Recommendation

Based on the options you selected, these are the recommendations for the experimental context:

Identification of muscle synergies (also called motor modules/ primitives) using methods such as non-negative matrix factorization

*Note: Consensus has not yet been reached regarding the physiological information contained within muscle synergies or their interpretation. Until further work is undertaken it is generally advisable to consider normalisation to MVC or peak across the task and interpret the results with caution (i.e., MVC normalisation is biased by muscles with large activation, and peak normalisation can give “meaning” to muscles that are barely used in a task).

Sourced from the paper Besomi et al. 2020

It is suggested to use the most stringent method below. Expand a method to learn more.

Yes: High probability that this is appropriate

No recommended methods are available for this category based on your selections.

Caution: Might be appropriate, considering specific issues

Normalisation to the maximum voluntary contraction (MVC) is appropriate for the extraction of muscle synergies but requires some considerations. Synergy extraction would be biased to describe the muscles that are activated at the higher level. As such, it may affect the number of extracted muscle synergies. Ideally, the MVC will be optimized for each muscle considered in the investigated task (i.e., separate MVCs for each muscle, but this may only be possible for isometric contractions).

See more

Normalisation to the maximum voluntary contraction (MVC) is appropriate for the extraction of muscle synergies but requires some considerations. Synergy extraction would be biased to describe the muscles that are activated at the higher level. As such, it may affect the number of extracted muscle synergies. Ideally, the MVC will be optimized for each muscle considered in the investigated task (i.e., separate MVCs for each muscle, but this may only be possible for isometric contractions). This method is less ideal than normalisation to the MVC in the same task/context as the task of interest because it DOES NOT control for physiological and anatomical issues (e.g., contraction type, muscle architecture, electrode placement, etc.).

See more

Although normalisation to the peak or mean EMG amplitude allows the extraction of muscle synergies, this method requires considerations. When evaluating the muscle synergies, because the amplitude of every muscle would be expressed as a percentage of their peak/mean value throughout the task of interest, the muscle synergies extraction will not be biased toward the muscle(s) with the highest activity. Normalising muscles that have sustained low-level activation or no activity across the task to their task peak/mean, is likely to give unwarranted weighting/ “meaning” to that muscle in the analysis.

See more

No: High probability that this is inappropriate

Normalisation to a standardised submaximal task is not appropriate for the extraction of muscle synergies. The submaximal task is unlikely to be equivalent between muscles (different muscles can be active at different MVC%) and between participants (the same muscle can be more active in one person than another). This could result in extracting synergies with different temporal profiles and different weights for each participant, preventing the finding of common patterns.

See more

Non-normalised EMG amplitude is not appropriate for the extraction of muscle synergies. The synergy extraction would be biased to describe the muscles which exhibit a high EMG amplitude for possible non-relevant reasons (e.g., subcutaneous fat, muscle architecture, electrode location). This would affect the number of extracted muscle synergies and complicate the interpretation of the results.

See more
Back

We value your feedback

Let us know how helpful you found the recommendations above and how we can improve:

If you found this tool helpful in planning or conducting your experiment, please consider citing the original publication:

Besomi, M., Hodges, P. W., Van Dieën, J., et al (2020). Consensus for experimental design in electromyography (CEDE) project: Amplitude normalization matrix. Journal of Electromyography and Kinesiology, 53, 102438.

https://doi.org/10.1016/j.jelekin.2020.102438