Psychometric Comparison of the NASA-TLX and PROS-TLX for Myoelectric Prosthesis Use

Background: Prosthesis users report high cognitive workload, or the mental effort needed to achieve task performance. The NASA Task Load Index (NASA-TLX) is the most common subjective workload measure. However, it was developed for general human–machine interfaces and may not capture prosthesis-specific demands. The recently proposed Prosthesis Task Load Index (PROS-TLX) includes additional domains relevant to prosthesis use, yet its psychometric properties have not been established. This study compared the convergent validity, test-retest reliability, and measurement error of the NASA-TLX and PROS-TLX across two groups.Methods: Ten individuals without limb difference performed a timed reach-and-grasp task using a myoelectric prosthetic emulator in both a physical and virtual environment. Additionally, ten individuals with upper limb difference performed an accuracy- and time-based activity of daily living using their customary myoelectric, multi-grip prosthesis. Both groups completed two sessions at least a week apart and rated workload using the NASA-TLX and the PROS-TLX in both sessions. Because the PROS-TLX has more total subscales, we modified the scoring of the PROS-TLX to align it with the NASA-TLX. For both surveys, total workload scores were a weighted average of the subscales, ranging from low (0) to high (100). We compared total workload scores between the surveys using Pearson's correlation and calculated the intraclass correlation coefficients, measurement error, and minimum detectable change for each survey. We also explored the role of session number, survey, and task environment on the survey subscales and the relationships between subscales and task performance outcomes.Results: For emulator users, the PROS-TLX and NASA-TLX total workload scores were moderately correlated in session 1 (r>0.63), suggesting good convergent validity. However, for prosthesis users, the surveys were weakly correlated in session 1 (r=0.13). The PROS-TLX demonstrated higher test–retest reliability point estimates (ICC=0.44–0.83) and lower measurement error (SEM=7.43-10.79) than NASA-TLX (ICC=0.18-0.45, SEM=10.52-17.29), though the confidence intervals were large. The prosthesis-specific subscale of uncertainty varied by task environment and correlated with performance outcomes.Conclusions: The PROS-TLX adds prosthesis-specific contributors to workload to the NASA-TLX with higher reliability and lower measurement error estimates, but wide confidence intervals limit conclusions. Larger studies in prosthesis users are needed.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-18
DOI
https://doi.org/10.21203/rs.3.rs-11392273/v1
Primary Topic
Muscle activation and electromyography studies
Type
preprint

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Psychometric Comparison of the NASA-TLX and PROS-TLX for Myoelectric Prosthesis Use

Deanna H. Gates, Mira E. Mutnick, Jessica Carlson, Sarah Isgro
1 links
Research Square
Muscle activation and electromyography studies
preprint

Psychometric Comparison of the NASA-TLX and PROS-TLX for Myoelectric Prosthesis Use

Deanna H. Gates, Mira E. Mutnick, Jessica Carlson, Sarah Isgro
preprint en

Abstract

Background: Prosthesis users report high cognitive workload, or the mental effort needed to achieve task performance. The NASA Task Load Index (NASA-TLX) is the most common subjective workload measure. However, it was developed for general human–machine interfaces and may not capture prosthesis-specific demands. The recently proposed Prosthesis Task Load Index (PROS-TLX) includes additional domains relevant to prosthesis use, yet its psychometric properties have not been established. This study compared the convergent validity, test-retest reliability, and measurement error of the NASA-TLX and PROS-TLX across two groups.Methods: Ten individuals without limb difference performed a timed reach-and-grasp task using a myoelectric prosthetic emulator in both a physical and virtual environment. Additionally, ten individuals with upper limb difference performed an accuracy- and time-based activity of daily living using their customary myoelectric, multi-grip prosthesis. Both groups completed two sessions at least a week apart and rated workload using the NASA-TLX and the PROS-TLX in both sessions. Because the PROS-TLX has more total subscales, we modified the scoring of the PROS-TLX to align it with the NASA-TLX. For both surveys, total workload scores were a weighted average of the subscales, ranging from low (0) to high (100). We compared total workload scores between the surveys using Pearson's correlation and calculated the intraclass correlation coefficients, measurement error, and minimum detectable change for each survey. We also explored the role of session number, survey, and task environment on the survey subscales and the relationships between subscales and task performance outcomes.Results: For emulator users, the PROS-TLX and NASA-TLX total workload scores were moderately correlated in session 1 (r>0.63), suggesting good convergent validity. However, for prosthesis users, the surveys were weakly correlated in session 1 (r=0.13). The PROS-TLX demonstrated higher test–retest reliability point estimates (ICC=0.44–0.83) and lower measurement error (SEM=7.43-10.79) than NASA-TLX (ICC=0.18-0.45, SEM=10.52-17.29), though the confidence intervals were large. The prosthesis-specific subscale of uncertainty varied by task environment and correlated with performance outcomes.Conclusions: The PROS-TLX adds prosthesis-specific contributors to workload to the NASA-TLX with higher reliability and lower measurement error estimates, but wide confidence intervals limit conclusions. Larger studies in prosthesis users are needed.

Research Square
University of Michigan (US)
U.S. Department of Defense, National Institutes of Health
Good health and well-being
Muscle activation and electromyography studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon