Complementary Therapies in Clinical Practice 31 (2018) 256e261

Contents lists available at ScienceDirect

Complementary Therapies in Clinical Practice

journal homepage: www.elsevier.com/locate/ctcp

Trunk and hip muscle activation during yoga poses: Do

sex-differences exist?
Lori A. Bolgla*, Leonard Amodio, Kristen Archer, Jillian Estes, Rachelle Leung,
Katherine Magoni, Alyssa Mullikin, Dylan Roberts, Mekenzie Weems, Debra Beazley
Department of Physical Therapy, Augusta University, 987 St Sebastian Way, Augusta, GA 30912, USA

a r t i c l e i n f o a b s t r a c t

Article history: Objective: To compare core activation during yoga between males and females.
Received 19 January 2018 Methods: Surface electromyography was used to quantify rectus abdominis (RA), abdominal obliques
Received in revised form (AO), lumbar extensors (LE), and gluteus maximus (GMX) activation during four yoga poses. Data were
13 March 2018
expressed as 100% of a maximum voluntary isometric contraction. Mixed-model 2
2 analyses of
Accepted 14 March 2018
variance with repeated measures were used to determine between-sex differences in muscle activity.
Results: Females generated greater RA activity than males during the High Plank (P < 0.0001) and
Keywords:
Dominant-Side Warrior 1 (P ¼ 0.017). They generated greater AO (P < 0.0001) and GMX (P ¼ 0.004) ac-
Complementary medicine
Surface electromyography
tivity during the High Plank (P < 0.0001). No between-sex EMG activity differences existed for the Chair
Gender and Upward Facing Dog.
Conclusion: Findings have provided preliminary evidence for between-sex differences in muscle acti-
vation during yoga poses. Clinicians should consider such differences when prescribing yoga to improve
muscle strength and endurance.
© 2018 Published by Elsevier Ltd.

1. Introduction
Yoga, originating in India over 3000 years ago, offers a holistic
approach to health and wellness through physical postures (asana),
breathing techniques (pranayama), and meditation (dyana) [1]. The
National Institutes of Health has classified yoga as a form of Com-
plementary and Alternative Medicine (CAM) mind-body therapy.
The National Center for Complementary and Alternative Medicine
defines CAM as a group of diverse medical and healthcare systems
and practices not presently considered part of conventional
healthcare [2]. As such, yoga exists as both a stand-alone inter-
vention for the exercise enthusiast and more importantly as a
complementary therapy for the chronic medically-impacted indi-
vidual. Yoga is used in combination with conventional western-
based medicine and sometimes in place-of-or-alternative-to con-
ventional western-based interventions [3].
* Corresponding author.
E-mail addresses: lbolgla@augusta.edu (L.A. Bolgla), lamodio@augusta.edu
(L. Amodio), krarcher@augusta.edu (K. Archer), jestes@augusta.edu (J. Estes),
rleung@augusta.edu (R. Leung), kmagoni@augusta.edu (K. Magoni), amullikin@
augusta.edu (A. Mullikin), dyroberts@augusta.edu (D. Roberts), mweems@
augusta.edu (M. Weems), dbeazley@augusta.edu (D. Beazley).
Yoga has been shown to benefit individuals with low back pain
(LBP) because it incorporates core strengthening and stabilization
exercises [4]. Exercises thought to positively address LBP are those
that require greater electromyographic (EMG) activity. Researchers
have quantified the relative core muscle activity during core
strengthening and stabilization exercises [5e10]; however, more
limited data exist for core activation during yoga poses [11,12].
Beazley et al. [12] recently reported the relative activation of core
muscles during the following poses: High Plank (Plank), Chair,
Upward-Facing Dog (Dog), and Dominant-Side Warrior 1 (Warrior).
They concluded that many of the poses could improve core strength
and endurance. More important, activation levels during many of
the poses were similar to those reported during commonly pre-
scribed exercises for the treatment of LBP.
A limitation of many studies has been the use of mixed-sex
cohorts of subjects. Findings from more recent studies have
shown that males and females exhibit different levels of muscle
activity during rehabilitation exercises [13,14], suggesting the need
for prescribing sex-specific exercises. To our knowledge, re-
searchers have not investigated potential sex-differences in muscle
activity during yoga poses. The purpose of this study was to
determine if males and females generate different levels of core
muscle activity during yoga poses. Due to the exploratory nature of

https://doi.org/10.1016/j.ctcp.2018.03.006
1744-3881/© 2018 Published by Elsevier Ltd.
 L.A. Bolgla et al. / Complementary Therapies in Clinical Practice 31 (2018) 256e261 257

this study, we used the null hypothesis that males and females
would generate similar levels of activation.

2. Methods

This study came from a secondary analysis of previously re-

ported cross-sectional data [12]. The primary purpose of this
investigation was to quantify the relative amount of EMG activity of
core muscles during yoga in untrained individuals. The secondary
purpose was to compare the relative amount of core muscle acti-
vation during yoga to traditional exercises prescribed for in-
dividuals with LBP.

2.1. Subjects

Fifteen males (mean age ¼ 25.4 ± 2.6 y; height 179.5 ± 7.2 cm;
and mass 81.2 ± 9.0 kg) and 15 females (mean age ¼ 24.0 ± 1.3 y;
height 168.7 ± 6.1 cm; and mass 62.1 ± 8.7 kg) participated. Sub-
jects between the age of 18 and 40 years with less than four weeks
of yoga experience were recruited from a local university. All were
healthy with no history of spine or upper/lower extremity surgery
or any significant spine or lower extremity injury in the past two
years. The investigators explained the benefits and risks of this
study to all participants, who then signed an informed consent
document approved by the Institutional Review Board.

2.2. Procedures

We followed procedures previously described [12]. Briefly,

subjects completed gentle stretching exercises for the trunk and
lower extremity. Then, each was instructed in the following yoga
poses (Figs. 1e4): Chair, Plank, Dog, and Warrior [11]. We used
these poses because they were similar to those clinically prescribed
for individuals with LBP [15]. During the Chair, subjects stood and
flexed the knees 45 (as if to sit in a chair) while keeping their backs

Fig. 2. The Chair pose.

straight, upper extremities overhead, and palms facing inward.

Subjects assumed a full push-up position for the Plank. For the Dog,
subjects were positioned in prone and then pushed upward using
the upper extremities to extend the spine. This exercise was chosen
because of its similarity to the McKenzie extension exercise. For the
Warrior, subjects lunged toward the same side as the dominant
hand with the upper extremities outstretched to the side, keeping
the trunk and non-dominant-side lower extremity facing forward.
Stance width was the length of the lower limb on the dominant
hand side. Subjects lunged to the position where the tibia on the
dominant hand side was vertical to the floor. This exercise was
chosen to facilitate core stabilization during a frontal plane
movement.
Next, EMG electrodes were placed over the rectus abdominis
(RA), abdominal obliques (AO), lumbar extensors (LE), and gluteus
maximus (GMX). The skin over these muscle bellies was cleaned
(and shaved, if needed) with isopropyl alcohol. Trigno™ wireless
sensors (Delsys®, Boston, MA) were placed parallel over each belly
[16,17]. Electrode placement was confirmed by observing electrical
signals on an oscilloscope as an investigator applied muscle resis-
tance in accordance with common manual muscle testing tech-
niques [18]. To enable normalization of the raw EMG data, subjects
Fig. 1. The High Plank pose.
258 L.A. Bolgla et al. / Complementary Therapies in Clinical Practice 31 (2018) 256e261
Fig. 3. The Upward-Facing Dog pose.
Fig. 4. Dominant-Side Warrior 1 pose.
performed three maximum voluntary isometric contractions
(MVIC) for each muscle in accordance with the “make” test [19].
Subjects completed one practice trial and three test trials [20]. Test
positions for each muscle were used in accordance with previously
described methods [21]. Subjects received strong verbal encour-
agement during each trial and rested 30 s between trials.
For testing, subjects held each pose for a 20-s period and EMG
activity during the last 15 s was used for analysis [11]. They also
rested at least 1 min between poses to minimize fatigue. The order
of testing was counterbalanced to reduce the possibility of order
bias [20]. Subjects received verbal feedback, if needed, to maintain
the proper form. After testing, subjects were instructed to refrain
from any physical activity, other than normal walking, for a 24-h
period to minimize the potential for muscle soreness.
2.3. EMG instrumentation and analysis
A four-channel wireless EMG system collected all EMG data,
which were sampled at 2000 Hz and band-pass filtered between 20
and 450 Hz. Unit specifications also included a common mode
rejection ratio greater than 80 dB. All data were root-mean-squared
(RMS) over a 30-ms moving window [21]. For the MVICs, a com-
puter algorithm determined the maximum RMS amplitude over a
moving 500-ms average window across the MVICs [22]. The win-
dow with the highest amplitude was used to normalize all data as a
percentage of MVIC (% MVIC). The average amplitude of EMG data
during the last 15 s of individual yoga poses was expressed as %
MVIC and used for statistical analysis.
2.4. Statistical analysis
Separate mixed-model 2 (sex) X 4 (yoga pose) analyses of
variance with repeated measures were used to determine between-
sex differences in average muscle amplitudes across poses. All
statistical analyses were conducted using IBM SPSS version 24.0
with a 0.05 level of significance. The level of significance was
adjusted using the sequentially rejective Bonferroni test to account
for multiple pairwise comparisons, thus protecting against possible
type I error [23].
3. Results
EMG activity for the RA was low to moderate (9.3e28.2% MVIC)
across poses. An interaction and main effect existed (P < 0.0001)
and the post-hoc analysis showed that subjects generated signifi-
cantly greater RA activity during the Plank (P < 0.0001) compared
to the other exercises. Females generated 2.2 times (38.7% vs 17.7%
MVIC) greater (P < 0.0001) RA activity than males during the Plank
and 1.9 times (12.2% vs 6.5% MVIC) greater (P ¼ 0.017) RA activity
during the Warrior (Fig. 5). EMG activity for the AO was moderate
to high (17.5e44.5% MVIC) across poses. An interaction and main
effect also existed (P < 0.0001) and the post-hoc analysis showed
Fig. 5. Mean electromyographic activity of the rectus abdominis during the yoga poses
expressed as 100% of a maximum voluntary isometric contraction (% MVIC).
 L.A. Bolgla et al. / Complementary Therapies in Clinical Practice 31 (2018) 256e261 259

that subjects generated significantly greater AO activity during the

Plank (P < 0.0001) compared to the Chair, Dog, and Warrior. They
also generated greater AO activity during the Dog compared to the
Chair (P < 0.0001) and Warrior (P < 0.0001). Females generated 1.8
times (57.6 vs. 31.4% MVIC) greater (P < 0.0001) AO activity than
males during the Plank (Fig. 6). EMG activity for the LE was low to
moderate (10.8e38.6% MVIC). Only a main effect existed
(P < 0.0001), with all subjects generating significantly greater LE
activity during the Chair (P < 0.0001) compared to the Plank, Dog,
and Warrior (Fig. 7). EMG activity for the GMX was low (14.3e20.4%
MVIC); no main effect existed (P ¼ 0.09). However, an interaction
effect existed in that females generated 1.9 times (18.7% vs 10.0%
MVIC) greater (P ¼ 0.004) activity than males during the Plank
(Fig. 8).

4. Discussion

The purpose of this study was to determine if males and females

generate different levels of core muscle activity during yoga poses. Fig. 7. Mean electromyographic activity of the lumbar extensors during the yoga poses
expressed as 100% of a maximum voluntary isometric contraction (% MVIC).
To interpret the relative amount of EMG activity, we will use the
following recommendations: 0e20% MVIC (low); 21e40% MVIC
(moderate); 41e60% MVIC (high); and greater than 60% MVIC (very
high) [24]. Moderate levels of EMG activity have been thought to
improve muscle endurance while high and very high levels of EMG
activity may produce greater strength gains. Findings from this
study have provided information of differences in expected gains
(e.g., muscle endurance or strength) based on sex. These data will
provide clinicians evidence for the development and imple-
mentation of sex-specific interventions.

4.1. Relative muscle activation

4.1.1. Rectus abdominis

All subjects, on average, generated a moderate level of activation
(28.2% MCIV) during the Plank, which was significantly greater than
the other poses (9.3e12.7% MVIC). This finding suggested that all
subjects, regardless of sex, would benefit from this exercise to
improve RA endurance. However, when comparing activation be-
tween males and females, females required significantly higher
activation (38.7% MVIC) just below the threshold for achieving
Fig. 8. Mean electromyographic activity of the gluteus maximus during the yoga poses
strengthening gains. Conversely, male activation (17.7% MVIC) was expressed as 100% of a maximum voluntary isometric contraction (% MVIC).
considered low. Therefore, the Plank may represent an exercise

designed best for improving RA strength in females. Conversely,

clinicians should consider prescribing more demanding exercises
for males in need of improved RA endurance and strength.

4.1.2. Abdominal obliques

Fig. 6. Mean electromyographic activity of the abdominal obliques during the yoga
poses expressed as 100% of a maximum voluntary isometric contraction (% MVIC).
Although overall activation of the AO during the Plank (44.5%
MVIC) was greater than the RA (28.2% MVIC), a similar pattern of
relative activation between males and females occurred. Females
generated high-to-very high activation (57.6% MVIC), which sug-
gested that the Plank would result in strengthening gains. Males, on
the other hand, exhibited moderate activation (31.4% MVIC) that
would improve AO endurance. This finding was clinically important
for the following reasons. First, the Plank was more demanding for
females, which would be appropriate for those needing improved
AO strength. However, the Plank may not necessarily be indicated
for females with significant AO weakness. Second, overall AO
activation, regardless of sex, suggested sufficient activation for
strength gains. However, males would only benefit from the Plank
for endurance effects; more demanding AO exercises would be
required for strengthening effects.
It was noteworthy that subjects, regardless of sex, generated
260 L.A. Bolgla et al. / Complementary Therapies in Clinical Practice 31 (2018) 256e261
greater AO during the Dog compared to the Chair and Warrior
poses. This pattern highlighted sufficient AO activation for endur-
ance effects that would likely benefit males and females needing a
less demanding AO exercise. AO activation was 1.8 times greater
than RA activity in males and 1.5 times greater in females. This
pattern suggested that the AO may provide greater core stability
than the RA during positions incorporating trunk extension [12].
4.1.3. Lumbar extensors
All subjects, on average, generated moderate levels of LE activity
(38.6% MVIC) during the Chair and low levels during the other
poses (9.8e16.4% MVIC). Greater LE activation reflected its impor-
tance for controlling trunk flexion during the Chair [12]. Therefore,
males and females would benefit equally from the Chair to improve
LE endurance. The clinician would need to prescribe more
demanding exercises for individuals needing greater LE strength.
4.1.4. Gluteus maximus
On average, GMX activity was low (14.3e20.4% MVIC) across all
poses. This finding suggested that none of the poses would benefit
individuals in need to GMX endurance and strength [12].
Comparing the relative activation between males and females
highlighted an interesting pattern of differences. Males generated
GMX activity sufficient for endurance effects during the Dog (23%
MVIC) and Warrior (22.2% MVIC). Therefore, the Dog and Warrior
would be more beneficial for males with GMX weakness compared
to females. Unlike males, females almost generated sufficient GMX
activity during the Chair (18.6% MVIC) and Plank (18.7% MVIC) for
endurance effects. When prescribing the Chair and Plank, females
may achieve moderate GMX activity with even greater instruction
to activate the GMX during these poses. Additional investigations
would be needed to make this determination. In summary, all
subjects, regardless of sex, would require more demanding exer-
cises for GMX strength gains.
4.2. Clinical implications
Findings from this study were unique because they highlighted
between-sex differences in muscle activation during some of the
poses. Enough RA activity for endurance effects was noted in fe-
males, but not males, during the Plank. Therefore, males needing
improved RA endurance effects would require an exercise, other
than the Plank, to target this muscle. More demanding RA-specific
exercises would be needed for strength gains in males and females.
Regarding the AO, the Plank will benefit males requiring improved
endurance and females requiring strengthening effects. Like the RA,
males who may benefit from AO strengthening effects would need
more demanding exercises. It was noteworthy that all subjects,
regardless of sex, generated AO activity during the Dog for endur-
ance effects. Therefore, females with AO weakness who cannot
perform the Plank may initially benefit from the Dog. As AO
endurance improves, they could transition to the more demanding
Plank. During the Dog, males would gain similar endurance effects
as the Plank, offering them a choice of exercise for improved AO
endurance. The Chair will be equally appropriate for males and
females in need of LE endurance. Finally, GMX activity was rela-
tively low during all poses. However, males generated sufficient
activity during the Dog and Warrior for low-level endurance effects.
Clinicians may consider prescribing these poses to initially target
the GMX. However, females would need an exercise, other than the
Dog and Warrior, for improved GMX function.
4.3. Limitations
As reported previously, there were certain limitations associated
with this study [12]. First, only healthy individuals participated so
that findings cannot be generalized to individuals with musculo-
skeletal or neurological pathologies. Second, signal crosstalk from
adjacent muscles could have occurred with the use of surface EMG
but was minimized since electrodes were donned in a standardized
manner [16,17].
5. Conclusions
Yoga has continued to emerge as an effective alternative or
complement to physical therapy for treating non-specific LBP
[25,26]. To our knowledge, prior works only have examined the
relative activation of core muscles in individuals regardless of sex
[11,27]. Findings from the current study have provided important
information of differences in core muscle activity not only across
poses but also sex. Clinicians may use these results for the devel-
opment and implementation of not only evidence-based, but also
sex-specific, yoga interventions.
Funding
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
Conflicts of interest
The authors have no competing interests to declare.
Appendix A. Supplementary data
Supplementary data related to this article can be found at
https://doi.org/10.1016/j.ctcp.2018.03.006.
References
[1] H. Cramer, L. Ward, A. Steel, R. Lauche, G. Dobos, Y. Zhang, Prevalence, pat-
terns, and predictors of yoga use: results of a U.S. nationally representative
survey, Am. J. Prev. Med. 50 (2016) 230e235, https://doi.org/10.1016/
j.amepre.2015.07.037.
[2] P.M. Barnes, B. Bloom, R.L. Nahin, Complementary and alternative medicine
use among adults and children: United States, 2007, Natl. Health Stat. Rep. 12
(2008) 1e23.
[3] C. Woodyard, Exploring the therapeutic effects of yoga and its ability to in-
crease quality of life, Int. J. Yoga 4 (2011) 49e54, https://doi.org/10.4103/
0973-6131.85485.
[4] N.J. Patil, R. Nagarathna, P. Tekur, D.N. Patil, H.R. Nagendra, P. Subramanya,
Designing, validation, and feasibility of integrated yoga therapy module for
chronic low back pain, Int. J. Yoga 8 (2015) 103e108, https://doi.org/10.4103/
0973-6131.158470.
[5] R.A. Ekstrom, R.A. Donatelli, K.C. Carp, Electromyographic analysis of core
trunk, hip, and thigh muscles during 9 rehabilitation exercises, J. Orthop.
Sports Phys. Ther. 37 (2007) 754e762, https://doi.org/10.2519/
jospt.2007.2471.
[6] D. Czaprowski, A. Afeltowicz, A. Gebicka, et al., Abdominal muscle EMG-
activity during bridge exercises on stable and unstable surfaces, Phys. Ther.
Sport 15 (2014) 162e168, https://doi.org/10.1016/j.ptsp.2013.09.003.
[7] J.W. Youdas, M.M. Boor, A.L. Darfler, M.K. Koenig, K.M. Mills, J.H. Hollman,
Surface electromyographic analysis of core trunk and hip muscles during
selected rehabilitation exercises in the side-bridge to neutral spine position,
Sport Health 6 (2014) 416e421, https://doi.org/10.1177/1941738114539266.
[8] R.F. Escamilla, C. Lewis, D. Bell, et al., Core muscle activation during Swiss ball
and traditional abdominal exercises, J. Orthop. Sports Phys. Ther. 40 (2010)
265e276, https://doi.org/10.2519/jospt.2010.3073.
[9] S. Maeo, T. Takahashi, Y. Takai, H. Kanehisa, Trunk muscle activities during
abdominal bracing: comparison among muscles and exercises, J. Sports Sci.
Med. 12 (2013) 467e474.
[10] S.J. Atkins, I. Bentley, D. Brooks, M.P. Burrows, H.T. Hurst, J.K. Sinclair, Elec-
tromyographic response of global abdominal stabilizers in response to stable-
and unstable-base isometric exercise, J. Strength Condit Res. 29 (2015)
1609e1615, https://doi.org/10.1519/jsc.0000000000000795.
[11] M. Ni, K. Mooney, K. Harriell, A. Balachandran, J. Signorile, Core muscle
function during specific yoga poses, Compl. Ther. Med. 22 (2014) 235e243,
https://doi.org/10.1016/j.ctim.2014.01.007.
[12] D. Beazley, S. Patel, B. Davis, S. Vinson, L. Bolgla, Trunk and hip muscle
 L.A. Bolgla et al. / Complementary Therapies in Clinical Practice 31 (2018) 256e261
activation during yoga poses: implications for physical therapy practice,
Compl. Ther. Clin. Pract. 29 (2017) 130e135, https://doi.org/10.1016/
j.ctcp.2017.09.009.
[13] L. Bolgla, N. Cook, K. Hogarth, J. Scott, C. West, Trunk and hip electromyo-
graphic activity during single leg squat exercises do sex differences exist? Int.
J. Sports Phys. Ther. 9 (2014) 756e764.
[14] M.K. Dwyer, S.N. Boudreau, C.G. Mattacola, T.L. Uhl, C. Lattermann, Compari-
son of lower extremity kinematics and hip muscle activation during reha-
bilitation tasks between sexes, J. Athl. Train. 45 (2010) 181e190, https://
doi.org/10.4085/1062-6050-45.2.181.
[15] B.J. Hoogenboom, K. Keisel, Core Stabilization Training. Clinical Orthopaedic
Rehabilitation, third ed., Elsevier, Philadelphia, 2011, pp. 467e481.
[16] E. Criswell, Cram's Introduction to Surface Electromyography, second ed.,
Jones and Bartlett Publishers, Sudbury, MA, 2011.
[17] A. Rainoldi, G. Melchiorri, I. Caruso, A method for positioning electrodes
during surface EMG recordings in lower limb muscles, J. Neurosci. Meth. 134
(2004) 37e43, https://doi.org/10.1016/j.jneumeth.2003.10.014.
[18] F.P. Kendall, E.K. McCreary, P. Provance, M.M. Rodgers, W.A. Romani, Muscles.
Testing and Function with Posture and Pain, fifth ed., Lippincott Williams &
Wilkins, Baltimore, MD, 2005.
[19] R.W. Bohannon, Reference values for extremity muscle strength obtained by
hand-held dynamometry from adults aged 20 to 79 years, Arch. Phys. Med.
Rehabil. 78 (1997) 26e32.
[20] L.A. Bolgla, T.L. Uhl, Reliability of electromyographic normalization methods
for evaluating the hip musculature, J. Electromyogr. Kinesiol. 17 (2007)
261
102e111.
[21] L.A. Bolgla, M.F. Cruz, L.H. Roberts, A.M. Buice, T.S. Pou, Relative electromyo-
graphic activity in trunk, hip, and knee muscles during unilateral weight
bearing exercises: implications for rehabilitation, Physiother. Theory Pract. 32
(2016) 130e138, https://doi.org/10.3109/09593985.2015.1092059.
[22] M.M. Bamman, S.G. Ingram, J.F. Caruso, M.C. Greenisen, Evaluation of surface
electromyography during maximal voluntary contraction, J. Strength Condit
Res. 11 (1997) 68e72.
[23] S. Holm, A simple sequentially rejective multiple test procedure, Scand. J. Stat.
6 (1979) 65e70.
[24] M.P. Reiman, L.A. Bolgla, J.K. Loudon, A literature review of studies evaluating
gluteus maximus and gluteus medius activation during rehabilitation exer-
cises, Physiother. Theory Pract. 28 (2012) 257e268, https://doi.org/10.3109/
09593985.2011.604981.
[25] R. Chou, R. Deyo, J. Friedly, et al., Nonpharmacologic therapies for low back
pain: a systematic review for an American College of Physicians clinical
practice guideline, Ann. Intern. Med. 166 (2017) 493e505, https://doi.org/
10.7326/m16-2459.
[26] A.P. Goode, R.R. Coeytaux, J. McDuffie, et al., An evidence map of yoga for low
back pain, Compl. Ther. Med. 25 (2016) 170e177, https://doi.org/10.1016/
j.ctim.2016.02.016.
[27] M. Ni, K. Mooney, A. Balachandran, L. Richards, K. Harriell, J.F. Signorile,
Muscle utilization patterns vary by skill levels of the practitioners across
specific yoga poses (asanas), Compl. Ther. Med. 22 (2014) 662e669, https://
doi.org/10.1016/j.ctim.2014.06.006.