Gary D. Paige

Falls in the Elderly: Reliability of a Classification System

To determine risk factors for falls, previous studies have classified falls according to the contribution of factors both intrinsic and extrinsic to the host. Due partly to the lack of operational definitions and the absence of information on reliability, no consensus on classification has been reached. Consequently, in a 3‐year prospective study of falls occurring in a probability sample of community‐dwelling elderly (n = 1,358), a fall classification system was developed and tested for interrater reliability. The 366 falls in the first year of the study were independently classified by two reviewers on the basis of a narrative description and structured interview. The falls in the four major categories of the classification system included: falls related to extrinsic factors (55%), falls related to intrinsic factors (39%), falls from a non‐bipedal stance (8%) and unclassified falls (7%). The interrater reliability for the four major categories was 89.9% with a kappa of 0.828. The system provides operational definitions for types of falls and a reliable and flexible method for classifying falls in the elderly.

The influence of target distance on eye movement responses during vertical linear motion.

SummaryStudies of the linear vestibulo-ocular reflex (LVOR) suggest that eye movement responses to linear head motion are rudimentary. This may be due to inadequate control of target distance (D). As D approaches infinity, eye movements are not required to maintain retinal image stability during linear head displacements, but must become increasingly large as D shortens. The LVOR in the presence of visual targets (VLVOR) was tested by recording human vertical eye and head movements during self-generated vertical linear oscillation (averaging 2.7 Hz at peak excursion of 3.2 cm) while subjects alternately fixated targets at D=36, 142, and 424 cm. Response sensitivity rose from 0.10 deg/cm (5.8 deg/s/g) for D=424 cm to 0.65 deg/cm (37.5 deg/s/g) for D=36 cm. Results employing optical manipulations, including spherical lenses to modify accommodation and accomodative convergence, and prisms to modify fusional vergence without altering accommodation, imply that the state of vergence is the most important variable underlying the effect. Trials in darkness (LVOR) and with head-fixed targets (visual suppression of the LVOR) suggest that, while visual following and perhaps “mental set” influences results, the major proportion of the VLVOR response is driven by vestibular (presumably otolith) inputs.

Senescence of human visual-vestibular interactions: smooth pursuit, optokinetic, and vestibular control of eye movements with aging

Natural aging entails progressive deterioration in a variety of biological systems. This study focuses on visual and vestibular influences on human eye movements as a function of aging. Eye movements were recorded (search-coil technique) during visual, vestibular, and combined stimuli in subjects across a broad range of ages (18–89 years). Two types of visual following were assessed: smooth pursuit (SP) of a small discrete target, and optokinetic (OKR) following of a large-field striped image. The vestibulo-ocular reflex (VOR) was studied during head rotation in darkness. Visualvestibular interactions were recorded during rotation in two ways: when the optokinetic scene was earth-fixed, resulting in visual enhancement of the VOR (VVOR), and when the visual image was head-fixed, allowing visual suppression of the VOR (VSVOR). Stimuli consisted of horizontal sinusoidal oscillations over the frequency range 0.025–4 Hz. Trials were analyzed to yield response gain (peak horizontal eye/stimulus velocities) and phase (asynchrony, in degrees, between eye and stimulus velocity signals). VOR gain in young subjects was greatest (near 0.9) at 2.5–4 Hz but declined steadily with decreasing frequency, while phase hovered near zero until 0.1 Hz and then developed a progressively increasing lead. Effects of advancing age were small, given the modest head velocities presented, and were most noticeable as an increase in phase lead and decline in gain at the lowest frequencies (≤0.1 Hz). The two forms of visual following and all conditions of visual-vestibular interactions displayed more prominent age-dependent changes. OKR and SP response characteristics (0.25–4 Hz) closely resembled each other. Gain was greatest at 0.25 Hz, while phase was near 0°. As frequency increased, gain declined while phase lag rose. However, both gain and phase lag tended to be slightly greater for OKR than for SP responses. Both SP and OKR response properties deteriorated progressively with increasing age, as witnessed by a progressive decline in gain and increase in phase lag, even at modest frequencies (e.g., 0.25–1.0 Hz). VVOR responses were generally closer to the ideal of 1.0 in gain and 0° in phase than either the VOR or visual following alone. A subtle but significant age-dependent decline in VVOR performance occurred at the lowest frequencies. VSVOR response characteristics were close to those of the VOR and VVOR at 4 Hz, where visual influences on eye movements are generally inconsequential. As frequency declined, visual suppression became more robust and gain dropped. The SP stimulus seemed surprisingly more effective than the OK scene in suppressing the VOR, but this effect is predicted by a linear model of visual-vestibular interactions. As age increased, visual influences on the VOR became progressively weaker, in concert with deterioration of visual following. The subjective sensation of circular vection (CV), a psychophysical measure of VVI, was assessed during optokinetic stimulation at 0.025 Hz. Interestingly, the likelihood and intensity of CV increased with aging, suggesting that visual inputs to the perception of self-motion are enhanced in the elderly. This may represent a form of visual compensation for age-dependent loss of vestibular self-rotation cues. In brief, the VOR, visual following, and their interactions display specific changes in response properties as a function of natural aging. The modifications may be interpreted as age-dependent deteriorations in the performance of systems underlying the control of human eye movements.

Plasticity in human sound localization induced by compressed spatial vision

Auditory and visual target locations are encoded differently in the brain, but must be co-calibrated to maintain cross-sensory concordance. Mechanisms that adjust spatial calibration across modalities have been described (for example, prism adaptation in owls), though rudimentarily in humans. We quantified the adaptation of human sound localization in response to spatially compressed vision (0.5× lenses for 2–3 days). This induced a corresponding compression of auditory localization that was most pronounced for azimuth (minimal for elevation) and was restricted to the visual field of the lenses. Sound localization was also affected outside the field of visual–auditory interaction (shifted centrally, not compressed). These results suggest that spatially modified vision induces adaptive changes in adult human sound localization, including novel mechanisms that account for spatial compression. Findings are consistent with a model in which the central processing of sound location is encoded by recruitment rather than by a place code.

Visually-induced adaptive plasticity in the human vestibulo-ocular reflex

SummaryThe vestibulo-ocular reflex (VOR) is under adaptive control which corrects VOR performance when visual-vestibular mismatch arises during head movements. However, the dynamic characteristics of VOR adaptive plasticity remain controversial. In this study, eye movements (coil technique) were recorded from normal human subjects during sinusoidal rotations in darkness before and after 8 h. of adaptation to 2X binocular lenses. The VOR was studied at 7 frequencies between 0.025 and 4.0 Hz at 50°/s peak head velocity (less for 2.5–4 Hz). For 0.025 and 0.25 Hz, the VOR was tested at 4 peak head velocities between 50 and 300° /s. Before 2X lens adaptation, VOR gain was around 0.9 at 2.5–4.0 Hz and dropped gradually with decreasing frequency to under 0.6 at 0.025 Hz. Phase showed a small lead at the highest frequencies which declined to 0° as frequency decreased to 0.5–0.25 Hz, but then rose to 14° by 0.025 Hz. VOR gain was independent of head velocity in the range 50–300°/s at both 0.025 and 0.25 Hz. However, Phase lead rose with increasing head velocity, more so at 0.025 than at 0.25 Hz. After 2X lens adaptation, gain rose across the frequency bandwidth. However, the proportional gain enhancement was frequency dependent; it was greatest at 0.025 Hz (44%), and declined with increasing frequency to reach a minimum at 4 Hz (19%). Phase lead increased after 2X lens adaptation at lower frequencies, but decreased at higher frequencies. New velocity-dependent gain nonlinearities also developed which were not present prior to adaptation; gain declined as peak head velocity increased from 50 to 300°/s at both 0.025 (23% drop) and 0.25 Hz (15% drop). This may suggest an amplitude-dependent limitation in VOR adaptive plasticity. Results indicate both frequency and amplitude dependent nonlinearities in human VOR response dynamics before and after adaptive gain recalibration.

Tilt perception during dynamic linear acceleration

Abstract Head tilt is a rotation of the head relative to gravity, as exemplified by head roll or pitch from the natural upright orientation. Tilt stimulates both the otolith organs, owing to shifts in gravitational orientation, and the semicircular canals in response to head rotation, which in turn drive a variety of behavioral and perceptual responses. Studies of tilt perception typically have not adequately isolated otolith and canal inputs or their dynamic contributions. True tilt cannot readily dissociate otolith from canal influences. Alternatively, centrifugation generates centripetal accelerations that simulate tilt, but still entails a rotatory (canal) stimulus during important periods of the stimulus profiles. We reevaluated the perception of head tilt in humans, but limited the stimulus to linear forces alone, thus isolating the influence of otolith inputs. This was accomplished by employing a centrifugation technique with a variable-radius spinning sled. This allowed us to accelerate the sled to a constant angular velocity (128°/s), with the subject centered, and then apply dynamic centripetal accelerations after all rotatory perceptions were extinguished. These stimuli were presented in the subjects’ naso-occipital axis by translating the subjects 50 cm eccentrically either forward or backward. Centripetal accelerations were thus induced (0.25 g), which combined with gravity to yield a dynamically shifting gravitoinertial force simulating pitch-tilt, but without actually rotating the head. A magnitude-estimation task was employed to characterize the dynamic perception of pitch-tilt. Tilt perception responded sluggishly to linear acceleration, typically reaching a peak after 10–30 s. Tilt perception also displayed an adaptation phenomenon. Adaptation was manifested as a per-stimulus decline in perceived tilt during prolonged stimulation and a reversal aftereffect upon return to zero acceleration (i.e., recentering the subject). We conclude that otolith inputs can produce tilt perception in the absence of canal stimulation, and that this perception is subject to an adaptation phenomenon and low-pass filtering of its otolith input.

Nonlinearity and Asymmetry in the Human Vestibulo-ocular Reflex

The horizontal vestibulo-ocular reflex was studied in normal and unilaterally vestibulopathic human subjects; the latter before and after surgical vestibular ablation. Subjects were oscillated at 0.025-4 Hz, 50 o/s peak head velocity, and for 0.025 and 0.25 Hz, at 50-300 o/s peak head velocity. Gain (peak eye/head velocities), phase (delay in degrees between peak eye and head velocities), and asymmetry (including d.c. bias; the average eye velocity over the cycle) were assessed for each rotation. In patients preoperatively, gain was subnormal while phase lead was greater than normal only at lowest frequencies. As head velocity increased at 0.025 and 0.25 Hz, gain dropped in patients, but not in normals. Furthermore, d.c. bias at 0.25 Hz tended to shift abnormally toward the lesioned side in most patients. After surgery, gain dropped and phase lead rose for all stimuli, but showed varied recovery over 4 months. d.c. bias also rose after surgery, and showed an enhancement with increasing head velocity at 0.25 Hz. d.c. bias remained unchanged even after 4 months for 300 o/s rotations, but disappeared at low head velocity. d.c. bias (and other asymmetry measures) represents a lateralizing sign which can be improved by utilizing high head velocities at moderate frequencies.

Characteristics of the VOR in Response to Linear Acceleration

Abstract: The primate linear VOR (LVOR) includes two forms. First, eye‐movement responses to translation [e.g., horizontal responses to interaural (IA) motion] help maintain binocular fixation on targets, and therefore a stable bifoveal image. The translational LVOR is strongly modulated by fixation distance, and operates with high‐pass dynamics (>1 Hz). Second, other LVOR responses occur that cannot be compensatory for translation and instead seem compensatory for head tilt. This reflects an otolith response ambiguity‐that is, an inability to distinguish head translation from head tilt relative to gravity. Thus, ocular torsion is appropriately compensatory for head roll‐tilt, but also occurs during IA translation, since both stimuli entail IA acceleration. Unlike the IA‐horizontal response, IA torsion behaves with low‐pass dynamics (with respect to “tilt”), and is uninfluenced by fixation distance. Interestingly, roll‐tilt, like IA translation, also produces both horizontal (a translational reflex) and torsional (a tilt reflex) responses, further emphasizing the ambiguity problem. Early data from subjects following unilateral labyrinthectomy, which demonstrates a general immediate decline in translational LVOR responses, are also presented, followed by only modest recovery over several months. Interestingly, the usual high‐pass dynamics of these reflexes shift to an even higher cutoff. Both eyes respond roughly equally, suggesting that unilateral otolith input generates a binocularly symmetric LVOR.

Auditory spatial perception dynamically realigns with changing eye position

Audition and vision both form spatial maps of the environment in the brain, and their congruency requires alignment and calibration. Because audition is referenced to the head and vision is referenced to movable eyes, the brain must accurately account for eye position to maintain alignment between the two modalities as well as perceptual space constancy. Changes in eye position are known to variably, but inconsistently, shift sound localization, suggesting subtle shortcomings in the accuracy or use of eye position signals. We systematically and directly quantified sound localization across a broad spatial range and over time after changes in eye position. A sustained fixation task addressed the spatial (steady-state) attributes of eye position-dependent effects on sound localization. Subjects continuously fixated visual reference spots straight ahead (center), to the left (20°), or to the right (20°) of the midline in separate sessions while localizing auditory targets using a laser pointer guided by peripheral vision. An alternating fixation task focused on the temporal (dynamic) aspects of auditory spatial shifts after changes in eye position. Localization proceeded as in sustained fixation, except that eye position alternated between the three fixation references over multiple epochs, each lasting minutes. Auditory space shifted by ∼40% toward the new eye position and dynamically over several minutes. We propose that this spatial shift reflects an adaptation mechanism for aligning the “straight-ahead” of perceived sensory–motor maps, particularly during early childhood when normal ocular alignment is achieved, but also resolving challenges to normal spatial perception throughout life.

Caloric responses after horizontal canal inactivation.

Horizontal eye movements in squirrel monkeys were recorded in response to ice-water caloric stimuli before and after horizontal canal (HC) inactivation, achieved by a plugging procedure. Calorics were presented with the subjects supine and prone, and the maximum slow phase eye velocity (SPEV) of the induced nystagmus was assessed. SPEV responses from normal ears were always directed toward the stimulated side when the monkeys were supine, and toward the opposite side when they were prone. Supine responses were always greater than prone ones. After HC-plug, which abolishes the canals mechanical response to rotatory or convection current stimulation, small SPEV responses were routinely observed. However, they were always directed toward the stimulated side, regardless of head orientation. The results are consistent with the notion that the normal caloric response is composed of the sum of a convention current component which is dependent upon head orientation, a smaller position-dependent component of unclear origin, and a direct temperature effect on the canals sensory apparatus which is independent of head orientation.