Stroboscopic Vision as a Dynamic Sensory Reweighting Alternative to the Sensory Organization Test

Abstract

Context:
The sensory organization test (SOT) is a standard for quantifying sensory dependence via sway-referenced conditions (sway-referenced support and sway-referenced vision [SRV]). However, the SOT is limited to expensive equipment. Thus, a practical version of the SOT is more commonly employed-the clinical test for sensory integration in balance; however, it fails to induce postural instability to the level of SRV.

Objective:
Determine if Stroboscopic vision (SV), characterized by intermittent visual blocking, may provide an alternative to the SRV for assessing postural stability.

Design:
Descriptive laboratory study.

Setting:
Research laboratory.

Participants:
Eighteen participants (9 males, 9 females; age = 22.1 [2.1] y, height = 169.8 [8.5] cm, weight = 66.5 [10.6] kg).

Intervention:
Participants completed the SOT conditions, and then repeated SOT conditions 2 and 5 with SV created by specialized eyewear.

Main outcome measures:
A repeated-measures analysis of variance was completed on the time-to-boundary metrics of center-of-pressure excursion in the anteroposterior and mediolateral directions in order to determine the difference between the full-vision, SV, and SRV conditions.

Results:
Postural stability with either SRV or SV was significantly worse than with full vision (P < .05), with no significant difference between SV and SRV (P > .05). Limits of agreement analysis revealed similar effects of SV and SRV except for unstable surface mediolateral time-to-boundary.

Conclusions:
In general, SV was found to induce a degree of postural instability similar to that induced by SRV, indicating that SV could be a portable and relatively inexpensive alternative for the assessment of sensory dependence and reweighting.

Full text

Stroboscopic Vision as a Dynamic Sensory Reweighting Alternative
to the Sensory Organization Test
Kyung-Min Kim, Joo-Sung Kim, Jeonghoon Oh, and Dustin R. Grooms
Context:The sensory organization test (SOT) is a standard for quantifying sensory dependence via sway-referenced
conditions (sway-referenced support and sway-referenced vision [SRV]). However, the SOT is limited to expensive
equipment. Thus, a practical version of the SOT is more commonly employed—the clinical test for sensory integration in
balance; however, it fails to induce postural instability to the level of SRV. Objective:Determine if Stroboscopic vision
(SV), characterized by intermittent visual blocking, may provide an alternative to the SRV for assessing postural stability.
Design:Descriptive laboratory study. Setting:Research laboratory. Participants:Eighteen participants (9 males, 9 females;
age = 22.1 [2.1] y, height = 169.8 [8.5] cm, weight = 66.5 [10.6] kg). Intervention:Participants completed the SOT
conditions, and then repeated SOT conditions 2 and 5 with SV created by specialized eyewear. Main Outcome Measures:
A repeated-measures analysis of variance was completed on the time-to-boundary metrics of center-of-pressure excursion in
the anteroposterior and mediolateral directions in order to determine the difference between the full-vision, SV, and SRV
conditions. Results:Postural stability with either SRV or SV was significantly worse than with full vision (P<.05), with no
significant difference between SV and SRV (P>.05). Limits of agreement analysis revealed similar effects of SV and SRV
except for unstable surface mediolateral time-to-boundary. Conclusions:In general, SV was found to induce a degree of
postural instability similar to that induced by SRV, indicating that SV could be a portable and relatively inexpensive
alternative for the assessment of sensory dependence and reweighting.
Keywords:balance/posture, biomechanics, instrument-assisted interventions, sensory integration
Sensory dependence is the preferential contribution of visual,
vestibular, and somatosensory feedback for completing a motor
task or maintaining stability and can be modified by pathologies,
task conditions, or training.
1,2
The sensory organization test (SOT)
is the gold standard for quantifying sensory dependence by utiliz-
ing sway-referenced conditions. Specifically, the SOT engages a
sway-referenced support condition to knockdown the somatosen-
sory system and a sway-referenced vision (SRV) condition to
knockdown the visual system. The SOT provides a standardized
assessment of sensory contributions to postural stability and has
been used extensively to evaluate concussion and fall risk and
patients with balance disorders in order to inform targeted therapies
for specific sensory deficits.
1,3,4
However, the SOT is limited to expensive equipment; thus, the
clinical test for sensory integration in balance (CTSIB) was
developed.
5
In the CTSIB, the sway-referenced condition of the
SOT is imitated using a foam pad, and SRV is imitated using a
white dome over the head.
5
The foam-induced postural instability
is comparable to the sway-referenced support condition, but the
dome is unable to induce instability to the level of the SRV.
6–8
Thus, there remains a need to develop an adequate surrogate for
SRV conditions for the CTSIB.
Stroboscopic vision (SV) may provide an alternative to SRV
and is characterized by intermittent visual blocking via glasses
or goggles. SV has recently been shown to modify visual depen-
dence and postural instability in both stable and unstable surface
conditions.
9
However, a comparison to the gold standard sensory
dependence for vision SRV condition has yet to be examined. If SV
induces postural instability similar to that induced by SRV, then it
may offer a viable option for a visually perturbed sensory depen-
dence assessment. Thus, the purpose of this study was to compare
the postural instability effects of SV against SRV on both stable
and unstable surfaces. We hypothesized that both SV and SRV
would significantly increase postural instability relative to full
vision, but postural instability between SV and SRV would not
be different and have high agreement.
Methods
Eighteen young adults (9 males, 9 females; age = 22.1 [2.1] y,
height = 169.8 [8.5] cm, weight = 66.5 [10.6] kg) without any
history of musculoskeletal injury in the past 6 months or of vision
or balance disorders participated. The University of Miami
Institutional Review Board approved the study protocol requiring
informed consent.
All participants completed the SOT with the NeuroCom
EquiTest (Neurocom International Inc, Clackamas, OR), consistent
with the testing protocol described previously.
4
However, SOT
conditions 2 and 5, in which participants are asked to close their
eyes during testing, were replaced with SV created by specialized
eyewear (Nike SPARQ Vapor Strobes; Nike Inc, Beaverton, OR)
K.-M. Kim, J.-S. Kim, and Oh are with the Department of Kinesiology and Sport
Sciences, University of Miami, Miami, FL, USA. Grooms is with the Division of
Athletic Training, School of Applied Health Sciences and Wellness, College of
Health Sciences and Professions; and the Ohio Musculoskeletal & Neurological
Institute, Ohio University, Athens, OH, USA. Grooms (groomsd@ohio.edu)is
corresponding author.
1
Journal of Sport Rehabilitation, (Ahead of Print)
https://doi.org/10.1123/jsr.2019-0466
© 2020 Human Kinetics, Inc. TECHNICAL REPORT

that intermittently cycles between opaque and transparent for 100
milliseconds at a time.
9
Participants performed bipedal stances that
lasted for 20 seconds and were repeated 3 times. The averages were
used for statistical analysis. The study outcomes were time-to-
boundary (TTB) parameters of center-of-pressure excursion. We
computed TTB measures such as mean and SD of the minima in the
anteroposterior (AP) and mediolateral (ML) directions using pre-
viously reported methods.
10
A lower TTB value reflects poorer
postural stability.
For each of the TTB measures, a repeated-measures analysis
of variance was performed to determine if there was a difference
in postural stability on the fixed (stable) surface between 3 visual
conditions as follows: (1) full vision (SOT condition 1), (2) SV
(SOT condition 2), and (3) SRV (SOT condition 3). The same
analysis was repeated for these conditions on the sway-
referenced (unstable) surface (SOT conditions 4–6). In addition,
Cohen deffect sizes and associated 95% confidence intervals
(CIs) were calculated to provide the magnitude of postural
instability with either SV or SRV compared to full vision using
pooled SD. Cohen guidelines were used to determine the
strength of the effect size as follows: (1) less than 0.2 = trivial,
(2) from 0.21 to 0.5 = small, (3) from 0.51 to 0.8 = moderate, and
(4) greater than 0.8 = large.
11
To assess agreement between SRV and SV, we completed a
one-sample ttest to determine difference from zero for each
balance measure, Bland–Altman plots to assess agreement for
each measure, and Pearson rvalues with 95% CIs between
conditions for all measures. We also calculated R
2
values to
determine the variance of SRV performance explained by SV.
All statistical analyses were conducted using SPSS statistical
software (version 24.0; IBM Corp, Armonk, NY) with the alpha
level set a priori at P≤.05.
Results
Descriptive data for the 3 visual conditions with the fixed and
sway-referenced surfaces are presented in Tables 1and 2,respec-
tively. For postural stability on a fixed surface, there were
significant differences between the 3 visual conditions for the
AP mean and SD measures (Table 1). Postural stability with either
SRV or SV in the AP direction was significantly worse than with
full vision, but there were no significant differences between SRV
and SV. The disruptive effects of SRV were small (close to
moderate), and SV effects were moderate, but there was no
difference in effect size between SRV and SV relative to FV,
because their respective 95% CI overlapped. In contrast to AP
postural instability, ML stability was unaffected by SRV or SV on
the stable surface.
Thereweresignificant differences for all TTB measures for
postural stability on the unstable (sway-referenced) surface
(Table 2). Postural stability with SRV or SV was significantly
worse relative to full vision, but SV again did not significantly
differ from SRV. Postural instability with SV or SRV in the
AP direction was of large magnitude relative to FV. For ML
postural instability, SV also had large effects relative to FV,
while SRV had small to moderate effects. Again, relative to FV,
the 95% CIs of both SV and SRV overlapped, indicating no
effect differences.
The limits of agreement analysis also largely supported a
similar effect of SV relative to SRV. On the fixed surface, all
TTB metrics had good agreement (Figure 1) and moderate to
high correlation (Table 1,r=.49–.77). On the unstable surface,
the TTB mean and SD in the AP direction had good agreement
(Figure 2), with moderate to high correlation (Table 2,
r=.65–.75). However, TTB mean and SD in the ML direction
had condition differences greater than zero but still had
high correlations (Table 2,r=.79–.82). The R
2
values for
each variable between SRV and SV are reported in Tables 1
and 2.
Discussion
We found both SRV and SV significantly impaired standing
balance relative to full vision, with no significant difference
between SV and SRV. Specifically, we observed small to mod-
erate effects for AP postural instability on a stable surface, with
no ML effect and moderate to large effects for AP and ML
postural instability on an unstable surface with both visual
conditions. The limits of agreement analysis indicated that SV
and SRV induced comparable effects in all conditions for all
outcome metrics except for the ML direction during the unstable
surface condition. This lack of agreement may have been due to
the different mechanisms of visual knockdown. The SV condi-
tion reduces visual feedback by 50% (strobe rate of 100-ms
transparent\100-ms opaque) and is not specifictoanAPorML
corrective strategy, whereas the SRV produces inaccurate visual
feedback primarily in the AP direction. Nonetheless, SV ac-
counted for approximately 62% to 67% of the variance in SRV
for ML TTB mean and SD; the remaining 33% to 38% can
potentially be attributed to differences in the ways in which
visual information was manipulated. The different mechanisms
of visual perturbation, while resulting in generally similar pos-
tural instability, should be considered for SOT in patients with
postural instability depending on an origin of postural instability
(reduced sensory feedback versus inability to recognize accurate
sensory input).
Stroboscopic vision can be utilized dynamically and is
portable and relatively inexpensive ($200 or less), making it
an attractive alternative for sensory reweighting assessment.
The need for such a tool is especially apparent in settings dealing
with highly active patients, such as in sports medicine, with
emerging evidence indicating that common musculoskeletal con-
ditions induce sensory reweighting.
12–15
SOT may not be effec-
tive in athletic populations due to its reliance on a static bipedal
stance not being challenging enough to discover postural insta-
bility that may be manifested during more aggressive athletic
maneuvers. Thus, SV could be an alternative for progressive
sensory reweighting in more active populations. Of note, our
findings are limited to individuals without current injury and to
the NeuroCom system. Patients, especially those with disorders
affecting balance, may respond to SV differently; this requires
future research.
Conclusions
Stroboscopic vision was found to induce a degree of postural
instability similar to that induced by SRV and can be used in the
assessment of sensory reweighting as part of CTSIB, allowing future
investigators and clinicians to engage in more dynamic postural
stability assessments.
(Ahead of Print)
2K.-M. Kim et al

Table 1 TTB Measures of Center-of-Pressure Excursion During 20-second Bipedal Stance on Fixed (Stable) Surface
TTB
parameters RM ANOVA FV SRV SV
Effect size
a
One-sample
ttest
b
Pearson r
c
R
2
FV and SRV FV and SV SRV and SV
ML mean (s) F
2,34
= 1.52;
P= .23
64.82
(15.15)
62.27
(12.01)
59.45
(12.62)
−0.19
(−0.84 to 0.47)
−0.38
(−1.04 to 0.28)
0.22
(−0.43 to 0.88)
t
17
= 1.24;
P= .23
.71 (.36, .88);
P<.01**
.50
(.13, .77)
AP mean (s) F
2,34
= 3.73;
P= .03*
35.25
(12.60)
30.01
(9.87)
26.86
(13.56)
−0.46
(−1.12 to 0.20)
−0.64
(−1.31 to 0.03)
−0.27
(−0.93 to 0.38)
t
17
= 1.56;
P= .14
.77 (.47, .91);
P<.01**
.59
(.22, .83)
ML SD (s) F
2,34
= 1.47;
P= .24
40.14
(9.14)
38.19
(7.46)
36.08
(10.82)
−0.23
(−0.89 to 0.42)
−0.41
(−1.07 to 0.25)
0.29
(−0.37 to 0.95)
t
17
= .92;
P= .37
.49 (.30, .78);
P= .04**
.24
(.09, .61)
AP SD (s) F
2,34
= 3.63;
P= .04*
26.60
(10.68)
22.49
(7.34)
20.04
(9.38)
−0.45
(−1.11 to 0.21)
−0.65
(−1.32 to 0.02)
0.23
(−0.43 to 0.88)
t
17
= 1.69;
P= .11
.75 (.44, .90);
P<.01**
.56
(.19, .81)
Abbreviations: AP, anteroposterior; FV, full vision; ML, mediolateral; RM ANOVA, repeated-measures analysis of variance; SRV, sway-referenced vision; SV, stroboscopic vision; TTB, time-to-boundary. Note: Bolded R
2
values
were calculated from the Pearson correlation coefficients.
a
Cohen’sdestimate of effect size was calculated between 2 of 3 visual conditions using pooled standard deviation, along with its associated 95% confidence interval. A lower TTB measure indicates poorer balance; a negative value
of effect size represents a decline in balance performance.
b
One-sample ttests were performed for each measure to determine the difference between SRV and SV from zero.
c
Pearson correlation coefficients (r) were estimated
between SRV and SV conditions for each measure, along with their 95% confidence intervals.
*AP postural control with SRV or SV was significantly worse than postural control with full vision; SV did not significantly differ from SRV. **Significant correlation between SRV and SV conditions.
(Ahead of Print) 3

Table 2 TTB Measures of Center-of-Pressure Excursion During 20-Second Bipedal Stance on Sway-Referenced (Unstable) Surface
TTB
parameters
RM
ANOVA FV SRV SV
Effect size
a
One-sample
ttest
b
Pearson r
c
R
2
FV and SRV FV and SV SRV and SV
ML mean (s) F
2,34
= 7.77;
P<.01*
50.69
(10.18)
44.14
(10.95)
34.99
(17.65)
−0.62
(−1.29 to 0.05)
−1.09
(−1.79 to −0.39)
−0.62
(−1.29 to 0.05)
t
17
= 3.44;
P<.01
.79 (.51, .92);
P<.01**
.62
(.26, .85)
AP mean (s) F
2,34
= 16.56;
P<.01*
17.64
(4.86)
12.17
(3.94)
12.35
(4.43)
−1.24
(−1.97 to −0.50)
−1.14
(−1.86 to −0.41)
0.04
(−0.61 to 0.70)
t
17
=−.16;
P= .88
.75 (.44, .90);
P<.01**
.56
(.19, .81)
ML SD (s) F
2,34
= 5.77;
P<.01*
30.76
(5.92)
28.11
(7.03)
21.98
(10.91)
−0.41
(−1.07 to 0.25)
−1.00
(−1.69 to −0.31)
−0.67
(−1.34 to 0.01)
t
17
= 3.95;
P<.01
.82 (.57, .93);
P<.01**
.67
(.32, .86)
AP SD (s) F
2,34
= 7.34;
P<.01*
15.19
(4.62)
11.14
(3.12)
9.58
(5.70)
−1.03
(−1.72 to −0.33)
−1.08
(−1.78 to −0.38)
−0.34
(−1.00 to 0.32)
t
17
= 1.51;
P= .15
.65 (.26, .86);
P<.01**
.42
(.07, .74)
Abbreviations: AP, anteroposterior; FV, full vision; ML, mediolateral; RM ANOVA, repeated-measures analysis of variance; ML, mediolateral; SRV, sway-referenced vision; SV, stroboscopic vision; TTB, time-to-boundary. Note:
Bolded R
2
values were calculated from the Pearson correlation coefficients.
a
Cohen’sdestimate of effect size was calculated between 2 of 3 visual conditions using pooled standard deviation, along with its associated 95% confidence interval. A lower TTB measure indicates poorer balance; a negative value of
effect size represents a decline in balance performance.
b
One-sample ttests were performed for each measure to determine the difference between SRV and SV from zero.
c
Pearson correlation coefficients (r) were estimated between
SRV and SV conditions for each measure, along with their 95% confidence intervals.
*AP postural control with SRV or SV was significantly worse than postural control with full vision, while SV did not significantly differ from SRV. **Significant correlation between SRV and SV conditions.
4(Ahead of Print)

Figure 1 —Bland–Altman plots for time-to-boundary (TTB) measures of center-of-pressure excursion during 20-second bipedal stance on fixed (stable) surface; the dotted line indicates
zero, the solid line represents the average difference (bias line), and the dashed lines show the 95% confidence intervals of the average differences. AP indicates anteroposterior; ML, mediolateral;
SRV, sway-referenced vision; SV, stroboscopic vision; TTB, time-to-boundary.
(Ahead of Print) 5

Figure 2 —Bland–Altman plots for time-to-boundary (TTB) measures of center-of-pressure excursion during 20-second bipedal stance on sway-referenced (unstable) surface; the dotted line
indicates zero, the solid line represents the average difference (bias line), and the dashed lines show the 95% confidence intervals of the average differences. AP indicate anteroposterior; ML,
mediolateral; SRV, sway-referenced vision; SV, stroboscopic vision; TTB, time-to-boundary.
6(Ahead of Print)

Acknowledgments
This research did not receive any specific grant from funding agencies in
the public, commercial, or not-for-profit sectors. The authors have no
conflicts of interest associated with this work.
References
1. Peterka RJ. Sensory integration for human balance control. Handb
Clin Neurol. 2018;159:27–42. PubMed ID: 30482320 doi:10.1016/
B978-0-444-63916-5.00002-1
2. Forbes PA, Chen A, Blouin JS. Sensorimotor control of standing
balance. Handb Clin Neurol. 2018;159:61–83. PubMed ID:
30482333 doi:10.1016/B978-0-444-63916-5.00004-5
3. Pierchala K, Lachowska M, Wysocki J, Morawski K, Niemczyk K.
Evaluation of the Sensory Organization Test to differentiate non-
fallers from single- and multi-fallers. Adv Clin Exp Med. 2019;28:
35–43. PubMed ID: 30058785 doi:10.17219/acem/76374
4. Vander Vegt CB, Register-Mihalik JK, Ford CB, Rodrigo CJ,
Guskiewicz KM, Mihalik JP. Baseline concussion clinical measures
are related to sensory organization and balance. Med Sci Sports Exerc.
2019;51(2):264–270. PubMed ID: 30239494 doi:10.1249/MSS.
0000000000001789
5. Shumway-Cook A, Horak FB. Assessing the influence of sensory
interaction of balance. Suggestion from the field. Phys Ther.1986;
66(10):1548–1550. PubMed ID: 3763708 doi:10.1093/ptj/66.10.1548
6. Cohen H, Blatchly CA, Gombash LL. A study of the clinical test
of sensory interaction and balance. Phys Ther. 1993;73(6):346–351;
discussion 51–54. PubMed ID: 8497509 doi:10.1093/ptj/73.6.346
7. Hong SK, Park JH, Kwon SY, Kim JS, Koo JW. Clinical efficacy
of the Romberg test using a foam pad to identify balance problems:
a comparative study with the sensory organization test. Eur Arch
Otorhinolaryngol. 2015;272(10):2741–2747. PubMed ID: 25200890
doi:10.1007/s00405-014-3273-2
8. Wrisley DM, Whitney SL. The effect of foot position on the modified
clinical test of sensory interaction and balance. Arch Phys Med
Rehabil. 2004;85(2):335–338. PubMed ID: 14966723 doi:10.1016/
j.apmr.2003.03.005
9. Kim KM, Kim JS, Grooms DR. Stroboscopic vision to induce sensory
reweighting during postural control. J Sport Rehabil. 2017;26(5):1–
11. PubMed ID: 28605310 doi:10.1123/jsr.2017-0035
10. Hertel J, Olmsted-Kramer LC, Challis JH. Time-to-boundary mea-
sures of postural control during single leg quiet standing. J Appl
Biomech. 2006;22(1):67–73. PubMed ID: 16760569 doi:10.1123/
jab.22.1.67
11. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd
ed. Hillsdale, MI: Lawrence Erlbaum Association; 1988.
12. Song K, Burcal CJ, Hertel J, Wikstrom EA. Increased visual use in
chronic ankle instability: a meta-analysis. Med Sci Sports Exerc.
2016;48(10):2046–2056. PubMed ID: 27635773 doi:10.1249/MSS.
0000000000000992
13. Wikstrom EA, Song K, Pietrosimone BG, Blackburn JT, Padua DA.
Visual utilization during postural control in anterior cruciate
ligament- deficient and -reconstructed patients: systematic reviews
and meta-analyses. Arch Phys Med Rehabil. 2017;98(10):2052–2065.
PubMed ID: 28483655 doi:10.1016/j.apmr.2017.04.010
14. Claeys K, Brumagne S, Dankaerts W, Kiers H, Janssens L. Decreased
variability in postural control strategies in young people with non-
specific low back pain is associated with altered proprioceptive
reweighting. Eur J Appl Physiol. 2011;111(1):115–123. PubMed
ID: 20824281 doi:10.1007/s00421-010-1637-x
15. Kim KM, Kim JS, Oh J, Lee SY. Time-to-boundary analysis of postural
control following acute lateral ankle sprain. Gait Posture. 2019;67:151–
153. PubMed ID: 30340127 doi:10.1016/j.gaitpost.2018.10.002
(Ahead of Print)
Strobes and SOT 7