Simple non-invasive measurement of rapid eye vibration

https://doi.org/10.1016/S0022-460X(02)01150-1Get rights and content

Abstract

A simple, non-invasive method for the measurement of eye vibrations above 30 Hz is described. The method can be used in either laboratory or natural conditions, and is based on the cancellation of an illusion of motion that occurs when two nearby light sources flickering in counterphase above the flicker fusion limit are observed during eye vibration. In these conditions, the light sources appear to oscillate in space at a frequency equal to the difference between the vibration and flicker frequencies. The frequency of eye vibration can be determined by adjusting the flicker frequency until the illusion disappears (i.e., until the difference frequency becomes zero). The same set-up can also be used to determine the amplitude of eye vibration, by adjusting the spatial separation between the two light sources until the oscillation appears to be the result of their bouncing off each other upon contact. The reliability and sensitivity of this method are illustrated with data from three observers whose eyes were vibrated with a commercial massager applied onto their neck, and using three different settings for the speed of the massager.

Introduction

Dynamic responses of the human body to vibration have been the subject of study for a long time, and the effect of whole-body vertical vibration on visual performance has also been studied [1], [2], [3], [4], [5], [6], [7]. In laboratory conditions subjects can be exposed to whole-body vibration with specified parameters and some times a comparison has been made between the effect of body vibration when the visual stimulus is static and the effect of stimulus vibration when the observer is static. Body vibration and stimulus vibration are nominally identical in these cases, but it is known that body posture, seating conditions, and anthropometrical characteristics of the subjects result in the actual vibration of the head and eyes being different from its nominal level [8]. These variations result in individual differences in the actual vibration transmitted to the head and eyes [9]. Some studies have shown that stimulus vibration deteriorates vision more than the same magnitude of subject vibration at frequencies below 6 Hz [3], [10], but the opposite occurs at higher frequencies [10]. These results suggest that body, head, and eye resonance alter the nominal parameters of subject vibration. A fair comparison of the effect of stimulus vibration with that of subject vibration could be made if the actual vibration of the eyes (when the subject vibrates) could be measured and its parameters applied to the stimulus (when the subject is static). Indeed, visual performance during whole-body vibration depends on the quality of the retinal image, which in turn depends only on how the eye vibrates regardless of the parameters of the vibration source. Note also that the use of accelerometers on bite bars to determine head vibration is insufficient for this purpose, because the eyes do not vibrate rigidly with the cranium.

Out of the laboratory, the human body is daily exposed to high-frequency (30–80 Hz) vibration that must also affect visual performance. The activities providing this vibration include most forms of transportation and the operation of some types of industrial machinery, and also such mundane tasks as the use of an electric toothbrush or other home appliances (e.g., massagers). A precise description of the vibration source is difficult to achieve but, again, measuring the vibration of the eyes in these circumstances is all that is needed to study their effects on visual performance and allow a comparison with results obtained in the laboratory.

Eye vibrations of high frequency and low amplitude are not easy to measure. Conventional equipment for the recording of eye movements at high spatial and temporal resolutions either involves invasive methods (scleral coils) or is cumbersome and requires head restraint (pupil tracking or Purkinje-image methods). These characteristics make eye trackers difficult to utilize along with other apparatuses in complex experimental settings. Lee and King [11] proposed a blur-cancellation method in which the subjects alter the vibration of the stimulus until it matches the vibration of the eye, a method based on the principle that when eye and stimulus vibrate synchronously the perception of blur caused by vibration will disappear. The applicability of this method is restricted to laboratory conditions in which the same source makes the stimulus and the subject vibrate, so that the frequency of vibration is known and only the amplitude and phase of the vibration of the eyes needs to be determined.

The goal of this paper is to describe a simple non-invasive method for the measurement of high-frequency (above 30 Hz) eye vibrations, a method based on a cancellation strategy that is widely applicable in either laboratory or natural conditions. The method is based on an illusion of motion that occurs when a display that flickers beyond the critical fusion frequency (about 30 Hz; see Levinson [12]) is observed under mechanical vibration of the eyes [13]. The method allows measuring both the frequency and the amplitude of eye vibration, and its rationale is described next.

Fig. 1 illustrates the principle underlying the illusory perception of motion described by Peli and Garcı́a-Pérez [13] when the eyes vibrate while looking at two nearby point sources that are flickering in counterphase above the critical fusion frequency. With still eyes, the lowpass temporal characteristic of the visual system [12] filters out the flicker at each retinal location, and the light sources are perceived as static and continuous: each light source is flickered onto a single retinal location (Fig. 1a, left) and temporal lowpass filtering produces the perception of two continuously illuminated dots that are static on the retina (Fig. 1a, right; see the appendix for computational details). Yet, when the eye vibrates, these flickering lights are effectively swept over the retina continuously and, if the frequency of the vibration differs from the flicker frequency (Fig. 1b, left), the same temporal lowpass filtering results in the apparent relative distance between the light sources varying as a function of time, which causes the illusion by giving the impression of relative motion (Fig. 1b, right). On the other hand, eye vibration at the same frequency (or an integer multiple) of the flickering light results in each light source undergoing the same exact sweep on the retina during its ON phase, resulting in the perception of blurred images that maintain a constant spatial separation over time (Fig. 1c). In these conditions, the relative phases between flicker and vibration affect the spread of the perceived blur, but in any case the perceived image is that of two static dots. Finally, eye vibration in the absence of flicker merely results in blurred images (Fig. 1d) without any illusory motion.

Then, when the eyes vibrate at an unknown rate, the frequency of vibration can be determined by adjusting the counterphase flicker frequency of two nearby light sources so as to reach the situation depicted in Fig. 1c, where the amount of blur may vary depending on the relative phases of flicker and vibration but the two light sources will appear static. The flicker frequency at which the illusory motion is cancelled is indeed the frequency at which the eyes are vibrating.

Note also that the stimulus set-up illustrated in Fig. 1a can be used to determine the amplitude of the vibration. In the right panel of Fig. 1b, the illusory oscillatory paths of either light source do not overlap in space because the spatial separation of the two light sources is larger than the amplitude of the oscillation of either source on the retina. If the actual spatial separation of the light sources could be varied, the perception would vary from that of two oscillating dots whose paths are separated (when the separation between the light sources is similar or larger than that depicted in the left panel of Fig. 1a) to that of two oscillating dots whose paths intersect (for separations sufficiently smaller). Then, the amplitude of the vibration can be determined by adjusting the actual separation between the light sources so that they appear to just touch each other as they move along their oscillatory path.

Section snippets

Visual stimulus

The stimulus consisted of two sharp-edged, circular LEDs (each 4.5 mm in diameter) mounted on a precision slider that allowed varying the distance between the LEDs continuously in the range 1–32 mm. The slider could be positioned at any orientation on the frontal plane of the observer, and included a nonius scale that permitted accurate distance measurement (to 0.1 mm) of the actual edge-to-edge separation of the LEDs. The LEDs were made to flicker in square-wave counterphase through custom-made

Temporal frequency of eye vibration

Fig. 2 shows the frequency at which illusory motion was cancelled for each subject in each condition. The frequency of eye vibration measured with our cancellation method does not differ much from the frequency at which the massager vibrates in the stand-alone condition, although some minor differences can be observed. These minimal differences can reasonably be attributed to the effect of contact between the massager and the body (measurements with a stroboscope indicated that the fundamental

Conclusion

The cancellation method described here seems powerful for determining the frequency and amplitude of eye vibrations above 30 Hz, although the massager that was used to illustrate the workings of the method is not a good source for producing eye vibrations with specified parameters. Under the conditions of the measurements, data gathered in the same session (i.e., without postural changes) had little variability, whereas data gathered in different sessions (which involved postural changes and

Acknowledgements

This work was carried out at The Schepens Eye Research Institute, where MAGP was a Research to Prevent Blindness International Research Scholar also supported by Dirección General de Enseñanza Superior grant PB96-0597 and by a Schepens Eye Research Institute Career Enhancement grant to EP. EP was supported by National Institute of Health grants EY05957 and EY12890 and by NASA grant NCC-2-1039. The authors thank Elisabeth Fine for access to her eye tracker and Fernando Vargas-Martı́n and Shaun

References (16)

  • M.J. Griffin et al.

    A review of the effects of vibration on visual acuity and continuous manual control, part Ivisual acuity

    Journal of Sound and Vibration

    (1978)
  • M.J. Moseley et al.

    Sinusoidal and random whole-body vibrationcomparative effects on visual performance

    Aviation, Space, and Environmental Medicine

    (1982)
  • M.J. Moseley et al.

    Effects of display vibration and whole-body vibration on visual performance

    Ergonomics

    (1986)
  • M.J. Moseley et al.

    Whole-body vibration and visual performancean examination of spatial filtering and time-dependency

    Ergonomics

    (1987)
  • H. Seidel et al.

    Isolated and combined effects of prolonged exposures to noise and whole-body vibration on hearing, vision and strain

    International Journal of Occupational and Environmental Health

    (1988)
  • B. Harazin et al.

    Influence of whole-body vertical vibration on vision performance

    Journal of Low Frequency Noise and Vibration

    (1996)
  • B. Harazin

    Study of effects of whole-body vibration on visual acuity

    Journal of Low Frequency Noise, Vibration and Active Control

    (1999)
  • M.J. Griffin

    Vertical vibration of seated subjectseffects of posture, vibration level and frequency

    Aviation, Space, and Environmental Medicine

    (1975)
There are more references available in the full text version of this article.

Cited by (3)

  • Ocular vestibular evoked myogenic potentials to bone conducted vibration of the midline forehead at Fz in healthy subjects

    2008, Clinical Neurophysiology
    Citation Excerpt :

    These results from human subjects pointing to otolithic origin are in accord with the physiological evidence showing that in the guinea pig bone conducted vibration is an ineffective stimulus for primary semicircular canal afferents, whereas most primary otolith irregular neurons respond vigorously to bone conducted vibration. The n10 to Fz BCV is not caused by actual vibration of the eye ball (Garcia-Perez and Peli, 2003) because n10 occurs after the end of the vibration stimulus and in patients without vestibular function any vibration of the eyeball would be expected to be similar to that in normal subjects. However, patients with no vestibular function showed no n10 response.

View full text