kinaesthesis

The sense by which motion, weight, and the position of various body parts are perceived. Kinaesthesis depends on the sense organs (proprioceptors called muscle spindles and golgi tendon organs) that monitor the state of contraction of muscles, skin receptors, and the balance receptors in the ear. These all pass information to the brain so that a person is aware of body and limb position during movements. This awareness, known as kinaesthetic perception, is important for many physical activities, especially those involving gymnastic movements.

Kinaesthetic perception can be determined by testing the ability to distinguish between different weights when blindfolded, or by timing how long a blindfolded person can maintain balance while standing on one leg (see blind stork test).

The sense by which motion, weight, and position of various body parts are perceived. Kinaesthesis depends on the sense organs, especially proprioceptors, skin receptors, and the vestibular apparatus providing information about the state of contraction of muscles and the position of the limbs and the body in space.

Almost everything that a human being does involves the perception of the spatial locations of objects. The senses used in perceiving spatial locations are known as the spatial senses: they are vision, hearing, touch, and kinaesthesis. Kinaesthesis is the sense which enables us to appreciate the positions and movements of limbs, and depends on receptors in muscles, tendons, and joints, as well as on the sense of muscular effort involved in moving a limb or holding it in a given position. Movements of the head are detected by a set of specialized sense organs in the head, known as the vestibular sense organs.

The task of judging the spatial location of an object is complicated by the fact that sense organs are attached to mobile parts of the body. For instance, the receptive surface of the eye (the retina) is attached to a mobile eyeball, which in turn is attached to a mobile head. If we wish to know the direction of a seen object with respect to the torso, the position of the eyes and of the head must be taken into account, along with information about the retinal position of the image of the object. This type of process is here referred to as sensory integration.

The position of an object is often detected by more than one sense organ at the same time. For instance, we may hold an object which we can also see and hear. Furthermore, we usually see with two eyes and hear with two ears. Spatial information must be coordinated between different sense organs, either two organs of the same type or belonging to different spatial senses. This is the process of intersensory coordination.

Finally, after we have located an object we may wish to reach for it. This requires that sensory spatial information be coordinated with the motor commands which control the movements of limbs. This is the process of sensorimotor coordination.

Performance of any spatial task has an accuracy and a precision. 'Accuracy' is the extent to which the mean of a set of judgements deviates from the true value. 'Precision' is the extent to which spatial judgements are scattered about their mean position. A darts player is highly accurate, if the throws are evenly distributed about the target, even though the player is very imprecise because the throws are widely scattered. Another player is very inaccurate if the throws are well to one side of the target, even though all the throws land on the same spot.

1. Sensory integration
2. Intersensory coordination
3. Sensorimotor coordination.

1. Sensory integration

As an example of sensory integration, consider the act of estimating the direction of a seen object with reference to the body. Such a task involves estimating where the object is with respect to the eyes, how the eyes are oriented in the head, and how the head is oriented on the body. The visual direction of an object with respect to an eye is indicated by the position of its image on the retina. However, this task is simplified by the fact that we normally direct our gaze towards an object of interest and bring its image onto the centre of the retina (the fovea).

Information about the direction of the eyes in the head is provided by motor signals sent to the muscles that move the eyes, or hold them in a given position. An eye never has to move against a variable load, so that the muscular forces, and hence the motor signals, are always the same for a given position of an eye. There is no need for sense organs to indicate the position of the eyes; their position is always indicated by the sense of effort required to move them or hold them in position.

The direction of the head with respect to the body is indicated by sensory receptors in the muscles and joints of the neck; motor signals are unreliable indicators of head position because the effort required to hold the head in a given position on the body depends on the posture of the head with respect to gravity.

For the total task of judging the visual direction of an object, the information from these three components must be summed, or integrated. Since the eye and the head rotate about approximately the same vertical axis, one would expect the algebraic sum of the angular inaccuracies of the three components to equal the inaccuracy of the total task and the sum of the variabilities (precision) of the component tasks to equal the variability of the total task. This is why this case is referred to here as 'sensory integration'.

A similar state of affairs holds when we judge the direction of a hidden object which we touch with the finger. In this case information from the various joints of the arm is summed in estimating the direction of the object relative to the body. An extra factor is involved in this example because, in addition to summing information about the angular positions of the joints, it is also necessary to know the length of each segment of the limb.

Implicit knowledge about the spatial properties of our own body is known as the 'body schema'. This knowledge seems to be stored in the parietal lobes of the brain; damage to these areas results in anomalous experiences of the body (Critchley 1969). A patient with parietal lobe damage may complain that one of his arms does not belong to him, even though he is able to move it and feel with it, or he may feel that his arm is distorted, or not attached to the body. The body schema for a limb changes as the body changes during growth and persists after the limb has been amputated and this creates the illusion that the limb is still present. An amputee will attempt to use his phantom limb when doing habitual things.

2. Intersensory coordination

In intersensory coordination an object is detected by at least two sense organs and the person is required to coordinate the spatial information derived from these different sources. Consider the act of picking up a small handbell, looking at it, and ringing it. The direction of the bell is sensed by the eyes, by the ears, and by the hand, and yet these separate impressions normally seem to originate from one and the same bell.

Interesting things happen when the spatial information from the various sense organs is not in agreement, that is, when there is a sensory discordance. Such a situation may be induced artificially in several ways. For instance, a person may view a ringing bell through prisms which displace the retinal image to one side. A ventriloquist produces a sensory discordance by moving the lips of his dummy and keeping his own lips still. The same thing happens in the cinema, where the sound that seems to come from the actors actually originates from loudspeakers to one side of the screen. In other words, we misperceive the direction of a sound to make it conform to the direction of a visual object with which the sound is associated — an effect known as ventriloquism or visual dominance. Jackson (1953) did an experiment to determine how far a visual object has to be separated from an associated sound before the discordance becomes noticeable. Subjects reported that a hidden whistle appeared to originate from a silent steaming kettle if the kettle and whistle were separated by less than 30 degrees. A movement of an isolated sound source by this amount is easily detected.

An experiment by Rock and Victor (1964) provides a nice example of the dominance of vision over kinaesthesis. Subjects looked through a lens which caused a square object to appear rectangular. They selected a matching object from among a set of objects they could feel but not see and from a set they could see but not feel. Most subjects selected an object which matched the shape as seen rather than the shape as felt, and few subjects were aware of any conflict.

The conflict between audition and touch kinaesthesis was studied by Pick, Warren, and Hay (1969). Blindfolded subjects pointed with one hand to the felt position of the other hand which touched a loudspeaker that was emitting clicks. At the same time, subjects wore a pseudophone which apparently displaced the clicks by 11 degrees to one side. Subjects pointed to the true position of the other hand and ignored the discordant auditory information.

Thus, in a conflict situation, when the person is convinced that the object detected by one sense organ is the same as that detected by another, vision dominates audition and kinaesthesis, and kinaesthesis dominates audition.

If one wishes to determine how precisely a person can bring stimuli detected by different sense organs into coincidence, one must use stimuli that do not evoke a dominance effect. In the procedure shown in Fig. 1 a light, a small loudspeaker, and a small tactile-kinaesthesis 'button' are each mounted on a boom at arm's length. The subject is presented with pairs of stimuli in various positions and reports which member of each pair is to the left of the other. The subject also judges the position of each stimulus presented on its own. In an ideal system — one which makes best use of the available information — the variability of judgements about the relative positions of two stimuli should equal the sum of variabilities of the judgements about the positions of each stimulus taken separately. Auerbach and Sperling (1974) showed that the performance of human subjects on an auditory–visual localization task, with an apparatus like that shown in Fig. 1, conformed closely to the ideal.

When we direct our gaze towards an object which is straight ahead, it is objectively to the right of the left eye and to the left of the right eye, and yet we experience one object straight ahead. This is because the part of ourselves which we use in making directional judgements is somewhere on a line passing through the bridge of the nose and the centre of the head. This point is known as the visual egocentre. Fig. 2 illustrates a simple procedure for demonstrating that lines which extend out from each eye are perceived to lie in a plane midway between the eyes, which is to say that they are referred to a common egocentre in the median plane of the head.

Stereoscopic vision is a special case of intersensory coordination. In this case objects are seen by the two eyes, but in slightly different directions, because the two eyes are not in the same place. As long as the disparity in the relative positions of the two images is not too large, we experience only one object, but at the same time use the spatial disparity as a clue to the relative distances of objects.

Another interesting case of intersensory coordination is provided by the way we use information from the two ears to judge the direction of sound sources. This is a highly sophisticated mechanism which depends on the detection of relative intensities and times of arrival of sounds at the two ears (see binaural hearing).


Fig. 1. Schematic representation of an apparatus used to measure the accuracy and precision of intersensory localization.



Fig. 2. Procedure for demonstrating that visual directions in the two eyes are referred to a common egocentre. Each line must point accurately to the pupil of an eye, and fixation must be maintained on the point where the two lines meet. When this is done a 'fused' image of the two lines is seen extending towards the bridge of the nose.

3. Sensorimotor coordination

White, Castle, and Held (1964) described the normal development of visual motor coordination. During the first month the child is able to pursue objects with the eyes and head, and by the second month these movements become more refined and show signs of predicting the future position of moving objects. Arm movements are unrelated to vision at this stage. The grasp reflex is present but is wholly under tactile control. Infants under 1 month of age do not attend to objects within arm's reach, probably because of inadequate accommodation and convergence. In the second and third months the infants visually attend to near objects and begin to take a visual interest in their own arms. The first visually directed swiping movements of the arm develop, but the child grasps an object only if the hand touches it. In the third month the swipe gives way to a more directed arm movement, and the child looks back and forth between object and hand. In the third and fourth months, the child watches the two hands as they contact and manipulate each other, thus producing a double feedback experience. In the fifth month, this double arm action comes under visual control and gradually gives rise to the ability to reach rapidly and grasp an object. White (1970) reported that, for infants nurtured in an environment enriched by a variety of objects hanging within reach, the onset of sustained observation of the hand occurred at a mean age of 50 days, rather than at 60 days as in infants reared in a 'normal' environment.

The spatially coordinated behaviour of adult humans can be adjusted to the changing size and shape of parts of the body during growth, to the demands of novel environments, and to compensate for injury. Because of their intelligence, humans have dispensed with narrowly specialized sense organs and limbs and have instead evolved highly flexible mechanisms that reach their highest expression in learned skills. There are several ways of studying the flexibility of spatially coordinated behaviour. One way is to rotate or transplant sense organs or tendon insertions surgically: a method applicable to humans only when radical surgery is required for medical reasons. A second procedure is to study animals reared in anomalous sensory environments, or people with severe sensory deficits, from an early age. Finally, the flexibility of spatially coordinated behaviour may be studied by temporarily distorting the visual input by placing prisms or lenses in front of the eyes. Some representative experiments of each kind will now be described.The effects of surgical rotation of the eyeAs part of a treatment for a detached retina, Barrios, Recalde, and Mendilaharzi (1959) severed each rectus muscle of one eye in several human patients, rotated the eyeball through 90 degrees, and sutured each muscle back onto a stump of tendon which was 90 degrees away from the muscle's normal insertion. The patients were allowed to use both eyes during the six-month recovery period. At the end of this period, when tested using only the rotated eye, the patients reported that the visual scene appeared rotated 90 degrees. Furthermore, pursuit eye movements and visually directed movements made with the unseen hand occurred in a direction at right angles to the movement of the visual targets. The total absence of adaptation in these patients was probably due to suppression in vision in the rotated eye during the recovery period. Human beings are certainly able to compensate behaviourally for the rotation of the visual scene produced by optical means, as the pioneering work of Stratton (1897) demonstrated. When a similar experiment was done on cats, the animals showed accurate visually guided paw placement and obstacle avoidance when seeing with the rotated eye, but only if, during training, the good eye was kept closed. In this experiment, the projection of nerve fibres from the retina of the rotated eye onto the visual cortex was found to be unchanged. The behavioural compensation was obviously due to changes at a higher level.The effects of restricted rearing on visual motor coordinationOne can study the kinds of sensorimotor experience required for the development of visual motor skills by rearing animals in environments which restrict experience in specific ways. In the most famous of these experiments (by Held and Hein 1958) pairs of kittens were reared in darkness, except for a certain period each day when they were placed in an illuminated striped carousel apparatus, as shown in Fig. 3. One kitten of each pair was always placed in the box so that its feet did not touch the ground, and the other kitten was always placed on the other end of the rotating lever so that it could walk and thereby cause itself and its passive partner to be moved round inside the striped drum. Both kittens had the same visual experience, but only for the active kitten was this related to the act of walking. Sensory stimulation which results from self-produced movements is known as reafference, and sensory stimulation which occurs independently of self-produced movements is called exafference. Held and Hein found that only the active kitten developed the ability to avoid a cliff, blink at an approaching object, or extend its paw to a surface. They concluded that reafference is necessary for the development of visual motor skills.

Held, Gower, and Diamond subsequently showed that the paw-placing response developed in immobilized kittens which had experienced only diffuse light. The passive experience in the carousel must have interfered with the maturation of this response, which undermines the claim that reafferent stimulation is necessary for development of visual motor skills.

Held and Hein developed tests for abilities which, they claimed, do not develop without reafferent visual experience. They reared kittens with opaque collars round their necks which prevented them from seeing their limbs (Fig. 4). These kittens could extend their paws towards prongs (Fig. 5) but could not hit them, except by chance, and they could not strike a ball dangled in front of them. Held and Bauer conducted a very similar experiment with monkeys, who for the first 35 days after birth wore a collar which occluded their arms. When the arms were allowed to come into view the monkeys could not reach accurately towards a bottle. It was concluded that 'an infant Primate initially fails to reach accurately for attractive visible objects with a limb that it has never previously viewed'.

Note that the monkeys were not allowed to touch the bottle before the collar was removed, and therefore had not learned to relate seen objects with anything that the unseen arm did. Walk and Bond repeated the experiment but allowed the monkeys to touch one end of a rod, the other end of which they could see projecting above the edge of the collar which occluded the hand. After this exposure, visually guided reaching was tolerably accurate after the collar was removed. Therefore, sight of the hand is not required for the development of visual motor coordination, only some experience that links its motion to a seen object.

Thus, visual motor skills seem to develop in the presence of any type of information which informs the animal about the accuracy of performance. Certain types of early deprivation have a general debilitating effect, but visual motor learning is usually very specific to the conditions under which it occurs.The effects of visual distortions on visual motor coordinationWhen stimuli impinging on one sense organ are spatially distorted with respect to those impinging on the other sense organs, there is a sensory discordance. The types of visual distortion that have been studied include sideways displacement, tilt, inversion, left–right reversal, magnification, and curvature. Anyone with a wedge prism can perform the following simple experiment. A few numbers are marked on the edge of a piece of card which is then placed horizontally under the chin as in Fig. 6. With the prism before one eye and the other eye closed, the finger is directed towards a number on the far side of the card and allowed to come into view. This arm is then returned to the side of the body, after which the aiming movement is repeated several times to each of the numbers in random order. The error in pointing will be very evident for the first few trials, but accuracy is soon restored. When the prism is removed, it will be found that the first few aims will be off target in the opposite direction to the error first experienced when the prism was in place. This after-effect illustrates that adaptation to a visual distortion is not merely a question of deliberate compensation.The nature of the changes underlying adaptation to displaced visionIt is agreed that visual motor adaptation to distorted vision does not involve changes in the sense organs, or in the muscular system that controls arm movements. The change must be in the way sensory or motor signals are coded in the central nervous system. There is no evidence of a change in the internal calibration of the position of the retinal image. This is not surprising because, when a person points to an object, its image falls on the fovea, which is a very distinctive landmark. Other experiments, reviewed in Howard (1981), have failed to reveal any significant effects of visual motor learning on the apparent directions and shapes of objects, even when the objects are not fixated. It seems that simple visual sensations are insulated from events outside vision. There is an old theory, known as the motor theory of perception, in which it is claimed that the way we see is determined by motor behaviour. It would seem that this theory is wrong with respect to simple visual sensations.

Visual motor adaptation has been found sometimes to involve a change in the sense of eye position. For instance, Craske (1967) found that the objectively determined position of the eyes, when the subject was attempting to look straight ahead in the dark, was shifted after exposure to displacing prisms. Similar experiments have demonstrated that there may also be a change in the felt position of the head on the body.

If training one arm with prisms affects the way the other arm points, the adaptation is said to show intermanual transfer. Intermanual transfer indicates that the change must have occurred in the sense of position of the head or the eyes — parts common to both arms. There is general agreement that intermanual transfer is only partial, so that most of the adaptive change to displaced vision must involve a change in the sense of position of the trained arm — a change in the way information from joint receptors is coded, or in the spatial coding of motor commands.

Evidence that motor learning may contribute to adaptation of the visual motor system comes from experiments by Taub and Goldberg on monkeys in whom the sensory roots of the spinal cord had been severed. These monkeys could be trained to reach with the unseen arm towards visual targets and to adapt their pointing when viewing the targets through displacing prisms. This must have been motor learning because the animals lacked sensory inputs from the arm.

Thus it seems that all systems beyond the most peripheral processes in sense organs and muscles are capable of adapting to unusual circumstances, given the correct set of constraints and demands. The one exception seems to be that the central calibration of the position of the retinal image is immutable.The conditions under which adaptation to displaced vision occursHeld proposed that sensations arising from active movement (i.e. reafferent stimuli) are necessary for visual motor adaptation to displaced vision. In one experiment Hein and Held asked subjects to watch one of their own hands through a displacing prism as the hand was waved from side to side, either actively by the subject or passively by the experimenter. Only after the active condition did subjects show evidence of having adapted to the prisms when they were tested in an aiming test. They concluded that self-produced movement coupled to visual reafferent stimulation (sight of the moving arm) is necessary for a change of visual motor coordination. However, subjects probably paid more attention to what they were doing when moving their own arms and, with this factor controlled, other investigators have demonstrated that self-produced movement is not necessary for visual motor adaptation.

Howard and Templeton (1966) argued that adaptation occurs in response to many forms of discordant information: the important thing is that salient information regarding the discordance (not necessarily consciously perceived) should be available to the subject. The sensory consequences of self-produced movement may be a particularly potent source of information, but the most reasonable general conclusion is that any consistent relationship between stimuli within a given sense or between stimuli in different senses, or any relationship between stimuli and responses, will be learned and the system changed accordingly.

The posterior parietal lobe of the cerebral cortex is well placed to serve as a centre for the higher control of coordination of vision, somaesthesis, eye movements, and limb movements. It has been shown that cells in this region of the monkey's brain respond when the animal sees an object of interest, such as food, towards which it is likely to reach. Many of the visually responsive neurons also respond when the eyes move, and some also respond to stimulation of touch receptors.

Subcortical centres in the brain are also involved in sensorimotor coordination. For instance, it has been shown that cells in the basal ganglia, which receive highly processed spatial information, probably from the parietal lobes, are active when the animal is tracking a moving visual object. Damage to the basal ganglia in humans, such as is associated with Parkinsonism, results in akinesia, or an inability to initiate movements towards a visual object. This disorder is manifest only in certain contexts, which suggests that the basal ganglia are concerned with organizing responses to specific visual information. Our knowledge of the neurology of sensorimotor coordination is very fragmentary. Perhaps the development of artificial automata will help to provide theoretical insights which will guide future work.


Fig. 3. Apparatus used by Hein and Held for equating motion and consequent visual feedback for an actively moving animal (A) and a passively moving animal (P).



Fig. 4. Kitten wearing a collar that prevents sight of limbs and torso.



Fig. 5. Apparatus for testing the accuracy of visually guided paw placement in the cat.



Fig. 6. A simple apparatus for demonstrating adaptation of pointing to displaced vision.


(Published 1987)

— Ian P. Howard

Bibliography
• Aldridge, J. W., Anderson, R. J., and Murphy, J. T. (1980). 'The role of the basal ganglia in controlling a movement initiated by a visually presented cue'. Brain Research, 192.
• Auerbach, E., and Sperling, P. (1974). 'A common auditory-visual space: evidence for its reality'. Perception and Psychophysics, 16.
• Bailey, J. S. (1972). 'Arm–body adaptation with passive arm movements'. Perception and Psychophysics, 12.
• Barrios, R. R., Recalde, E. M., and Mendilaharzi, C. (1959). 'Surgical rotation of the eyeball'. British Journal of Psychology, 43.
• Blakemore, C., van Sluyters, R. C., Peck, C. K., and Hein, A. (1975). 'Development of cat visual cortex following rotation of one eye'. Nature. New Biology, 257.
• Craske, B. (1967). 'Adaptation to prisms: changes in internally registered eye position'. British Journal of Psychology, 58.
• Critchley, M. (1969). The Parietal Lobes.
• Fisher, G. H. (1960). 'Intersensory localization in three modalities'. Bulletin of the British Psychological Society, 14.
• Held, R., and Hein, A. (1958). 'Adaptation of disarranged hand–eye coordination contingent upon re-afferent stimulation'. Perceptual Motor Skills, 8.
• Howard, I. P. (1981). Human Visual Orientation.
• — —  and Templeton, W. B. (1966). Human Spatial Orientation.
• Jackson, C. V. (1953). 'Visual factors in auditory localization'. Quarterly Journal of Experimental Psychology, 5.
• Pick, H. L., Warren, D. H., and Hay, J. C. (1969). 'Sensory conflicts in judgements of spatial direction'. Perception and Psychophysics, 6.
• Rock, I., and Victor, J. (1964). 'Vision and touch: an experimentally created conflict between the two senses'. Science,143.
• Stratton, G. M. (1897). 'Upright vision and the retinal image'. Psychological Review, 4.
• Templeton, W. B., Howard, I. P., and Lowman, A. E. (1966). 'Passively generated adaptation to prismatic distortion'. Perceptual Motor Skills, 22.
• White, B. L. (1970). 'Experience and the development of motor mechanisms in infancy'. In Connolly, K. (ed.), Mechanisms of Motor Skill Development.
• — —  Castle, P., and Held, R. (1964). 'Observations on the development of visually-directed reaching'. Child Development, 35.

---