thus increasing strain in the plantar fascia, which wraps around the MTP joints, presumably aecting their
stability27. Several lines of evidence suggest that weak foot muscles may be partly a consequence of features in
modern shoes that support the longitudinal arch and passively stien the foot21,28,29. As these studies showed,
individuals who habitually wear minimal footwear have intrinsic foot muscles with large cross sectional areas
and dynamically stier longitudinal arches than individuals who habitually wear modern shoes. Weak intrinsic
Figure1. A toe spring describes the curve upward of the sole of a shoe. (A) e upward curvature below
the metatarsal heads orients the toe box dorsally relative to the rest of the shoe. (B) Custom-made sandals
with varying degrees of toe spring angle were manufactured to mimic the stiness and shape of toe springs
commonly found in commercially available shoes. (C) e sandals were secured with minimal rope laces that
could be adjusted by a buckle and did not restrict the placement of reective markers.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

3
Vol.:(0123456789)
Scientific RepoRtS | (2020) 10:14643 | https://doi.org/10.1038/s41598-020-71247-9
www.nature.com/scientificreports/
foot muscles may thus be an evolutionary mismatch caused by the foot not being entirely adapted for modern
shoes30. Until recently, humans were either barefoot or wore minimal shoes. Although the rst evidence for
minimal footwear dates back to 10,000 years ago31,32, most shoes until very recently were minimal and did not
have arch supports, cushioning, and other supportive features that increase comfort and reduce the work that
the foot muscles have to do33.
Here we focus on how toe springs aect the foot’s ability to function as a sti lever, especially during the
propulsive phase of stance. While it is generally recognized that toe springs facilitate the forefoot’s ability to roll
forward at the end of stance, toe springs may also have some eect on arch stiness via the windlass mechanism.
It is reasonable to hypothesize that toe springs continually engage the windlass mechanism by permanently
orienting the toes in a dorsiexed position when they might otherwise be in a neutral, horizontal position and
thereby elevate the arch. Without a toe spring, loading the arch should cause a ‘reverse windlass’ eect in which
the toes are plantarexed as the arch is compressed during walking or running12. However, a toe spring could
prevent that motion from occurring, eectively stiening the arch by preventing compression. is stiening
eect should be pronounced at midstance, when the foot is loaded by body mass prior to dorsiexion of the toes
at heel li. Following this traditional perspective of the windlass mechanism, a toe spring could thus passively
reduce the need for intrinsic foot muscles to actively resist arch deformation. Another related eect that toe
springs could have on the foot concerns energy loss at the MTP joints during the propulsive phase of each step.
It is well established that the digital exor muscles do a signicant amount of work as the MTP joints dorsiex
Figure2. e ability to dorsiex the toes relative to the rest of the foot at the metatarsophalangeal (MTP)
joints during the propulsive phase is one of the key evolved features that enable humans to walk and run
bipedally eectively and eciently. (A) Dorsiexion at the MTP joints helps stien the foot through a windlass
mechanism. During this action, dorsiexion of the toes creates tension in the plantar aponeurosis that tends to
pull the calcaneus towards the metatarsal heads. is motion creates an upward force in the longitudinal arch.
(B) During propulsive phase, the metatarsal heads and the distal phalanges are the only points of contact with
the ground on the trailing leg and hence become load-bearing. As a result, the ground reaction force (vGRF)
acts on thedistal phalanges at a distance R from the MTP joint center togenerate a moment that causes the
MTP joints to dorsiex. (C) e intrinsic exor muscles are active (Fex)at the end of stance phase, balancing
the dorsiexion moments at the MTP joints(with r as the lever arm of the acting exor muscles).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

4
Vol:.(1234567890)
Scientific RepoRtS | (2020) 10:14643 | https://doi.org/10.1038/s41598-020-71247-9
www.nature.com/scientificreports/
during this phase34, and previous studies have estimated that the work done by the digital exor muscles is
proportional to the amount of MTP joint rotation during push-o35. By passively dorsiexing the toes before
push-o, a toe spring could thus decrease the total angle through which the toes rotate while these muscles are
active. ese eects on foot biomechanics would reduce the total work required of the intrinsic foot muscles,
possibly helping to explain their observed atrophy in individuals who habitually wear modern shoes.
Although toe springs aect foot biomechanics during walking and running, this study explores how the toe
spring aects intrinsic foot biomechanics during walking because it is the most common gait. While toe springs
may have general eects on overall gait, as has been demonstrated in studies of prosthetic toe shape and shoe
midsole stiness2,18, here we focus on the immediate eect of toe springs on intrinsic foot biomechanics to test
discrete hypotheses about how they potentially aect foot function. We focus on the medial longitudinal arch
and the MTP joints during midstance and propulsive phase and use kinematic and force data to test the general
hypothesis that shoes with a toe spring will aect stiness of the foot-shoe-complex and the total work done
at the MTP joints. We also test two specic hypotheses. Hypothesis 1 is that during midstance, the stiness of
the medial longitudinal arch will increase with greater toe spring angles since the dorsiexed position of the
toes activates the windlass mechanism. Hypothesis 2 is that during the propulsive phase, increasing toe spring
angles will gradually decrease the total angle through which the toes rotate and subsequently decrease the total
work at the MTP joint.
Methods
Participants. Data were collected from 13 participants (9 male, 4 female), ranging in age from 19 to 33years
old (mean ± SD: 22 ± 3.1years). Average weight was 74 ± 7.5kg and average height was 182 ± 6cm. All partici-
pants were apparently healthy and had no current injuries or conditions that would cause gait abnormalities.
Written informed consent was obtained from each subject. e study protocol was approved by Harvard’s Com-
mittee on the Use of Human Subjects and conducted in accordance with the Declaration of Helsinki.
Footwear design. Participants walked on the treadmill barefoot and in four pairs of custom-made sandals
with varying degrees of toe spring angle. e sandals consisted of a top sole, rubber outsole, foam midsole (thick-
ness 2mm), and curved berglass plate that ran the length of the sandal and curved upwards at the ball of the
foot to the tip of the sandal (Fig.1B). e upwards curvature under the toes was either 10°, 20°, 30° or 40°. e
10° condition was chosen as the lowest prole to ensure a minimum of natural foot roll-over during the pro-
pulsive phase. e sandals were secured with minimal rope laces that could be adjusted by a buckle and did not
restrict the placement of reective markers. Two sandal sizes were used, depending on the participant’s foot size
(24cm and 28cm length). We chose to use sandals rather than shoes for this study because of their relative ease
of construction, and because they allowed us to place a detailed marker set on the foot (see below).
e sandals were designed to mimic the stiness and shape of toe springs commonly found in commercially
available shoes36. Before the experiment, the bending stiness of the sandal was measured with a uniaxial tensile
and compression testing machine (Model HC 10, Zwick GmbH & Co. KG, Ulm, Germany). e test set-up for
measuring shoe bending stiness has been described in detail elsewhere37. In brief, the rearfoot portion of the
sandal was clamped down on a xed platform set to align the rotational axis of the machine with the anatomical
MTP joint bending axis. e distance between the midpoint of the metatarsal axis and the force application line
was 50mm. e sandal was bent by liing and lowering a sha by 40mm. Using the corresponding force to
the deformation curve enabled the calculation of torque and the bending angle. e average bending stiness
was calculated based on the torque–angle curve from 10 loading cycles at 2Hz. e measured average bending
stiness of the sandal was 6.38 ± 1.58 Nm/rad, being similar to the bending stiness of modern shoes (Adidas
adizero: 7.00 Nm/rad, Nike Zoom Streak 6: 9.4 Nm/rad)38.
Experimental treatment. Participants walked on a split belt treadmill (Bertec Corporation, Columbus,
OH, USA) instrumented with separate force plates under each belt, which were used to measure the ground
reaction forces (GRF) in each leg individually for each step. e order of the sandals and barefoot condition were
randomized. Before the barefoot walking trial, participants were also recorded standing still for 10s. is was
done to provide a neutral representation of each participant’s foot so that we could normalize the joint angles
calculated during walking to each participant’s neutral foot posture, where joint angles were set at 0 degrees.
Participants walked in each condition for a few minutes until they felt comfortable. ey walked at a speed pro-
portional to leg length as determined by a convention known as a Froude number that follows the principle of
dynamic similarity39. A Froude number of 0.15 was chosen for each subject because it is a comfortable, moderate
walking pace. Leg length was measured as the distance from a subject’s greater trochanter to the ground. Kinetic
and kinematic data were collected simultaneously across a 30s data collection period for each walking trial
(barefoot, 10°, 20°, 30° and 40° sandal). Ten steps were exported for further analysis.
Acquisition of kinematic and kinetic data. Motion data were captured at 200Hz using an eight-camera
3D optoelectronic motion capture system (Oqus, Qualysis, Gothenburg, Sweden). GRF data were synchronously
captured with the motion data at 2000Hz using the Qualisys Track Management soware (Qualisys, Gothen-
burg, Sweden). In order to quantify three-dimensional motions of the foot and shank, een retro-reective
markers (12.0mm diameter) were placed on each subject’s right knee, ankle, and foot. ese were placed on
bony landmarks dened by Leardini etal.40, which dene the leg and foot as ve separate segments: the shank
(lower leg between the knees and ankle) the hallux (toes), metatarsals (forefoot), midfoot, and calcaneus (rear-
foot).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

5
Vol.:(0123456789)
Scientific RepoRtS | (2020) 10:14643 | https://doi.org/10.1038/s41598-020-71247-9
www.nature.com/scientificreports/
Data analysis. Marker trajectories and GRF data were exported to Visual3D (C-motion Inc., Germantown,
MD, USA) for post-processing and analysis. A recursive fourth-order Butterworth low-pass lter (10Hz cuto
frequency) was used to process both kinematic and GRF data. e same cuto frequency was used for both
GRF and kinematic data to avoid artifacts in inverse dynamics calculations that occur when dierent lter cut-
o frequencies are used41. Contact time (from heel contact to toe-o) was calculated with a 50N vertical GRF
threshold. All data curves were time normalized to the stance phase duration for plotting and visual inspection.
Joint kinematics. e geometrical denitions of MLA and MTP joint angle were based on skin-markers
located on the calcaneus (Cal), sustentaculum tali (ST), base of the rst metatarsal bone (FMB), head of the
rst metatarsal bone (FMH), and distal end of the rst proximal phalanx (PM)40,42. e MLA and MTP joint
measurements were calculated as angles between the projections of two vectors on the sagittal plane of the foot
overall and the forefoot segment, respectively, as dened in Leardini etal.40. For the MLA angle, the vector on
the proximal segment is bounded by marker Cal_proj and ST, where Cal_proj is the projection of Cal on the x–y
plane of the foot. e vector on the distal segment is bounded by markers ST and FMH42. For the MTP joint, the
vector on the proximal segment is bounded by marker FMB and FMH, and the vector on the distal segment is
bounded by markers FMH and PM. To evaluate the eect of toe springs on arch kinematics, MLA and MTP joint
angles were analyzed at 50% of stance phase, although peak forces acting on the arch occur later in stance during
walking. is allowed us to isolate the toe spring eect on MLA deformation, since this is prior to the initiation
of normal MTP dorsiexion due to heel li during walking. To evaluate the eect of toe springs on MTP joint
kinematics during thepropulsive phase, peak MTP joint dorsiexion angle and the corresponding MLA angle,
as well as peak MTP joint angular velocity, were quantied. Further, total range of MTP joint dorsiexion angle
was dened as therange through which the toes rotate from the moment when the COP aligns with the MTP
joint to peak dorsiexion angle.
Joint kinetics. Quasi-stiness of the midtarsal joint during midstance (dened as the slope of the joint’s
moment–angle relationship) was computed using the MLA joint angles and corresponding joint moment43.
e quasi-stiness was computed only when the COP was anterior to the midtarsal joint center (dened by the
marker on the sustentaculum tali) until the heel le the ground. MTP joint moment and power were dened to
be zero until the resultant GRF vector moved anterior to the MTP joint center19,35. Power was calculated using
the following equation:
P=M×ω
, where M is the moment and ω is the angular velocity at the MTP joint,
derived from the kinematic data (Fig.3B). Negative and positive work were then quantied by taking the integral
of power over time of push-o (i.e., from the moment when the resultant GRF vector originates anterior to the
MTP joint center to the moment the toes lied o the ground). Distance of travel of COP aer it moved anterior
to the MTP joint was also quantied. All calculations were performed using Visual3D, and custom MATLAB
(e MathWorks, Natick, MA, USA) and R (R Core Team 2019, Vienna, Austria) scripts.
Statistical analysis. Means and standard deviations (mean ± SDs) were calculated and a Shapiro–Wilk
test of normality was performed for all variables. Further comparison of kinematic and kinetic measurements
between the barefoot and the four sandal conditions was performed using a one-way repeated measures ANOVA
for normally distributed outcome parameters, or a Friedman test for those measurements that were not normally
distributed. When a signicant main eect between conditions was observed, Bonferroni-adjusted post-hoc
analysis was performed. For the non-normally distributed measurements, a Wilcoxon signed-rank test was per-
formed. For all tests signicance was set at α = 0.05, using IBM SPSS Statistics, version 25 (IBM, Armonk, New
York, USA).
Results
At midstance, the increasingly angled toe springs kept the toes at slightly but signicantly increasing degrees of
dorsiexion (p < 0.05, repeated measures ANOVA). Bonferroni-adjusted post-hoc analysis revealed signicant
dierences in MTP joint dorsiexion between the dierent toe spring conditions (Table1). However, the total
degree of toe dorsiexion at midstance was relatively small, reaching a maximum of 6.28° (± 2.35°) with the 40°
sandal (Fig.3). Variations in toe spring angle did not have a signicant eect on midtarsal joint quasi-stiness
(p > 0.05, Friedman ANOVA).
At push-o, a notable dierence of 10.55° for peak MTP joint angle was found between the barefoot and
10° sandal condition. Contrary to expectation, no dierences were found for peak MTP joint angle between all
sandal conditions (p > 0.05, repeated measures ANOVA) (Fig.3, Table1). However, total range of MTP joint
dorsiexion angle changed between the barefoot and all sandal conditions. Between barefoot and 10° sandal con-
dition, the total MTP joint range dropped signicantly by 29.42% (p < 0.05, Friedman ANOVA). With increasing
toe-spring angle, the total MTP joint range further decreased by up to 15.92% between the 10° (19.72° ± 7.71°)
and 40° sandals (16.58° ± 8.68°) (p < 0.05, Friedman ANOVA). e change in total MTP joint range corresponds
well with the MTP joint dorsiexion angle at the moment when the COP passed the MTP joint center. e MTP
joint angle at that moment increased with increasing toe spring angles, from 11.88° ± 8.17° in the 10° sandal to
15.82 ± 8.93° in the 40° sandal (p < 0.05, Friedman ANOVA). Along with the changes in total range of MTP joint
dorsiexion angle, signicant dierences were found in the time when the COP passed the MTP joint center
during stance phase (p < 0.05, Friedman ANOVA). e COP passed the MTP joint center signicantly earlier
in the 10° sandal (79.77 ± 6.19% stance phase) compared to barefoot (82.31 ± 7.11% stance phase) and the 40°
sandal (82.76 ± 7.22% stance phase).
While peak MTP joint velocity was signicantly higher in the barefoot condition, there were also no dier-
ences in peak angular velocity between the sandal condition (p > 0.05, Friedman ANOVA) (Fig.4A, Table1).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

6
Vol:.(1234567890)
Scientific RepoRtS | (2020) 10:14643 | https://doi.org/10.1038/s41598-020-71247-9
www.nature.com/scientificreports/
e most interesting nding to emerge from the data during propulsive phase is that peak moment increased
with the 10° sandals compared to the barefoot condition, but decreased gradually with increasing toe spring
angle by up to 31.63% between the 10° and 40° sandals (Fig.4B, Table1). is behavior in peak moment is likely
related to the fact that the distance of travel of COP distal to the MTP joint increased between barefoot and 10°
sandal condition, but decreased signicantly with increasing toe-spring angle by up to 22.26% between the 10°
(39.26 ± 18.31mm) and 40° sandals (30.55 ± 11.16mm)(p < 0.05, Friedman ANOVA) (Table1).
Negative MTP joint work was greatest in the barefoot and 10° sandal (−2.81 ± 2.08 and −2.76 ± 2.12J, respec-
tively) and least in the 40° sandal (−1.81 ± 1.65J). A comparison between the sandals showed a gradual decrease
by 2.5%, 15.3%, 24.2%, and 35.6% relative to the 10° sandal for the 20°, 30° and 40° sandals, respectively (p < 0.05,
Friedman ANOVA) (Table1). Bonferroni-adjusted post-hoc analysis revealed a signicant dierence between the
10° and 40° sandals. Positive MTP joint work was signicantly dierent between barefoot and 10° as well as 20°
sandals (p < 0.05, Friedman ANOVA). No signicant dierence was found between the sandal conditions, but the
data indicate a gradual, slight decrease from the 10° sandal (0.28 ± 0.33J) to the 40° sandal (0.11 ± 0.09J) (Table1).
Discussion
e present study was designed to model and then test the eects of toe springs in shoes on foot biomechanics
during walking. We hypothesized that toe springs would increase the stiness of the medial longitudinal arch
during midstance by engaging the windlass mechanism. We further hypothesized that the negative work at
Figure3. (A) Mean temporal proles of the metatarsophalangeal (MTP) joint dorsiexion angle during
normalized stance phase duration. Subjects walked barefoot (green) and in four curved sandal conditions
(purple: 10°, blue: 20°, yellow: 30°, red: 40°). Toe springs increased MTP joint dorsiexion at midstance
but decreased dorsiexion at the end of stance. (B) Mean temporal proles of the MTP joint power during
normalized stance phase duration.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

7
Vol.:(0123456789)
Scientific RepoRtS | (2020) 10:14643 | https://doi.org/10.1038/s41598-020-71247-9
www.nature.com/scientificreports/
the MTP joints would decrease during the propulsive phase because toe springs would reduce the total angle
through which the toes rotate. While we did not nd any change in arch stiness due to toe springs, our results
indicate that toe springs decrease the negative work at the MTP joints during push-o. As expected, greater toe
spring angles decreased the total range of MTP joint rotation. Linked to this, we found that more angled toe
springs delayed the time when the COP passed the MTP joint center and reduced the distance traveled by the
COP anterior to the MTP joint. As a result of these kinematic changes, the corresponding joint moment and
resulting negative work performed decreased. Comparison of results from the toe spring conditions to those
of the barefoot condition revealed that toe springs seem to be able to compensate for the negative eects of sti
shoes on MTP work requirements.
e windlass mechanism predicts that dorsiexion of the toes increases tension on the plantar aponeurosis,
which stiens the longitudinal arch as a whole12. However, the small increases in MTP joint dorsiexion angles
caused by toe springs in this study did not aect measured arch stiness. is result corresponds to the ndings
of Welte etal.14, who used static loading to compress the arches of sitting participants at dierent toe dorsiexion
angles, and found that engaging the windlass mechanism did not increase arch stiness. A possible explanation
might be the exibility of the plantar aponeurosis, which stretches as the arch is loaded44–46. us, our results
can be interpreted as further evidence that the windlass mechanism does not contribute substantially to stien-
ing the longitudinal arch at midstance. Another possible explanation for the lack of toe spring eect on arch
stiness at midstance in our study could have been the relatively small MTP joint dorsiexion angles achieved
in the toe spring conditions. Although we designed sandals with toe spring angles of up to 40°, and a bending
stiness that is similar to conventional shoes, the maximum MTP joint dorsiexion angle caused by toe springs
was less than 10°. Although we do not know why these angles were so low, it is possible that during midstance,
when the arch is compressed by force from above and the plantar aponeurosis tenses, the windlass is unwound.
is unwinding of the windlass, described as a ‘reverse windlass’12,15, could plantarex the toes at the MTP joint
and counteract the curvature of the toe spring. However, this mechanism needs further testing.
In contrast to what we document at midstance, our results indicate that during the propulsive phase toe
springs aect MTP joint dynamics, as evidenced by the signicant decrease in negative work associated with
increasing toe spring angles. Paradoxically, negative work was highest during walking barefoot and with 10°
sandals. However, these values were highest in these conditions for dierent reasons. When barefoot, participants
achieved high peak MTP dorsiexion angles, necessitating high angular velocity and hence high magnitudes of
negative work. In contrast, the 10° sandal showed a reduced MTP dorsiexion angle but caused the COP to move
more distally during toe-o, eectively increasing the acting moment and hence negative work. e distal shi
of the COP in sandals is likely a consequence of pushing o against a relatively sti platform in the sole, which
is also reected by an earlier passing of the COP relative to the MTP joint center. ese ndings are broadly
consistent with other studies linking dierences in MTP joint dynamics with shoe stiness47–50. While there were
no dierences in peak MTP dorsiexion angles across sandals with dierent degrees of toe spring, there were
signicant dierences in time when the COP passed the MTP joint center. With increasing toe spring angles,
the COP passed the MTP joint center later in stance phase. e delay in timing might explain the reduced total
MTP joint range and magnitude of distal COP motion. Along with these ndings, our results show a signicant
decline in negative work coincident with the reductions in MTP joint range and COP travel due to increasing
toe spring angles. In addition, the data suggest a gradual downturn of the peak angular velocity with increasing
toe spring angles, but this dierence is not signicant between conditions. us, toe springs seem to counteract
the negative eects of sti shoes on MTP work requirements. While sti shoes do stien the MTP joints, toe
springs might compensate for the eects of increased COP travel distal to the MTP joints, and further reduce
total MTP joint range and possibly peak angular velocity, thereby reducing negative work.
Table 1. Mean values and standard deviations of medial longitudinal arch (MLA) and metatarsophalangeal
(MTP) joint kinematics and kinetics during stance phase for walking barefoot and in sandals with varying toe
spring angles. bf Value signicantly dierent from barefoot. 10 Value signicantly dierent from 10° toe spring
angle. 20 Value signicantly dierent from 20° toe spring angle. 30 Value signicantly dierent from 30° toe
spring angle. All values are signicant at the p < 0.05 level.
Barefoot 10° sandal 20° sandal 30° sandal 40° sandal Main eect
p-Value
MTP joint dorsiexion at midstance (°) 0.49 (0.72) 1.23 (1.74) 3.01 (1.91) bf,10 5.41 (2.28) bf,10,20 6.28 (2.35) bf,10,20,30 > 0.001
Time of stance phase when COP aligns with MTP joint (°) 82.31 (7.11) 79.77 (6.19)bf 80.54 (7.31) 81.38 (6.70)10 82.76 (7.22)10 0.021
MTP joint dorsiexion at moment when COP aligns with MTP joint (°) 14.42 (9.87) 11.88 (8.17)bf 13.09 (7.96) 10 14.60 (9.09) 10,20 15.82 (8.93) 10,20,30 0.006
Peak MTP joint dorsiexion (°) 41.43 (5.14) 30.88 (5.82) bf 31.54 (5.33) bf 31.89 (5.93) bf 31.83 (5.93) bf > 0.001
Total MTP joint range (°) 27.94 (10.23) 19.72 (7.71) bf 18.93 (7.58) bf 17.94 (7.93) bf,10 16.58 (8.68) bf,10,20 > 0.001
Peak MTP joint velocity (rad/s) 5.44 (1.95) 3.36 (1.40)bf 3.44 (1.33)bf 3.37 (1.01)bf 2.91 (1.98)bf > 0.001
Peak MTP joint moment (Nm) 7.16 (3.67) 10.15 (4.54)bf 8.88 (4.47) 8.56 (4.11) 7.54 (3.92) 10 0.003
Distance of travel of COP distal to MTP joint (mm) 30.72 (8.16) 39.26 (18.31)bf 38.26 (17.3) 34.20 (10.89) 30.55 (11.16)10,20,30 0.022
Negative work (J) −2.81 (2.08) −2.74 (2.12) −2.38 (2.26) −2.13 (1.79) −1.81 (1.65) bf,10 0.011
Positive Work (J) 0.06 (0.10) 0.28 (0.33)bf 0.22 (0.25)bf 0.16 (0.20) 0.11 (0.09) 0.016
Midtarsal joint quasi-stiness at midstance (Nm/°) 0.20 (0.04) 0.22 (0.07) 0.20 (0.05) 0.22 (0.06) 0.22 (0.05) 0.262
Content courtesy of Springer Nature, terms of use apply. Rights reserved

8
Vol:.(1234567890)
Scientific RepoRtS | (2020) 10:14643 | https://doi.org/10.1038/s41598-020-71247-9
www.nature.com/scientificreports/
e decrease in negative work at the MTP joints suggests that intrinsic foot muscles have to perform less
eccentric muscle work to control MTP joint dorsiexion during the propulsion phase of gait19. Farris etal.15,19
found that the intrinsic foot muscles play an important role in helping to stien the MTP joints as they are being
dorsiexed at the end of a step in walking and running. By reducing moments at the MTP joints, toe springs likely
relieve the intrinsic foot muscles of some of the work necessary to stien these joints. While the dierences in
joint work among conditions measured in this study are relatively small, the intrinsic foot muscles themselves
are also small, meaning that a higher proportion of the available fascicles will need to be contracted to produce
a given amount of energy than in larger limb muscles. Furthermore, these small dierences in muscle work
likely add up to substantial dierences over time when considering that the average individual in industrialized
countries takes 4,000 to 6,000 steps per day51. us, habitually wearing shoes with toe springs could inhibit or
de-condition the force generating capacity of intrinsic foot muscles. While the direct consequences of weak foot
muscles are not fully known, it is likely that they could increase susceptibility to at foot and associated problems
such as plantar fasciitis21,23. is painful condition, which is recognized as an injury caused by excessive and
repetitive loading of the foot’s longitudinal arch26, is the most commonly treated foot problem among habitually
shod populations52. Farris etal. 15 recently suggested that the intrinsic foot muscles contract to prevent strain in
the plantar aponeurosis under high loads, and thus weakening the intrinsic foot muscles may limit their ability
to perform this function.
It is crucial to bear in mind that the possible link between toe springs and plantar fasciitis needs further test-
ing. Unfortunately, this study included only habitual shoe users, whose feet might already have been adapted to
shoes with toe springs. Further studies are needed to investigate the eect of toe springs on habitually barefoot
Figure4. Toe springs did not aect the metatarsophalangeal (MTP) joint velocity but did change the acting
MTP joint moment during propulsive phase. (A) plots the MTP joint angle against the MTP joint moment. (B)
plots the MTP joint angleagainst the MTP joint velocity. Both plots display the mean values of all participants
during walking barefoot (green) and in four curved sandal conditions (purple: 10°, blue: 20°, yellow: 30°, red:
40°).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

9
Vol.:(0123456789)
Scientific RepoRtS | (2020) 10:14643 | https://doi.org/10.1038/s41598-020-71247-9
www.nature.com/scientificreports/
individuals. Additional limitations that need further testing are walking speed and gaits. While this study tested
only one walking speed, and did not look at running, future investigations should also test greater speeds that
increase the demands on arch stability and muscle activity. is study also did not assess intrinsic foot muscle
activity; therefore, uncertainty remainsregarding if changes in muscle activity reect the alterations in MTP
joint work. It is possible that the intrinsic foot muscles contract isometrically as the toes are being dorsiexed
during walking, and that changes in power at the MTP joints among toe spring conditions reect dierences
in elastic energy storage and release, rather than changes in intrinsic foot muscle work. Recent static loading
experiments from Kelly etal.53 have suggested that intrinsic foot muscle fascicles actually contract concentri-
cally at high loads, but further research is necessary to determine whether this holds true during walking and
running. Additional caution should be taken when interpreting the MTP joint kinetics. MTP moment, power
and work were calculated from the moment when the COP passed anterior to the MTP joint. is approach has
been used in previous studies (Rolian etal., 2009; Farris etal., 2019), but might slightly overestimate the moment
and power calculations at the toe joints when compared to more complex methods requiring independent force
measurements from multiple forceplates54. Nevertheless, we expect this eect would be consistent across condi-
tions used in the present study, and therefore should not aect our overall conclusions.
Notwithstanding these limitations, we conclude that toe springs have important heretofore unrecognized
biomechanical eects on foot function that merit consideration, especially since they have become increasingly
exaggerated in modern athletic shoes55. As shown here, toe springs can alter the natural biomechanics of the
foot during walking principally by altering total work at the MTP joint and thus potentially reducing the work
required by the intrinsic foot muscles. Consequently, while toe springs may increase comfort by reducing the
eort of the foot muscles, they may increase susceptibility to plantar fasciitis and other foot-related problems.
at said, considerably more work will need to be done to understand more fully the eects of toe springs on
foot function and overall gait. Future studies might explore the impact of toe springs in combination with other
shoe features, including insoles, shoe stiness, and cushioning. Also, future research should investigate how toe
springs could aect more general aspects of gait such as center of mass mechanics, which have previously been
shown to be aected by MTP joint stiness and shoe sole curvature2,3. Studies could also explore additional
walking speeds and running to provide a more comprehensive understanding of how toe springs aect gait that
might help improve footwear design and use. Finally, future research should incorporate techniques such as
EMG to explore how the mechanical eects of toe springs observed here relate to actual neuromuscular output
and control during gait.
Received: 14 April 2020; Accepted: 12 August 2020
References
1. Nigg, B. M. Biomechanics of Sport Shoes 1st edn. (University of Calgary, Calgary, 2010).
2. Honert, E. C., Bastas, G. & Zelik, K. E. Eect of toe joint stiness and toe shape on walking biomechanics. Bioinspir. Biomim. 13,
66007. https ://doi.org/10.1088/1748-3190/aadf4 6 (2018).
3. Adamczyk, P. G., Collins, S. H. & Kuo, A. D. e advantages of a rolling foot in human walking. J. Exp. Biol. 209, 3953–3963. https
://doi.org/10.1242/jeb.02455 (2006).
4. Scha, P. S. & Cavanagh, P. R. Shoes for the insensitive foot: the eect of a “rocker bottom” shoe modication on plantar pressure
distribution. Foot Ankle 11, 129–140. https ://doi.org/10.1177/10711 00790 01100 303 (1990).
5. Fernández, P. J. et al. Evolution and function of the hominin forefoot. Proc. Natl. Acad. Sci. USA 115, 8746–8751. https ://doi.
org/10.1073/pnas.18008 18115 (2018).
6. Fernández, P. J., Holowka, N. B., Demes, B. & Jungers, W. L. Form and function of the human and chimpanzee forefoot: implica-
tions for early hominin bipedalism. Sci. Rep. 6, 30532. https ://doi.org/10.1038/srep3 0532 (2016).
7. Harcourt-Smith, W. E. H. & Aiello, L. C. Fossils, feet and the evolution of human bipedal locomotion. J. Anat. 204, 403–416. https
://doi.org/10.1111/j.0021-8782.2004.00296 .x (2004).
8. Latimer, B. & Lovejoy, C. O. Metatarsophalangeal joints of Australopithecus afarensis. Am. J. Phys. Anthropol. 83, 13–23. https ://
doi.org/10.1002/ajpa.13308 30103 (1990).
9. Lewis, O. J. Functional morphology of the evolving hand and foot (Clarendon Press and Oxford University Press, Oxford and New
York, 1989).
10. Susman, R. L., Stern, J. T. & Jungers, W. L. Arboreality and bipedality in the Hadar hominids. Folia Primatol. 43, 113–156. https
://doi.org/10.1159/00015 6176 (1984).
11. Venkadesan, M. et al. Stiness of the human foot and evolution of the transverse arch. Nature 579, 97–100. https ://doi.org/10.1038/
s4158 6-020-2053-y (2020).
12. Hicks, J. H. e mechanics of the foot. II. e plantar aponeurosis and the arch. J. Anat. 88, 25–30 (1954).
13. Sichting, F., Holowka, N. B., Ebrecht, F. & Lieberman, D. E. Evolutionary anatomy of the plantar aponeurosis in primates, including
humans. J. Anat. https ://doi.org/10.1111/joa.13173 (2020).
14. Welte, L., Kelly, L. A., Lichtwark, G. A. & Rainbow, M. J. Inuence of the windlass mechanism on arch-spring mechanics during
dynamic foot arch deformation. J. R. Soc. Interface 1, 4. https ://doi.org/10.1098/rsif.2018.0270 (2018).
15. Farris, D. J., Birch, J. & Kelly, L. Foot stiening during the push-o phase of human walking is linked to active muscle contraction,
and not the windlass mechanism. J. R. Soc. Interface 17, 20200208. https ://doi.org/10.1098/rsif.2020.0208 (2020).
16. Bruening, D. A., Pohl, M. B., Takahashi, K. Z. & Barrios, J. A. Midtarsal locking, the windlass mechanism, and running strike
pattern: a kinematic and kinetic assessment. J. Biomech. 73, 185–191. https ://doi.org/10.1016/j.jbiom ech.2018.04.010 (2018).
17. Caravaggi, P., Sforza, C., Leardini, A., Portinaro, N. & Panou, A. Eect of plano-valgus foot posture on midfoot kinematics during
barefoot walking in an adolescent population. J. Foot Ankle Res. 11, 55. https ://doi.org/10.1186/s1304 7-018-0297-7 (2018).
18. Cigoja, S. et al. e eects of increased midsole bending stiness of sport shoes on muscle-tendon unit shortening and shortening
velocity: a randomised crossover trial in recreational male runners. Sports Med. 6, 9. https ://doi.org/10.1186/s4079 8-020-0241-9
(2020).
19. Farris, D. J., Kelly, L. A., Cresswell, A. G. & Lichtwark, G. A. e functional importance of human foot muscles for bipedal locomo-
tion. Proc. Natl. Acad. Sci. USA 116, 1645–1650. https ://doi.org/10.1073/pnas.18128 20116 (2019).
20. Kelly, L. A., Cresswell, A. G., Racinais, S., Whiteley, R. & Lichtwark, G. Int rinsic foot muscles have the capacity to control deforma-
tion of the longitudinal arch. J. R. Soc. Interface 11, 20131188. https ://doi.org/10.1098/rsif.2013.1188 (2014).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

10
Vol:.(1234567890)
Scientific RepoRtS | (2020) 10:14643 | https://doi.org/10.1038/s41598-020-71247-9
www.nature.com/scientificreports/
21. Holowka, N. B., Wallace, I. J. & Lieberman, D. E. Foot strength and stiness are related to footwear use in a comparison of mini-
mally- vs. conventionally-shod populations. Sci. Rep. 8, 3679. https ://doi.org/10.1038/s4159 8-018-21916 -7 (2018).
22. Mann, R. & Inman, V. T. Phasic Activity of Intrinsic Muscles of the Foot. J Bone Joint Surg Am 46, 469–481 (1964).
23. McKeon, P. O., Hertel, J., Bramble, D. & Davis, I. e foot core system: a new paradigm for understanding intrinsic foot muscle
function. Br. J. Sports Med. 49, 290. https ://doi.org/10.1136/bjspo rts-2013-09269 0 (2015).
24. Zipfel, B. & Berger, L. R. Shod versus unshod: e emergence of forefoot pathology in modern humans?. e Foot 17, 205–213.
https ://doi.org/10.1016/j.foot.2007.06.002 (2007).
25. Tong, K. B. & Furia, J. Economic burden of plantar fasciitis treatment in the United States. Am. J. Orthop. 39, 227–231 (2010).
26. Wearing, S. C., Smeathers, J. E., Urry, S. R., Hennig, E. M. & Hills, A. P. e pathomechanics of plantar fasciitis. Sports Med. 36,
585–611. https ://doi.org/10.2165/00007 256-20063 6070-00004 (2006).
27. Cheung, R. T. H., Sze, L. K. Y., Mok, N. W. & Ng, G. Y. F. Intrinsic foot muscle volume in experienced runners with and without
chronic plantar fasciitis. J. Sci. Med. Sport 19, 713–715. https ://doi.org/10.1016/j.jsams .2015.11.004 (2016).
28. Chen, T.L.-W., Sze, L. K. Y., Davis, I. S. & Cheung, R. T. H. Eects of training in minimalist shoes on the intrinsic and extrinsic
foot muscle volume. Clin. Biomech. (Bristol, Avon) 36, 8–13. https ://doi.org/10.1016/j.clinb iomec h.2016.05.010 (2016).
29. Ridge, S. T. et al. Walking in minimalist shoes is eective for strengthening foot muscles. Med. Sci. Sports Exercise 51, 104–113.
https ://doi.org/10.1249/MSS.00000 00000 00175 1 (2019).
30. Lieberman, D. E. What we can learn about running from barefoot running: an evolutionary medical perspective. Exercise Sport
Sci. Rev. 40, 63–72. https ://doi.org/10.1097/JES.0b013 e3182 4ab21 0 (2012).
31. C onnolly, T. J. et al. Getting beyond the point: textiles of the terminal pleistocene/early holocene in the Northwestern Great Basin.
Am. antiq. 81, 490–514. https ://doi.org/10.1017/S0002 73160 00039 66 (2016).
32. Trinkaus, E. Anatomical evidence for the antiquity of human footwear use. J. Archaeol. Sci. 32, 1515–1526. https ://doi.org/10.1016/j.
jas.2005.04.006 (2005).
33. Tenner, E. Our Own Devices. e Past and Future of Body Technology (Alfred A. Knopf, New York, 2003).
34. Stefanyshyn, D. J. & Nigg, B. M. Mechanical energy contribution of the metatarsophalangeal joint to running and sprinting. J.
Biomech. 30, 1081–1085. https ://doi.org/10.1016/S0021 -9290(97)00081 -X (1997).
35. Rolian, C., Lieberman, D. E., Hamill, J., Scott, J. W. & Werbel, W. Walking, running and the evolution of short toes in humans. J.
Exp. Biol. 212, 713–721. https ://doi.org/10.1242/jeb.01988 5 (2009).
36. Toon, D. T. Dissertation. Loughborough University, 2008.
37. Krumm, D., Schwanitz, S. & Odenwald, S. Development and reliability quantication of a novel test set-up for measuring footwear
bending stiness. Sports Eng. 16, 13–19. https ://doi.org/10.1007/s1228 3-012-0106-7 (2013).
38. Hoogkamer, W., Kipp, S. & Kram, R. e biomechanics of competitive male runners in three marathon racing shoes: a randomized
crossover study. Sports Med. 49, 133–143. https ://doi.org/10.1007/s4027 9-018-1024-z (2019).
39. Alexander, R. M. & Jayes, A. S. A dynamic similarity hypothesis for the gaits of quadrupedal mammals. J. Zool. 201, 135–152. https
://doi.org/10.1111/j.1469-7998.1983.tb042 66.x (1983).
40. Leardini, A. et al. Rear-foot, mid-foot and fore-foot motion during the stance phase of gait. Gait Posture 25, 453–462. https ://doi.
org/10.1016/j.gaitp ost.2006.05.017 (2007).
41. Kristianslund, E., Krosshaug, T. & van den Bogert, A. J. Eect of low pass ltering on joint moments from inverse dynamics:
implications for injury prevention. J. Biomech. 45, 666–671. https ://doi.org/10.1016/j.jbiom ech.2011.12.011 (2012).
42. Caravaggi, P. et al. Reliability of medial-longitudinal-arch measures for skin-markers based kinematic analysis. J. Biomech. 88,
180–185. https ://doi.org/10.1016/j.jbiom ech.2019.03.017 (2019).
43. Kern, A. M., Papachatzis, N., Patterson, J. M., Bruening, D. A. & Takahashi, K. Z. Ankle and midtarsal joint quasi-stiness during
walking with added mass. PeerJ 7, e7487. https ://doi.org/10.7717/peerj .7487 (2019).
44. Caravaggi, P., Pataky, T., Goulermas, J. Y., Savage, R. & Crompton, R. A dynamic model of the windlass mechanism of the foot:
evidence for early stance phase preloading of the plantar aponeurosis. J. Exp. Biol. 212, 2491–2499. https ://doi.org/10.1242/
jeb.02576 7 (2009).
45. Fessel, G. et al. Changes in length of the plantar aponeurosis during the stance phase of gait: an invivo dynamic uoroscopic study.
Ann. Anat. = Anatomischer Anzeiger 196, 471–478. https ://doi.org/10.1016/j.aanat .2014.07.003 (2014).
46. Gefen, A. e invivo elastic properties of the plantar fascia during the contact phase of walking. Foot Ankle Int. 24, 238–244. https
://doi.org/10.1177/10711 00703 02400 307 (2003).
47. Willwacher, S., König, M., Potthast, W. & Brüggemann, G.-P. Does specic footwear facilitate energy storage and return at the
metatarsophalangeal joint in running?. J. Appl. Biomech. 29, 583–592. https ://doi.org/10.1123/jab.29.5.583 (2013).
48. Stefanyshyn, D. J. & Nigg, B. M. Inuence of midsole bending stiness on joint energy and jump height performance. Med. Sci.
Sports Exerc. 32, 471–476 (2000).
49. Oleson, M., Adler, D. & Goldsmith, P. A comparison of forefoot stiness in running and running shoe bending stiness. J. Biomech.
38, 1886–1894. https ://doi.org/10.1016/j.jbiom ech.2004.08.014 (2005).
50. Oh, K. & Park, S. e bending stiness of shoes is benecial to running energetics if it does not disturb the natural MTP joint
exion. J. Biomech. 53, 127–135. https ://doi.org/10.1016/j.jbiom ech.2017.01.014 (2017).
51. Altho, T. et al. Large-scale physical activity data reveal worldwide activity inequality. Nature 547, 336–339. https ://doi.o rg/10.1038/
natur e2301 8 (2017).
52. Cole, C., Seto, C. & Gazewood, J. Plantar fasciitis: evidence-based review of diagnosis and therapy. Am. Fam. Physician 72,
2237–2242 (2005).
53. Kelly, L. A., Farris, D. J., Cresswell, A. G. & Lichtwark, G. A. Intrinsic foot muscles contribute to elastic energy storage and return
in the human foot. J. Appl. Physiol. 126, 231–238. https ://doi.org/10.1152/jappl physi ol.00736 .2018 (2019).
54. Bruening, D. A. & Takahashi, K. Z. Partitioning ground reaction forces for multi-segment foot joint kinetics. Gait Posture 62,
111–116. https ://doi.org/10.1016/j.gaitp ost.2018.03.001 (2018).
55. Cavanagh, P. R. e Running Shoe book. [e Complete Story of the Running Shoe; Its History and Evolution, How It Is Manufactured
and Tested, How It Can Help Prevent Injuries, and a Ten-Point Plan to Help You Choose the Shoe at’s Right for You] (Anderson
World, Mountain View Calif, 1980).
Acknowledgements
We thank Annika Haenel, Henry Korb, and Henrik Zopf for their help with manufacturing the custom-made
sandals. A special thank goes to omas L. Milani for his support. Funding for this project was provided by the
American School of Prehistoric Research (Harvard University).
Author contributions
D.E.L., F.S., N.B.H. and O.H. designed the study. N.B.H. and O.H. collected the data, and F.S., N.B.H. and O.H.
processed the data. F.S. and N.B.H. analyzed the data and prepared all gures. F.S., N.B.H., D.E.L. and O.H.
wrote the manuscript.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

11
Vol.:(0123456789)
Scientific RepoRtS | (2020) 10:14643 | https://doi.org/10.1038/s41598-020-71247-9
www.nature.com/scientificreports/
Funding
Open Access funding provided by Projekt DEAL.
Competing interests
e authors declare no competing interests.
Additional information
Correspondence and requests for materials should be addressed to F.S.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access is article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this license, visit http://creat iveco mmons .org/licen ses/by/4.0/.
© e Author(s) 2020
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com