'Falling heads': investigating reflexive responses to head-neck perturbations

Abstract

Background: Reflexive responses to head-neck perturbations affect the injury risk in many different situations ranging from sports-related impact to car accident scenarios. Although several experiments have been conducted to investigate these head-neck responses to various perturbations, it is still unclear why and how individuals react differently and what the implications of these different responses across subjects on the potential injuries might be. Therefore, we see a need for both experimental data and biophysically valid computational Human Body Models with bio-inspired muscle control strategies to understand individual reflex responses better.

Methods: To address this issue, we conducted perturbation experiments of the head-neck complex and used this data to examine control strategies in a simulation model. In the experiments, which we call 'falling heads' experiments, volunteers were placed in a supine and a prone position on a table with an additional trapdoor supporting the head. This trapdoor was suddenly released, leading to a free-fall movement of the head until reflexive responses of muscles stopped the downwards movement.

Results: We analysed the kinematic, neuronal and dynamic responses for all individuals and show their differences for separate age and sex groups. We show that these results can be used to validate two simple reflex controllers which are able to predict human biophysical movement and modulate the response necessary to represent a large variability of participants.

Conclusions: We present characteristic parameters such as joint stiffness, peak accelerations and latency times. Based on this data, we show that there is a large difference in the individual reflexive responses between participants. Furthermore, we show that the perturbation direction (supine vs. prone) significantly influences the measured kinematic quantities. Finally, 'falling heads' experiments data are provided open-source to be used as a benchmark test to compare different muscle control strategies and to validate existing active Human Body Models directly.

Keywords: 3D finite element modelling; Head–neck perturbations; Motor control; Muscle modelling; Musculoskeletal model; Reflex behaviour.

Fig. 1

Vertical displacement of ’falling heads’ experiment. The vertical displacement of the supine (a) and prone (b) position is shown for all participants and all trials (light blue solid lines). For the supine case, we additionally show the simulation trajectory (as dark blue line with asterisks)

Fig. 2

Effective stiffness. The effective stiffness of the supine position (a) and the prone position (b) is shown for all participants and all trials (black diamonds). For the supine case, we additionally show the simulation value (red triangle)

Fig. 3

Differences in vertical displacement for different ages and sexes. Experimental vertical displacement trajectories for both the supine (a, b) and the prone position (c, d) are shown. The differences of age and sex are highlighted with different colours

Fig. 4

Experimental latency times. The latency times for the SCM and trapezius muscles in both the supine (a, c) and prone position (b, d) are shown. The mean and standard deviation were calculated for different age and sex groups

Fig. 5

Experimental net moment. The net moment $M_{\text{net}}$ plotted over the angle is shown here for a representative participant (participant 4), in both experiments (supine and prone position). The different colours represent the three separate trials

Fig. 6

Dynamic quantities. Comparison of different dynamic quantities for both the experimental results (participant 17, all three trials, displayed in colour) and the simulation result (displayed in black)

Fig. 7

Results of controller variation. Simulation results showing the vertical and rotational displacement trajectories for both the reflex controller (a, c) and the lambda controller (b, d). In comparison to the simulation results (displayed in colour), the mean value of the experimental data is shown with a black solid line, the standard deviation of the experimental data is shown as a grey area and all experimental trajectories are shown as dashed grey lines

Fig. 8

Sketch of the volunteer placement. The volunteer were placed in supine (a) and prone (b) position. The participant’s head was supported by a trapdoor released at the start of the experiment. The three recorded markers are labelled as M1, M2 and M3 in the figure. Here, $\phi =0$ represents the starting position, where the head is at rest