Effects of neck muscle vibration on subjective visual vertical in an altered visual frame of reference

The human sense of vertical (subjective visual vertical) is strongly infl uenced by inputs from the visual, vestibular and somatosensory systems. Sensory information conveyed by each system is assigned a different “weight” that can vary depending on the individual, the environment, the availability of said information, the task, as well as from unanticipated or abrupt changes in the sensory input stream itself [1,2].


Introduction
The human sense of vertical (subjective visual vertical) is strongly infl uenced by inputs from the visual, vestibular and somatosensory systems. Sensory information conveyed by each system is assigned a different "weight" that can vary depending on the individual, the environment, the availability of said information, the task, as well as from unanticipated or abrupt changes in the sensory input stream itself [1,2].
The rod and frame test (RFT) has been used previously to investigate the effects of an altered visual frame of reference on subjective visual vertical [3,4]. The RFT requires an individual to align a luminous rod as best they can with gravity (i.e. true vertical). Their vision, however, is restricted by a square frame that can be rotated clockwise or counterclockwise. This provides the individual with an illusory visual frame of reference from which to base their estimation of vertical. Participants' subjective visual vertical (quantifi ed as their error from true vertical) can provide insight into the role of individual sensory systems in perceiving vertical as well as individual participant's reactions to a visual illusion. Previous work has shown that participants' errors are generally in the direction of the frame's orientation [3][4][5].
Muscle vibration has noted effects on proprioceptive integration, and, by stimulating both primary and secondary muscle afferents, appears to provide the illusion of muscle lengthening [6][7][8] [10].
In this study, we were interested in examining how the head orientation and perception illusions generated by ventral neck vibration would affect a participant's subjective sense of vertical in an altered (illusory) visual frame of reference. In this case, as our participant's heads rested on a chin piece, they were unable to move their head in response to the vibratory stimulus. While it is known that the illusory effects of muscle vibration can be diminished by light touch with a static object, we postulated that our participant's sense of vertical might still be affected by the vibratory illusion of self-motion, even without an overt ability to move their head [11,12]. In addition, by utilizing an RFT we provided an illusory close visual frame of reference which we believed might infl uence the effects of the neck vibration. In this way, we aimed to examine the effect of two sensory illusions (one visual modality and the other somatosensory) used in conjunction on an individual's subjective sense of vertical.

Experiment
During the experiment, participants stood in front of a modifi ed version of Oltman's rod and frame apparatus (Oltman 1968) [13]. Each participant rested their chin on an ophthalmological chin rest that was adjusted to center their vision into a 24" long, 12" square tunnel. The tunnel rotated around a central axis and their peripheral vision was restricted such that the only reference and visual input they could use was the frame (tunnel) and a monitor at the far end of the frame.
The monitor displayed a white rod on a black background that could be rotated using a wireless gamepad controller (Logitech F710 Gaming Controller) held at waist level. Participants were asked to use the controller to align the rod, as best they could, with gravity (i.e. true vertical) over the course of 16 different trials. In between trials, participants were asked to close their eyes while the position of the frame was changed between 18 degrees clockwise, and 18 degrees counterclockwise, where the inclination of the frame has been found to elicit its greatest effect [14]. The starting position of the rod was either 18 degrees clockwise, or 18 degrees counterclockwise relative to the position of the frame, depending on the trial. Prior to testing, participants were provided an opportunity to interact with the gamepad controller to become familiar with the control features Figure 1.  Table 1.

Statistical analysis
Data normality was assessed using Shapiro-Wilk and Kolmogorov-Smirnov tests.
As data were not normally distributed, baseline, frame tilt, vibration and combined frame/vibration conditions were Citation: Malaya

Results
There was a statistically signifi cant difference in baseline, frame, vibration and combined conditions,  2 (2) = 109.212, p < 0.001 Table 2. Across all participants, in the frame only conditions, frame right was not signifi cantly different than frame left (p > 0.05).
In the combined conditions, there was no signifi cant difference between the side of frame tilt and the same side of tilt frame only conditions (p > 0.05).

Discussion
This study found that participants were strongly affected by the position of a close visual frame when attempting to orient a rod to vertical. Consistent with previous work utilizing an RFT, their errors were most often made in the direction of frame tilt [3][4][5]. As several investigations have shown the effects of altered close visual frame of reference on subjective visual vertical to be robust through a range of different postures and body orientations, it is little surprise that our participants were likewise affected [3][4][5]. It is of note, however, that the RFT infl uenced our participants even though their heads were stabilized in a chinrest. This suggests that the infl uence of the close visual frame of the RFT is independent of, or even primary over, relevant somatosensory information from the ground and the chinrest as it relates to subjective visual vertical.
The lack of response to neck vibration was likely multifactorial. Malmstrom, et al. has previously shown a lower response to vibration applied to the ventral musculature as compared to dorsal, with stepping-in-place tasks [15]. However, the authors still saw signifi cant change in task with both ventral and dorsal vibration, while the current experiment saw no effect.
It is possible that the haptic contact from the chin rest provided corrective somatosensory information as to offset or overcome the vibratory illusion. This is supported by a study by Bove, et al. which found that light fi nger touch of a stationary surface reduced body tilt from short duration, single-sided muscle vibration of dorsal neck musculature and the sternocleidomastoid muscle. Similarly, Lackner, et al. [12] reported that haptic contact with a stable bar attenuated center of pressure changes from vibration of the right peroneal tendons in heel-to-toe stance participants. However, Karnath, et al. found signifi cant differences in participant's ability to align a laser pointer with their "subjective straight ahead," in spite of utilizing a chin rest [16]. This suggests the chinrest may not have been solely responsible for disrupting the vibration illusion.
Indeed, it is more likely that the rod and frame device provided enough of a structured visual context as to diminish the effects of the vibratory illusion. This is supported by previous work showing that proprioceptive vibratory stimulation was less impactful in a structured visual environment as compared to darkness [17,18]. This suggests that, in the context of the RFT (a structured visual environment), the vibratory illusion was diminished because, even though the close visual reference was confl icting with somatosensory inputs, it was still our participants' primary means of establishing verticality.
Gaerlan, et al. found that healthy adults primarily relied on their visual system for maintaining postural balance [19].
In a study of construction workers at elevation, all measured   [20]. It has been argued that "appropriate close visual references increase the ability to maintain balance" and that being elevated a distance from the ground removes some of the visual information needed to maintain postural balance [20]. While our study did not have an "appropriate" visual reference (that is, the frame did not accurately represent gravitational vertical), the simple task of standing during an RFT was not challenging enough to induce a fall in our participants and, as there was a close visual reference from the frame, the somatosensory input from the vibratory illusion was not upweighted to counter the confl icted visual input [21].
It is also possible that the somatosensory information from the neck, given altered visual input, was down weighted compared to somatosensory information from the feet. Interestingly, they found that trained dancers were more affected by tendon vibration and took longer to adjust to the illusion than untrained individuals. It has been argued that dance training shifts and individual's sensorimotor dominance from visual to proprioceptive [22][23][24][25][26][27][28][29][30][31][32][33][34]. While we did not collect any information on the athletic background of our participants, it was the case in our sample that the visual system appeared to be the primary estimator of verticality. While this appears to be dependent on population, it is notable that the RFT used in Thallasinos, et al. [5] did not have a three-dimensional, physical frame. Instead, the rod and frame were displayed on a large screen. It is possible then, that the visual system retains primacy for establishing vertical unless that information stream is actually absent. In Thallasinos et al., the participants were affected by vibration due to the lack of a physical, close referenced frame. Comparatively, in this study, the presence a close visual reference overrode the vibratory illusion. Said another way, in the presence of altered, even incorrect visual information in close reference, the visuals will remain weighted most heavily. More work in this area is needed to examine these effects, however. In particular, future work on this topic should consider enlarging and distancing the frame, and removing any other close, structured visual references, and examining the effects of muscular vibration on subjective visual vertical.