Introduction

Accurate joint position sense (JPS) is important for the execution of intended movements. Sensory information providing us with our joint position sense is known to be attributed to proprioceptors, including muscle spindles. Aberrant neck proprioceptive inputs into the central nervous system (CNS) may occur as a result of pain1, injury to proprioceptors, and/or a mismatch between abnormal neck afferent input and information from the vestibular and visual systems2,3. A decrease in head-neck joint position sense (JPS) accuracy in people experiencing both neck pain4 and neck injury5 has been reported. However, head-neck positioning accuracy was evaluated by moving the head with respect to the stable trunk in those studies, so the vestibular system is likely engaged in shaping head-neck position sense. In a more recent study, Chen and Treleaven used a neck torsion technique where the examiner gently held the participant’s head while the participant rotated their torso6. These authors argued that the neck torsion technique was superior to conventional tests of head-neck joint position sense (JPS) because activation of vestibular afferents was avoided. Nevertheless, the specific contribution of neck muscle spindles to head-neck position sense is not yet conclusive because of likely activation of both vestibular afferents and local neck mechanoreceptors of adjacent connective tissue structures (joints/ligaments). To clarify the role of neck muscle spindles in head-neck position sense, a tool that modulates neck muscle spindles inputs independently from these afferent systems is essential.

Skeletal muscle, as all striated muscle exhibits thixotropic behaviour. Muscle conditioning is a tool that exploits this thixotropic behaviour7. Muscle conditioning has been used by researchers to non-invasively study limb proprioception and the role of muscle spindles of limb muscles7. Muscle spindles play a significant role in providing the central nervous system (CNS) with information concerning the length and movement parameters of skeletal muscles. This allows the CNS to interpret where body segments are orientated in relation to each other and in space. The aim of the muscle conditioning protocol, originally developed by Proske et al.8 is to experimentally alter the mechanical sensitivity of muscle spindles which in turn alters the muscle spindles’ afferent resting discharge. The consequence of muscle conditioning, therefore, is an experimentally manipulated level of spindle afferent activity artificially signalling muscle length and the position of the corresponding joint to the CNS. The key step in muscle conditioning is a muscle contraction which abolishes stable cross bridges which spontaneously form in a relaxed muscle. Once the muscle contraction ends, new stable cross bridges will once again form at whatever length the muscle is at the time of the muscle contraction and subsequent relaxation. If a muscle is initially shortened, then contracted for a few seconds before the muscle is relaxed at this shortened length for at least 3 s, the result is the formation of stable cross bridges between actin and myosin filaments at this new length. Passively increasing the length of the muscle to an intermediate length does not detach these stable cross bridges but gives rise to an increase in the mechanical sensitivity of the muscle spindle, and an increased level of resting discharge of muscle spindle afferents. In contrast, if the muscle is initially lengthened, then contracted and maintained at this long length, passively shortening the muscle induces mechanical insensitivity of the muscle spindle to subsequent stretch because passive shortening results in slack of intrafusal fibres because of the persisting stable cross bridges within the intrafusal sarcomeres and a decreased level of resting discharge of muscle spindle afferents7. In other words, muscle conditioning alters the level of resting discharge of the spindle afferent fibres independently to the actual or physical length of the muscle. In this way, this experimental tool is used to induce systematic differences in errors in limb JPS (i.e., mechanically sensitive muscle spindles associated with more accurate JPS compared to mechanically insensitive muscle spindles leading to less accurate JPS)9.

This study has investigated two hypotheses.

Hypothesis #1

When dorsal neck muscles (DNM) are conditioned to leave their muscle spindles mechanically sensitive, like in limb muscles, head-neck and head-trunk position sense error will be reduced compared to when neck muscles are conditioned to leave muscle spindles mechanically insensitive.

Hypothesis #2

When the head is moved relative to the trunk about the neck, head-neck position sense error after muscle conditioning to leave DNM spindles slack will result in an error significantly smaller than when the trunk is moved about the fixed head.

There are two objectives of this study.

Objectives:

  1. 1.

    To investigate whether experimentally induced alterations in the relative quantity of dorsal neck muscle proprioceptive input by muscle conditioning results in a change in head-neck proprioceptive acuity evaluated by moving the head with respect to the stable trunk.

  2. 2.

    To investigate the effect of dorsal neck muscle conditioning on neck-trunk proprioceptive acuity tested by moving the trunk with respect to the stable head, to clarify whether head movement was a contributing factor in the accuracy in perception of head-neck- to trunk joint position sense.

Methods

Experimental design

This study consisted of two experiments. Participants lay on a bench with the head-neck or neck-trunk able to move only in the sagittal plane. For the first experiment, participants lay supine, and the head-neck rotated about the cervico-thoracic junction. For the second experiment, participants were positioned side-lying, the head was fixed while the trunk section was rotated. All experiments were approved by the RMIT Human Research Ethics Committee and conducted in accordance with the 1964 Declaration of Helsinki. Written informed consent was obtained from all participants prior to the commencement of the experiments including written informed consent for publication of identifying images in an online open-access publication.

Participants

Twenty-five young, healthy participants, with no self-reported that they were in good health with no history of neck pain or injury in the previous 6 months, neurological disease, vestibular dysfunction or known motion sickness completed the study. Volunteers were also excluded if they were taking any medications that may have had an impact on nervous or skeletal muscle function. The dizziness handicap inventory was used for those participants who completed the side-lying experiments (i.e., second experiment) to exclude any person suffering dizziness10. The first experiment, conducted in the supine position, consisted of 2 parts. Seven participants (6 males and 1 female) completed the experiment which investigated the JPE magnitude associated with determining the position of a pre-presented target angle of 10 degrees head-neck flexion. Nine different participants completed the same experiment, but the presented target angle was 15 and 20 degrees. All participants completing experiments in the supine posture were aged between 18 and 45 years.

The second experiment was conducted in the side-lying posture and a further 9 different participants (4 females, 5 males) (age range between 18 and 45 years) completed this experiment where the pre-presented target angle was 20 degrees only.

Participants were recruited through advertisements distributed within the university.

Equipment

Experiment 1 – Supine

Participants lay supine on a table which consisted of a head piece moveable in the sagittal plane and a fixed trunk piece. Head-neck to trunk joint position sense was investigated by rotating the head piece of the table around the stationary trunk (see Fig. 1, panels A and B).

Experiment 2 – Side lying

Participants were positioned on their left side on a motorised table which consisted of 2 separate sections, a fixed head section which held the head firmly but comfortably in a neutral position, and a motorised, moveable trunk section. Head-to-trunk joint position sense was investigated by rotating the trunk section of the table around the stationary head (see Fig. 1, Panels C and D).

Fig. 1
figure 1

Panels A and B, participants were positioned supine; the head piece of the table rotates around the fixed torso section of the table. Panel A, head movement into extension and Panel B, head movement into flexion. Panels C and D, a motorised table where participants were positioned side-lying. Note the fixed head-piece section of the table (Panels C and D) with the motorised trunk piece of the table rotating about the head piece. In panel C a participant lays on their right side. While the head remains fixed, the trunk piece is rotated, extending the neck on the stationery head. In both supine and side-lying postures the fulcrum of movement between head-neck and neck-trunk was the cervico-thoracic junction.

Full size image

Head-neck, neck-trunk joint position sense protocol

The accuracy in perception of the location of a pre-presented head-neck (experiment 1), and neck-trunk (experiment 2) joint angle was assessed. All participants were blind folded throughout the experiment. The protocol consisted of 2 phases, the learning phase and the testing phase (see Fig. 2).

Fig. 2
figure 2

Summary of experiment protocol.

Full size image

Phase 1 – Learning phase

Prior to each presentation of the target angle, previous history of neck muscle activation and length change was abolished. For experiment 1, this was achieved by participants actively moving their neck from full extension to full flexion 5 times before resting their head in a neutral position8. Immediately, the target angle was presented. For participants who completed experiment 2 (side lying), dorsal neck muscles were contracted at a shorter than neutral length (neck-trunk aligned with the head) of dorsal neck muscles7. The neck-trunk was then passively moved to the neutral (head-trunk aligned) position and immediately the target angle was presented.

Participants’ head-neck or neck-trunk was moved passively to the target angle (10 and 20 degrees in the supine position; 20 degrees only in the side-lying position) presented in neck flexion. These head-neck and neck-trunk angles were chosen as they reflected approximately the midpoint of the full range of movement of the head-neck or neck-trunk. In addition, for experiment 1, we were interested to learn whether a change in the location of the target angle from 10 to 15 to 20 degrees resulted in a difference in accuracy of JPS after muscle conditioning. The target angle was presented in flexion where dorsal neck muscles were longer than the neutral (head-trunk aligned) position to ensure that dorsal neck muscle spindles would be under stretch and mechanically sensitive. This position was held for 5 s so participants could learn the location of the presented target head-neck angle before the head-neck or neck-trunk was returned to a neutral, head-trunk aligned position.

Each participant was presented with the target angle 3 times so that the participant had opportunity to memorise (learn) the location of the presented target angle.

Phase 2, testing Phase - Presentation of the target angle

After the learning phase, the dorsal neck muscles were conditioned after which, the head-neck or neck-trunk was passively moved towards the target angle at a constant speed. We conducted preliminary experiments in the side lying posture (not published) measuring when participants detected movement of the table after both forms of muscle conditioning. Hold long muscle conditioning (resulting in slack muscle spindles) did affect the speed of table movement required to detect movement. To account for this effect, table movement speed was adjusted so that it was 25% faster than the mean perception of movement detection after hold long muscle conditioning. Participants were asked to indicate, by pressing a trigger, when they perceived that their head-neck or neck-trunk had reached the pre-presented target angle. For the first experiment, 5 trials were completed for each target angle after the 2 forms of muscle conditioning. For the second experiment where participants lay in the side-lying position, 6 trials were completed.

For the first experiment where participants were positioned supine, head-neck movement and position was recorded using a static accelerometer and electrogoniometer unit (Scientific Concepts, Melbourne Australia) which was mounted onto the forehead of the participant using adhesive tape. The output of the device was streamed to the data acquisition system (PowerLab/8SP, ADIstruments, Castle Hill, NSW, Australia).

In the second experiment where the participants were positioned in the side-lying position, neck-trunk position was recorded using a potentiometer goniometer that was mounted on the proximal end of the trunk piece of the segmented table (see Fig. 3, C, D). The output from the potentiometer was streamed to the data acquisition system (PowerLab/8SP, ADInstruments, Castle Hill NSW, Australia).

Dorsal neck muscle conditioning

Muscle conditioning was used to systematically alter the resting state of dorsal neck muscle fibres. Prior to muscle conditioning each participant was asked to perform a maximum voluntary contraction (MVC) of their dorsal neck muscles. This was repeated 3 times. Their MVC was calculated, and they were then prompted to practice contracting their dorsal neck muscles at 20% MVC using visual feedback when side-lying or verbal feedback when participants were supine. A voluntary muscle contraction of 10% MVC has been shown to be sufficient to eliminate previous history11.

For muscle conditioning the neck was first passively moved into either flexion (hold-long) or extension (hold-short). The participants were then asked to contract their dorsal neck muscles with a force of approximately 20% MVC with the instruction – “please extend your neck as if you are looking at the wall behind you”. This head position and muscle contraction was held for a period of 3 s after which time the participant was asked to relax their neck muscles for a further 5 s. The head-neck or neck-trunk was then returned to the neutral position and immediately the perception of the location of the target joint angle trial commenced (Fig. 3). Participants were required to keep their neck muscles relaxed for the duration of the trial as the effect of muscle conditioning is maintained as long as the dorsal neck muscles remain relaxed8.

Fig. 3
figure 3

Muscle conditioning protocol for dorsal neck muscles as conducted in supine (A) and side-lying (B) body postures. Panel C represents the 2 forms of muscle conditioning protocol – hold short and hold long. Horizontal lines represent changes in length of dorsal neck muscles (note, horizontal lines are separated for display purposes only). During hold long conditioning, dorsal neck muscles are lengthened from a neutral position (length). During hold short conditioning, dorsal neck muscles are shortened from the same midline length. Grey boxes indicate 20% MVC contraction of dorsal neck muscles.

Full size image

Surface EMG (sEMG) monitored neck muscle activity to confirm that neck muscles were active or at rest during muscle conditioning. After skin cleaning using 70% ethyl alcohol, for experiments conducted in the supine posture, bipolar electrodes (Blue Sensor, Blue Sensor, Medicostest, Denmark) were attached over either side of the dorsal neck muscles lateral to the C4 spinal process with an interelectrode distance (IED) of 50 mm12. The reference electrode was positioned over the T1 spinous process. For each sternocleidomastoid muscle recording, the electrodes were placed over the muscle belly in parallel with muscle fibres, with an interelectrode distance of 35 mm and the reference electrode at the level of the C6 spinous process12. For experiments in the side-lying posture, a bipolar wireless EMG system was used (Wave plus wireless EMG, Cometa, Milan, Italy) using the same IEDs as in the supine posture. The sEMG signals for all experiments were amplified with gain of 1000 times in voltage, filtered using a bandpass filter, 10–1000 Hz and a 50 Hz notch filter and recorded using a data acquisition system, Chart for Windows V7 (ADInstruments, Castle Hill NSW, Australia), sampled at 2000 Hz with 16 bits resolution. All sEMG data were visually inspected during the experiment and later repeated offline to ensure that dorsal neck muscles contracted and then remained relaxed as required for the muscle conditioning protocol.

Any trials where the sEMG signal was more than two standard deviations above baseline for longer than 40 milliseconds were excluded from further analysis. These parameters were used as an indicator of muscle activation and, therefore, the trial was excluded due to likely activation of muscle fibres including fusimotor fibres13.

Data analysis

Head-neck and neck-trunk joint position error was calculated for each trial using three parameters; constant error (CE), absolute error (AE) and variable error (VE)14. Constant error is the signed difference between the presented and subsequent perception of location of target angle; AE is the absolute difference between the presented target angle and the participant’s subsequent perception of location of the target angle; and VE is the square root of the CE of each (individual) trial for each condition (muscle condition and direction of presented target angle) minus the mean variance of the CE for all trials for each condition. Outliers were defined as the mean of the absolute error + 2 X standard deviations15 and were excluded from analysis.

A priori power analysis was conducted using G*Power (version 3.1.2) to determine the required sample size for a univariate analysis of variance (ANOVA) with two groups (experiment 1 and experiment 2). The analysis aimed for a power level of 0.80, an alpha level of 0.05, and an expected medium effect size (Cohen’s f) of 0.2516,17. Additionally, the partial eta-squared effect size was estimated at 0.06, and the correlation among repeated measures was assumed to be 0.50. The power analysis indicated that a total sample size of 24 participants would be necessary to achieve the desired power of 0.80. This sample size would enable the detection of a significant effect of group differences in the dependent variable, should such an effect exist. These parameters provide a robust framework for the proposed study, ensuring that the analysis will have adequate sensitivity to identify meaningful effects.

All statistical analyses were carried using IBM SPSS Statistics (version 29). Joint position error (defined as AE, CE and VE) was analysed against muscle condition (hold short and hold long) and posture (supine and side lying) using General Linear Model (GLM) univariate analysis of variance18. The assumptions (normality, homoscedasticity, and independence of the errors) behind the method were tested and were satisfied. In this analysis, joint position error is called the response variable while muscle condition and posture are called treatment effects. Even though participants are not of interest to be studied in this research, this variable is considered as a blocking effect and needs to be included in the analysis to account for it as a nuisance factor and/or to reduce the error term used in performing the test for the significance of the treatment effects. Moreover, pairwise comparisons were conducted for the statistical evaluation of CE and AE between muscle conditions considering the posture status. Parametric and non-parametric statistical tests for pairwise comparisons were also applied for CE and AE, as the sample size is small, and the results were the same. Note that the normality assumption was tested for all the data and was met. Significance was set at p < 0.05.

Results

Twenty-five participants complete this study in total. None of the participants experienced neck pain or discomfort associated with the conduct of the experiments. The mean value of error scores (CE, AE and VE) of individual trials for each participant were obtained and used for all subsequent analyses.

First experiment – Posture - supine

There were 2 participant groups. The first group of 7 participants were tested using a target angle of 10 degrees. Nine different participants completed the same experiment, but this time two target angles were used: 15 and 20 degrees. Each participant contributed 5 trials to each condition including (i) hold short with target angle presented at either 10° or 15° and 20° and (ii) hold long with target angle presented at either 10° or 15° and 20°. During each trial, the relaxed status of the neck muscles was checked by visual inspection of the sEMG, and trials with muscle activity were removed.

The first analysis compared CE, AE and VE scores between target angles 15 and 20 degrees (experiment 2, N = 9) after each form of muscle conditioning for AE, CE and VE. No significant differences were found between 15 and 20 the degrees target angle (see Fig. 4A).

We therefore used the participant data from the 20-degree target angle experiment and investigated whether there was a statistically significant difference in the magnitude of the joint position error (JPE) for CE, AE and VE between the target angle of 10 degrees (1st supine experiment, N = 7) and 20-degree target angle (2nd supine experiment, N = 9).

No significant differences were found between the 10- and 20-degrees target angles for the JPEs measured as AE, CE and VE (Fig. 4B). Therefore, a GLM univariate Anova analysis was conducted for AE, CE and VE scores using mean error scores from each participant from the 20-degree target angle experiment (experiment 2, N = 9) and 10-degree target angle experiment (experiment 1, N = 7) as the supine data set and the side lying data set (N = 9).

Fig. 4
figure 4

Panel A: Mean magnitude of Constant, Absolute and Variable Errors for participants (N = 9) perception of location of target angles 15 degrees (black filled circles) and 20 degrees (green filled squares). Panel B Mean magnitude of Constant, Absolute and Variable Errors for participants perception of location of target angles 10 degrees (black filled circles, N = 7) and 20 degrees (pink filled squares, N = 9) supine posture AEHS refers to absolute error after hold short conditioning. AEHL refers to absolute error after hold long conditioning. Likewise, CE refers to constant error and VE refers to variable error after hold short (HS) and hold long (HL) conditioning. No significant differences were found in the perception of location of any of the target angles after hold short or hold long conditioning. Error bars refer to 95% C.I.

Full size image

Joint position error (CE, AE, VE) analysis against muscle condition (Hold short and hold Long) and posture (Supine and side Lying)

Both the main effects (muscle condition; df = 1, p < 0.001 and posture; df = 1, p = 0.046) significantly contribute to the joint position error CE. Their interaction is also significant (df = 1, p = 0.017)) even though the blocking variable, participants, is not significant (df = 15, p = 0.477).

For AE, both main effects (muscle condition; df = 1, p = 0.006 and posture; df = 1, p < 0.001) significantly contribute to the joint position error AE. Their interaction is significant (df = 1, p = 0.019). Even though the blocking variable, participants, is not significant (df = 9) it should not be removed from the analysis (See Table 1).

Finally, for VE, both the main effects (muscle condition and posture) do not significantly (p = 0.262 and p = 0.249 respectively) contribute to the joint position error VE. Their interaction is not significant (p = 0.234) as well. The blocking variable, participants, is also not significant (p = 0.660).

Table 1 Statistical results of joint position error CE, AE and VE, using GLM univariate ANOVA method. Partial Eta square values and observed power values (parenthesis), p values and F values (parenthesis) are presented.
Full size table
Fig. 5
figure 5

Mean values of Absolute Error (AE), Constant Error (CE) and Variable Error (VE) after muscle conditioning (hold short and hold long) in the supine posture (black filled circles) and side lying posture (pink filled squares). Error bars represent 95% C.I. Note that when supine, the head was moved relative to the stationary trunk towards the head-trunk target angle. When side lying, the head remained stationary, and the trunk moved towards the trunk-head target angle.

Full size image

Pairwise comparisons of muscle condition considering posture status for CE

Pairwise comparisons between muscle conditions (hold short and hold long) in the supine posture for CE showed no significant difference (p = 0.333) and the mean difference between hold long and hold short was 1.500. For the side lying posture, a significant difference (p < 0.001) was found between hold long and hold short with a mean difference of 7.843 (see Fig. 5, left panel).

Pairwise comparisons of condition considering posture status for AE

Pairwise comparisons between muscle conditions (hold short and hold long) in the supine posture for AE showed no significant difference (p = 0.708) with a mean difference 0.412. For the side lying posture, a significant difference (p = 0.002) was found between hold long and hold short with a mean difference 4.870 (see Fig. 5, middle panel).

Pairwise comparisons of condition considering posture status for VE

Pairwise comparisons between muscle conditions (hold short and hold long) in the supine posture for VE showed no significant difference (p = 0.061) with a mean difference 1.188. For the side lying posture, there was also no significant difference (p = 0.964) was also found between hold long and hold short with a mean difference 0.037 (see Fig. 5, right panel).

Discussion

This study revealed three significant findings. First, dorsal neck muscle conditioning results in systematic differences in the perception of the location of the target angle of the trunk relative to the head when constant and absolute errors are considered. When conditioned as hold long that leaves dorsal neck muscle spindles mechanically less sensitive, accuracy in the perception of the location of the target angle given by the magnitude of the absolute and constant errors is decreased, but, only when participants are positioned in the side-lying posture (see Table 1; Fig. 5). In the supine posture, there was no significant difference in the magnitude of error after the 2 forms of muscle conditioning. Finally, there was no significant difference in the size of the variable error associated with dorsal neck muscle conditioning in both supine and side lying postures which indicates that the scatter of error around the mean value was consistent across the different forms of muscle conditioning and changes in body posture.

Our results indicate that when neck muscle proprioceptive afferent input is experimentally reduced after hold long conditioning, vestibular signals are reweighted enabling increased accuracy in head-trunk joint position sense when compared to the side lying posture and same form of muscle conditioning19,20,21. Additionally, our results indicate that muscle conditioning of neck muscles is a useful tool to modulate neck muscle spindle input independently from the vestibular system.

Muscle conditioning, a method that systematically alters the mechanical sensitivity of muscle spindles has been mainly applied to limb muscles but rarely tested on neck muscles to date. We are aware of only one previous study that had used muscle conditioning applied to the neck muscles which found that head-neck position sense was altered in response to conditioning22. However, there is some question whether the method used in that study effectively controlled for muscle history effects given that participants were seated during the experiment, because this posture would necessarily require neck muscles to actively support the head against gravity and muscle contraction attenuates muscle history (i.e., muscle conditioning) effect. In contrast, the current study maintained neck muscles in a relaxed state after muscle conditioning and we found that head-neck position sense was dependent on the form of muscle conditioning applied. In addition, Owens et al. only applied muscle conditioning at a shorter than test length, which increases the mechanical sensitivity of muscle spindles. Our study applied muscle conditioning at both a shorter than test length and longer than test length. This protocol allowed us to directly compare changes in head-neck/head-to-trunk JPS between the two defined mechanical states of intrafusal muscle fibres of the muscle spindles, that is, whether the muscle spindles were tight (hold short) or slack (hold long). The present results are consistent with previous reports which applied muscle conditioning to the limb muscles which concluded that hold-long conditioning causes larger joint position error23. We propose that the increase in CE indicating a pronounced bias to over-estimate the location of the pre-presented target angle occurring after hold-long conditioning in the side lying posture (see Fig. 5) was the result of decreased spindle afferent input from dorsal neck muscles as a consequence of the decreased mechanical sensitivity of dorsal neck muscle spindles as occurs in limb muscles7. This finding was also reflected in the significantly increased magnitude of the absolute error in the side lying posture after hold long conditioning.

On the other hand, there was no significant difference in VE associated with muscle conditioning for both supine and side-lying postures. Variable error has been described as an estimate of ‘system noise’ – which limits the transfer of actual proprioceptive information24. It is hypothesised that system noise may occur in the presence of nociceptive signals25. Thus, VE was not affected by muscle conditioning, which does not induce nociception.

One criticism on the application of muscle conditioning to neck muscles may be the more complex morphology of neck muscle proprioceptors than those in limbs, including complex spindle arrays and the presence of spindle-tendon organ complexes26,27,28. However, another study used muscle conditioning in feline lower back muscles, which also exhibit a complex morphology of muscle spindles like the neck. This study provided supporting evidence of conditioning dependent changes in the discharge rate of muscle spindle afferents where hold-long conditioning induced a lower discharge rate of muscle spindle afferents in comparison to hold-short conditioning where the discharge rate was higher29. Therefore, an alteration in head-neck position sense by muscle conditioning found in the present study is likely attributed to a change in discharge rate of neck muscle proprioceptor afferents.

There are limitations to this study. First, and most importantly, the sample size is small. This results in a lack of generalisability in the results of this present study. Secondly, there is uncertainty of the influence of the muscle conditioning protocol on the ventral neck muscles. A high density of muscle spindles present in the deep neck extensor muscles28,30 are also present in the deep flexor muscles27. Although we monitored neck muscle activity using sEMG, information on deep neck flexor muscles was not obtained. Application of the neck muscle conditioning protocol to the neck flexor muscles will clarify the contribution of proprioceptors in the ventral neck muscles to the head-neck position sense. This is being examined in a current study.

Neck pain and/or injury is a common musculoskeletal condition in society31,32,33,34. A significant proportion of individuals with neck pain and/or injury complain of concomitant symptoms such as dizziness35 and postural instability2. It is proposed that these concomitant symptoms are a consequence of aberrant neck proprioceptive inputs occurring because of pain1, injury to proprioceptors, and/or a mismatch between abnormal neck afferent input and normal information from the vestibular and visual systems2,3. There is abundant literature pointing to altered balance and increased risk of falls in people with neck pain, especially in the elderly (see for example36,37,). If neck pain is associated with muscle injury or reduced neck movement then taking advantage of muscle thixotropic effects in neck muscles may be considered a rehabilitative tool to increase spindle discharge prior to changes in posture.

Further, the application of the muscle conditioning protocol to neck muscles may be a useful tool for examining the functional association of neck proprioceptors with other organs such as limb muscles and the cardiovascular system.

In conclusion, the results of this pilot study have shown that when the neck-trunk is moved toward the target angle, dorsal neck muscle conditioning-dependent alterations in the accuracy of locating the position of a pre-presented target angle occurs. In addition, vestibular influences may reweight to positively influence head-neck joint position sense accuracy when the head and neck are both moved in the supine posture. The use of the muscle conditioning protocol applied to neck muscles can be a non-invasive tool to further investigate the role of neck muscle proprioceptors in clinical sequelae associated with sensorimotor disturbances associated with neck disorders.