COORDINATION OF HEAD AND EYE MOVEMENI-S FIXATE CONTlNUOUS AND INTERMITTENT TARGETS

TO

M. A. GRESTY Royal Air Force Institute of Aviation Medicine. Farnborough. Hants (Rrcviced 29 Jugust 1973)

Abstract-To compare head-eye co-ordination with and without a target which was visible during the movement. subjects fixated in the direction of target lamps which: (i) flashed for 10 msec followed by 1 set darkness; (ii) were continuously illuminated for 3 sec. Ballistic head movements were produced which made up Sjand 75 per cent respectively of the gaze displacement in conditions (i) and (ii). Eye movements were a combination of voluntary step. vestibulo-ocular reflex, position correcting steps and slow movements. Saccades produced by passive head rotation were modified during similar fixation movements. Accuracy of responses to flash was high although response components showed wide variability.

INTRODL’CTlON

When the head and eyes are moved synchronously

how the reffex anticipatory saccade is modifted in target fixating movements, Furthermore it has been shown by Benson and Guedry (1970) that the amplitudes of slow phase movements and saccades produced by passive rotation are suppressed in the presence of a visual target which moves with the subject. This linding demonstrates that the vestibulc+ocular reflex may readily be modified by specific conditions of visual information.

so that the eyes may fixate a displaced target the eye movement consists of an initial wide amplitude, high velocity step (saccade) in the direction of the displacement followed by a slower (slow phase), smaller amplitude eye movement in the opposite direction. Mowrer (1932) attributed the slow phase eye displacement to the vestibulo-ocular reflex which was evoked by head It was the intention of the present experiment to examovement and demonstrated that the slow phase commine the relative involvement of vestibul@ocu!ar nyspensated for continuing head rotation thus stabilizing tagmus in voluntary head and eye fixation movements the eye with respect to the visual world. in the presence and absence of a visual frame of referRecently, more detailed examinations of the role of ence. This was achieved by contrasting the head and the vestibulo-ocular reflex in voluntary head and eye eye movements made to fixate a continuously visible movements have been made in man by Bartz (1966), target with those made to fixate in the direction of a Flemming et al. (1969) and Gresty (1973). and in pritarget which appeared for a brief period of time. Incimates by Bizzi et al. (1972). These authors have refined dentally, by measuring the accuracy with which the diMowrer’s original description. The slow phase of the rection of the brief target was fixated an estimate could eye movement compensates for the angular velocity of be made of how precisely spatial orientation is mainhead rotation and it is during this phase of coordinatained when head and eye displacements occur in the tion that the eyes achieve fovea1 fixation of the target. absence of a continuously visible frame of reference. Frequently the fixation is achieved by interrupting the slow phase eye movement with small saccades which EXPERIMEZITAL METHODS correct errors in position. Although the slow phase of vestibular nystagmus is The visual display consisted of a horizontal row of 13 shown to be useful under conditions of VOhintaryhead Gallium Arsenide light emitting diodes mounted in a 1.35m and eye movements in response to step displacements radius semicircular panorama. The lamps were spaced symmetrically at angles of I-%7’, 26.2”. 39%‘. 53.5’. 67.5’ and of target, it is not clear what role is played by the saccaSt.7’ to the right and left of a central lamp. The lamps die component of vestibuIc+ocular nysta_mus. Barnes (1973) has shown that during passive head rotation in emitted a deep red light. with a luminance of 170cd. mz and had rise times of less than 5 /~sec.Each subtended 5.1 min simulation of a voluntary displacement. the eye moveof visual arc at the subject who sat at the radius centre of ment included a saccadic component which anticipates the semicircle facing the central lamp. the future head position. The saccade occurs generally The angular position of the subject’s head was monitored just before peak head velocity is attained. In contrast, with a low torque position potentiometer which was the initial saccade produced when the head and eyes attached with a universal joint to a flying helmet chosen for are turned to fixate a target is usually finished we11 lightness and unobtrusiveness. The helmer fitted tightly over before peak head velocity. Thus the question arises of the subject’s head and ears and was held in place by a chin 395 YR.I.4ti

strap. Hortzontal e)e dtsplacements were recorded using silver jilter-chloride suction electrodes placed dosr to the external canthi.The signals from the EOG electrodes were processed using a ditferential dc. amplifier with a Hat response to above LOOHz and those from the position potentiometer were conditioned by a high impedance follower amplitisr. The amplifcd signals were displayed on an u.v. galvanomctsr recorder and stored on magnetic tape. The accuracy with which the eye displacement could be measurr’d was to =f’ and the head displacement to within 2 1’. The presentation of sequences of lights was controlled by a digital pulse generator. 1 .‘continuous target presentation” (CTP) condition provided the subject with continual visual information during his head and eye movements. It consisted of the illumination of the central lamp for 3 set immediately followed by a 3 set illumination of one of the 12 target lamps. In order to elicit head and eye movements in the absence of concurrent visual information the “target ffash” (TF) condition was devised. It consisted of the following sequence of events: (i) 3 set ilIumination of the central la”mpf (ii, 10 msec period of total darkness: (iii) 40 msec “Rash” illumination of one of the 11 target lamps; (iv) 1 set total darkness: and (v) Re-illumination of the target lamp for 2.5 sec. Under both conditions the subject initially fixated the central light until it was extinguished and then was free to move his head and eyes to fixate the target light when it appeared. In the case of TF the Bash had terminated long before any movement could be made so in this condition the subject was instructed to fixate in the direction in which he thought the Rash had occurred. When the target lamp reappeared he could fixate it accuratelv and thus correct any error he might have made whilst tixat~ng in the dark. Subjects were presented with the sequence of experimental conditions: TF. CTP, TF. CTP. Under each condition 24 trials avers run so that each of the 12 target lamps were presented twice. The order of presentation ofthe targets was randomized and a-as unknown to the subject. There was a pause of about 5 set between trials. At the beginning and end of the experiment and between conditions. the subject’s head was fixed in dead centre by means ofa chin rest and the 11-7’ and 26.1” iamps were iHuminated so that the EOG could be precisely cafibrated. Six subjects with normal vision were employed in the experiment and their results were used as the basis for the statistics quoted. Two additional subjects were also tested. One of these wore spectacles and the other meniscus contact lenses. Both optical aids corrected for myopic vision, but obviously differed greatly in the physical restraints they might respectivety exert to modify the normal patterns of head znd eye movements. The subjects were carefully aquainted with the experimental situation using examples of experimental conditions. but were given no opportunity to train in the situation. Stress was laid on the accuracy of fixation and not on speed. The whole experiment lasted about 15 min and was conducted in the dark after 10 min of dark adaptation.

head movements recorded durmy the tarfe,t spotttn: situation. The waveforms employed were hAt smusoids Lvith a fixed duration of O+jsec and peak amplirucies vsping betwen ZO’.‘secand iSO- sec. These values uere bassd on measurements made on the data from the target spotting situations. Recordings were taken of horizontal eye displacement by EOG and of the angular trloctty of the spin table. During the quiescent period between each deliver! uf a rotation. the subject was instructed to look in the direction which he thought was dead ahead. and he was given no cues for the moment when the rotation stimulus ‘11;~ 10 be applied.

Typical records of head and eye displacements made in the continuous target presentation and target hash conditions are illustrated in Fig. I. The addition of the head and eye displacement produces the function commonly termed ‘-gaze”. which is the direction of fixation. Hence the upper trace in each set of three represents the angular position of the eye in space with respect to the centre fixation point. Under both conditions. the time course of the head movement fotlowed a simple sigmoid curve [Fig. i(a)-(f)] which terminated in a constant displacement. By inspection it was apparent that the angular acceleration on initiation of the motion was always greater than the deceieraiion which brought the head to rest. The reaction times between the onset of the target lamp and the start of the head movement averaged over all subjects did not v-ary with angular displacement of the target, but were related to experimental condition. The mean reaction time for responses to CTP was 335 (SD 72) msec and for responses to TF was 308 (SD 5s) msec. a difference which is highly significant (P 0.01). Angular eye movements produced in response to CTP and TF were, in their early phases very similar. The eye movement under both conditions commenced with a high velocity saccade ( > 4003/sect made in the direction of the target [Fig. l(a)-(f)] which frequently slowed down as it reached the maximum displacement [e.g. Fig. f(a)]. The mean reaction times of the initial saccades made in response to CTP and TF, averaging over all subjects and target displacements, were 270 (SD 56) msec and 264 (SD 46) msec respectively and not significantly different. In responses to TF the slow phase movement of the eyes took place after peak eye displacement and before the reillumination of the target. The slow phase had the aooearance of an ex~onent~a[i~ decaying curve which matched head velohty and terminated in a conSq$emmrnr~ experiment stant displacement [Fig. l(a).(b)]. Throughout the .An zuoerimrnt aas ocrformed to collect more comprehensive hata concerning the vestibule-o-..‘“C’ re- motion the slow phase angular velocity almost perLLA‘ill,*flex fectly compensated for the angular velonty of the consponss to passive transient displace---*@ n the head. SubLI1\.IAIJ df jects were seated. with head and s houldets fixed and eyes current head rotations. On some occasions, before the target light appeared, the eyes executed stepwise “corclosed, on a servo-motor controlle :d spin table. The table was subjected to transient ho&c antal aneular displace- rcctive” movements [Fig. l(a),(b)]. Follovving the reap. . ___ct:_ _E .._l.__.___. ._ .__ “I _c Lrit! _l^ *____. __,__“WANJllall_Y ,,,__:,_,,~., _,,$a mcnts which simulated the velocity pr~nrs VI roru~nary pearance raryct il__