Authors: Gaurav Seth ¹, William Seiple ², Fernanda Garcia-Piña ², Smita Rao ¹˒*, Amy Hurst ¹˒*, John-Ross Rizzo ³˒*
¹ NYU Steinhardt School of Culture, Education, and Human Development | ² Lighthouse Guild, New York City | ³ NYU Langone Health
* = authors contributed equally as senior authors
Persons with blindness or low vision (pBLV) exhibit gait adaptations in laboratory studies often described as a cautious gait pattern. It remains unclear whether between-group gait differences observed between pBLV and sighted individuals in laboratory settings persist during real-world outdoor walking. This pilot study examines group differences in commonly reported spatiotemporal gait parameters between pBLV and healthy vision controls during indoor and outdoor walking using wearable motion capture.
Participants completed six trials of straight walking indoors (hallway) and outdoors (nearby New York City sidewalk) along an 8 m walkway at a self-selected comfortable pace. Lower-body kinematics were recorded using Xsens MVN Awinda (seven IMUs). Outcomes selected a priori were walking speed, step length, step width, and double-support time. Given small, imbalanced samples, analyses emphasized estimation, reporting group means, and Hedges’ g effect sizes (pBLV minus control).
Across both environments, pBLV showed shorter step length, slower walking speed, and greater double-support time than controls, consistent with cautious gait patterns in the literature. Between-group differences in walking speed and double-support time were more pronounced outdoors, whereas step width differences were evident indoors but not outdoors, suggesting some adaptations may be environment-dependent. These preliminary findings indicate that key between-group gait differences reported in laboratory studies persist during real-world outdoor walking, with select features accentuated in ecologically valid environments, underscoring the value of studying mobility outside controlled settings.
People with blindness or low vision (pBLV) adapt the way they walk. Research consistently shows they walk slower, take shorter steps, and spend more time with both feet on the ground. This pattern is known as "cautious gait." (Gazzellini et al., 2016; Hallemans et al., 2010; Kahaki et al., 2023; Nakamura, 1997)
Almost all of this research was done indoors, in controlled settings. It is unclear whether these findings reflect how people actually walk in the real world. (Schmuckler, 2001; Lappi, 2015)
Real-world walking is harder. Uneven surfaces, other pedestrians, and unpredictable environments may affect pBLV more than sighted individuals.
Wearable sensors now make outdoor gait research possible. This study uses body-worn motion capture to measure walking in a controlled hallway and on a New York City sidewalk. (Hafer et al., 2023; Tao et al., 2012; Warmerdam et al., 2020)
Study Design
Single-session study at Lighthouse Guild, New York City. Participants completed both indoor and outdoor walking in counterbalanced order.
10 pBLV and 3 age- and sex-matched sighted controls. 7 pBLV included in final analysis due to data loss.
Participants with conditions known to affect gait were excluded.
Walking Task and Instrumentation
Lower-body kinematics recorded using Xsens MVN Awinda.
6 trials of straight walking along an 8-meter walkway at a self-selected pace, indoors and outdoors.
Outcomes and Analysis
Four gait parameters measured: walking speed, step length, step width, and double-support time. (Kahaki et al., 2023)
Small and imbalanced sample, so analyses focused on estimation rather than formal hypothesis testing.
Group means and Hedges' g effect sizes reported, with pBLV compared against sighted controls. (Hedges, 1981; Rosenthal et al., 1994)
Figure 1: Distributions of spatiotemporal gait parameters during indoor and outdoor walking for pBLV (red) and control (blue) groups. Boxplots show step length, step width, walking speed, and double-support percentage for each environment. Boxes represent the interquartile range, center lines indicate medians, and whiskers denote the data range excluding outliers.
Step Length
pBLV took shorter steps than controls in both environments.
Difference was large and consistent indoors and outdoors (g = −1.3 both).
Step Width
pBLV had a wider stance indoors but similar width to controls outdoors.
Difference shifted from moderate (g = 0.8) indoors to negligible (g = −0.1) outdoors.
Walking Speed
pBLV walked slower than controls in both environments.
Difference was accentuated outdoors (g = −1.0 indoors, g = −1.6 outdoors).
Double Support %
pBLV spent more time with both feet on the ground in both environments (higher double support %).
Difference was large and conistent across settings (g = 1.0 indoors, g = 1.1 outdoors).
Table 1: Effect size estimates (Hedges’ g) for between-group differences in spatiotemporal gait parameters during indoor and outdoor walking. Positive numbers indicate metric was smaller in control participants.
Several core features of cautious gait (shorter step length, slower walking speed, and greater double-support %) persist outside controlled environments.
Environmental effects were most evident for walking speed and double-support percentage, where between-group differences between pBLV and controls were more pronounced during outdoor walking.
As a pilot study with a small and imbalanced sample, these findings should be interpreted as preliminary.
These findings support the value of studying gait outside the laboratory to better understand real-world mobility in pBLV.
Gazzellini S, Lispi ML, Castelli E, Trombetti A, Carniel S, Vasco G, et al. The impact of vision on the dynamic characteristics of the gait: strategies in children with blindness. Exp Brain Res. 2016 Sept;234(9):2619–27.
Nakamura T. Quantitative analysis of gait in the visually impaired. Disabil Rehabil. 1997 May;19(5):194–7.
Hallemans A, Ortibus E, Meire F, Aerts P. Low vision affects dynamic stability of gait. Gait Posture. 2010 Oct;32(4):547–51.
Kahaki ZR, Safarpour AR, Daneshmandi H. The spatiotemporal gait parameters among people with visual impairment: A literature review study. Oman Journal of Ophthalmology. 2023;16(3):427–33.
Schmuckler MA. What is ecological validity? A dimensional analysis. Infancy. 2001 Oct;2(4):419–36.
Lappi O. Eye tracking in the wild: The Good, the Bad and the Ugly. J Eye Mov Res. 2015 Oct;8(5).
Hafer JF, Vitali R, Gurchiek R, Curtze C, Shull P, Cain SM. Challenges and advances in the use of wearable sensors for lower extremity biomechanics. J Biomech. 2023 Aug;157(111714):111714.
Tao W, Liu T, Zheng R, Feng H. Gait analysis using wearable sensors. Sensors (Basel). 2012 Feb;12(2):2255–83.
Warmerdam E, Hausdorff JM, Atrsaei A, Zhou Y, Mirelman A, Aminian K, et al. Long-term unsupervised mobility assessment in movement disorders. Lancet Neurol. 2020 May;19(5):462–70.
Hedges LV. Distribution theory for Glass's estimator of effect size and related estimators. J Educ Stat. 1981 June;6(2):107–28.
Rosenthal R, Cooper H, Hedges L. Parametric measures of effect size. The handbook of research synthesis. 1994;621(2):231–44.
PhD Candidate, Rehabilitation Science
NYU Steinhardt
Lighthouse Guild (New York)
Lighthouse Guild (New York)
NYU Steinhardt
NYU Steinhardt
NYU Langone Health