Like all animals, and different from plants and dead matter, human beings are capable of active and purposive movement in their environment. This implies not only the ability to move oneself but also the ability to sense this self-motion. For it is because we can sense our motion, that we can adapt it to our needs and give it purpose, i.e. give it a certain direction, modulate the velocity, avoid bumping into things or other beings, etc. This feedback about self-motion results from an interaction of output from almost every sensory system we have: the visual, proprioceptive, tactile and auditory systems as well as the vestibular system. And apart from the sensory systems, cognitive factors such as mental set or expectation may also influence one's sense of self motion (Guedry, 1974).

The vestibular system is a small apparatus ill the inner ear, that is sensitive to accelerations. Thus it informs our brain whether we make active head movements or, are being passively accelerated (for instance in a bus or train). It is a rather hidden sense organ, one we are not aware of, ill the way we are aware of being able to see and to hear. Its presence usually reveals itself under somewhat extreme circumstances, like after a ride oil a roller coaster, on a ship in stormy weather or after having drunk a couple of jenevers too much. The reason we don't notice it under normal circumstances is because it then does what it is supposed to do, it helps us keep our equilibrium and it gives us feedback about our self-motion. It consists of two parts: the semi-circular canals and the otoliths. Whereas the semi-circular canals are sensitive to angular accelerations, the otoliths are sensitive to linear accelerations. Ill order to measure the contribution of the vestibular system to perceived self-motion, one has to somehow prevent all the other contributions. Tills is possible to some extent. Eyes can be blindfolded, ears call be plugged. To prevent motion information from the legs one may be moved passively by some motor driven device for instance. But, especially in the case of the otolith system, it is difficult, if not impossible, to switch off all auxiliary cues to the perceptual system. For when the body is accelerated linearly, so as to stimulate the otolith system, pressure cues oil the body as a result from the inertial force, are inevitable. Therefore it should be kept in mind that whenever the otolith response to linear horizontal oscillation is investigated, ill reality both the otolith and the proprioceptive systems are researched together. Apart from the problem of masking all other sensory systems, there is the simultaneously arising question of whether the vestibular system doesn't behave differently because all tile other sensory inputs are masked. This is all important issue when the vestibular response is measured in order to establish its role in the process of self motion perception. Connected to the latter question is the problem of mental set, or expectation. In the laboratory people that participate in psychophysical experiments are often asked to behave in a manner that isn't very natural. They may as a result emphasize or pay special attention to sensations they don't usually pay attention to. Also their expectation of what they are going to sense may influence their percepts. Most investigators seem to agree on the fact that mental set or expectation influence perceived self motion (See e.g. Guedry, 1974) and it has been suggested that they should be implemented as a part of the 'Internal model' of system analytical models about self motion perception (Helm et al. 1980).

The idea for the psychophysical method that is presented ill part 1 of this thesis arose as a result of a 'mental set' problem. For if, ill ail experimental situation, one is asked to attend to ones motion sensation, the fact that one attends to it may influence it. In every day life one isn't usually paying special attention to self motion. Instead, estimates of velocity or displacement are, without one being aware of it, used, in order to perform other tasks, i.e. like checking the speed of a strain of traffic before merging with it. Therefore it seems more natural to measure perceived self-motion in the process of performing a task for which the brain needs ail estimate of perceived self-motion. It was hypothesized that we might use percepts of object motion, to indirectly measure percepts of vestibular self-motion (see Wertheim 1994). That way eyes would not have to be blindfolded and attention would not be directed inwardly, towards ones owl] motion, but towards that of the external object.

Underlying this psychophysical method is the assumption that the bran] uses an estimate of self-motion as a reference, to decide whether ail attended object is moving earth-relatively or not. This assumption is easy to understand, if you imagine yourself to be in a rocket to the moon, viewing an Unidentified Flying Object (UFO) through the window. Besides you, the rocket and the UFO there is nothing but empty space. Now if the UFO doesn't move relative to the window, does this mean it doesn't move in space? No, of course riot, for that depends oil whether the rocket is moving in space or not: if it is, the UFO is as well, if it is riot, neither is tile UFO. The same is true for tile situation in which the UFO does move relative to the window of your rocket: if your rocket is stationary, then the UFO is moving, but if your rocket moves, with the same speed as, but in the direction opposite to, the UFO, then the UFO is stationary. Thus, it seems that in order to decide whether tile UFO moves ill space or not, you have to know the velocity with which your rocket moves in space.

Now suppose that you think that the UFO is stationary, but you don't know the speed of your rocket. You may then discover the speed of your rocket, by measuring the speed of the UFO relative to your rocket's window. If you replace the rocket's window by your eyes and the UFO by any object, the idea for the method presented in part I is born (see also Wertheim, 1994). Whenever you perceive an object as stationary in space, the velocity of this object relative to the eyes, may equal the brain's estimate of the velocity of the eyes in space, i.e. the brains estimate of self-motion. Thus, by measuring the velocity that the object has when it is perceived as stationary in space, one indirectly measures the perceived velocity of self-motion.

In the first chapters, this method is tried for linear horizontal (otolith) oscillation (chapter 1), horizontal canal stimulation (chapter 2), and horizontal smooth pursuit (chapter 2). In chapter 3 an unexpected visual illusion is reported that may have consequences regarding the validity of the method. These consequences, and those of some additional control experiments are discussed in chapter 4.

After having performed quite a number of experiments using the Wertheim (1994) method, the doubts and frustrations about it hid grown to such proportions that it was decided to try a different approach. The conviction had developed that it is impossible to measure the perceived vestibular response to horizontal self motion, and especially the perceived otolith response, without interference of other uncontrollable, visual or mental interferences. Therefore it seemed sensible to investigate the perceived otolith response whilst it interacts with those other self motion inputs, instead of trying to eliminate them. As a result in part II of this thesis the perceived otolith response during linear horizontal accelerations is studied whilst it interacts with mental set and with the visual self motion perception system.

Apart from the fact that, in the investigation of the otolith response during linear horizontal self motion perception, the proprioceptive system can't be turned off, there is another complicating factor. The stimulus to the otolith and proprioceptive systems, is a change in the inertial force vector, which may be caused in two different ways. It may be caused by a change in direction with respect to gravity, and it may be caused by a linear acceleration. For, as Einstein stated in his equivalence principle: " A gravitational field of force at any point in space is in every way equivalent to an artificial field of force resulting from acceleration, and no experiment can possibly distinguish between them. " (White, 1940, p.678). Yet it seems, that in our every day life, we are capable to distinguish between these two causes of otolith sensation. Apart from the fact that we have our visual and muscular (neck, joints) systems to clarify the cause of inertial stimulation, we also have, besides ail otolith system, a semi-circular canal system which senses angular acceleration of the head. When we change the direction of our head relative to gravity not only the otoliths are stimulated but the canals as well, whereas during linear acceleration, only the otoliths react. Finally, our capability to distinguish gravity from Linear acceleration is also ascribed to the low-pass characteristics of the vestibular system, causing tilt outputs for low frequency stimuli (gravity) and linear displacement outputs for higher frequencies (Mayne, 1974).

With respect to this Issue it should be noted however that, different from other animals, human beings have, in the past century, become regular passengers and drivers of all kinds of vehicles. The experiments presented in this thesis all involve passive selfmotion. The sense of rather high linear accelerations and decelerations unaccompanied by stimulation of the leg muscles and joints, Is something we have become quite used to. But such sensations are different from the active human motion patterns with which the vestibular system has evolved. By the loss of muscle and joint information the selfmotion perception system is in a way, unpaired. A symptom of this injury is motion sickness, one of the side effects that has appeared with our ability to propel ourselves with vehicles. How then does the human perceptual system cope with all this 'passively' moving about? Somehow the system compensates the loss of muscle and joint information.

It seems fair to assume, that human cognition, in the form of past experience, knowledge of the possibilities of the vehicle and, related to thus, expectation and mental set, become extra important to the self-motion perception system in order to function optimally in these 'unnatural' situations. Expectation, in the shape of 'the subjective vertical' for instance is assumed to be involved in the developing of motion sickness (Bles and Bos, 1994). It is not unthinkable that expectation also assists tile otolith system in distinguishing changes in the direction of gravity from linear accelerations. The people that participate in experiments, for instance, usually know what kind of passive motion they will undergo, they can see what sort of device they enter before the start of the experiment, and accordingly they form expectations. They know beforehand whether their chair can turn or not and whether the rail along which it moves is horizontal or not. What if they do not know such things beforehand? Chapter 5 is concerned with this question.

Related to the role of expectation or mental set, in the perception of self motion, is the role of vision. For unlike the vestibular and proprioceptive system, the visual system is capable of 'looking ahead'. We can see that we are approaching a bend in the road or a hill, and thus we expect to sense it a certain amount of time later. Does this mean that the brain tends to attach more value to what it sees than to what it senses vestibularly? That is, what happens to the overall percept of self-motion when input to two systems does not coincide in terms of phase or amplitude? This question was investigated in chapter 6, in terms of the veridicality of perceived self-motion and the perceived earth relative stability of the visual environment.

• Bos J. E. and Bles W. (1997) A vestibular based motion sickness incidence model. In press.
• Guedry F. E. (1974) psychophysics of vestibular sensation. In: Handbook of Sensory Physiology (Kornhuber H. H. ed.) Springer Verlag Berlin, New York, Vol. VI/2, 3-154.
• Henn V., Cohen B. and Young L R. (1980) Visual-vestibular interaction in motion perception and the generation of nystagmus. Neurosciences Research Program Bulletin, published by the MIT press, ISSN 0028-3967, Vol. 18, no. 4.
• Mayne R. (1974) A systems concept of tile vestibular organs. In: Handbook of Sensory Physiology (Kornhuber H. H, ed) Springer Verlag Berlin, Vol. VI/2, 493580,
• Wertheim (1994) motion perception during self-motion: The direct versus inferential controversy revisited, Behavioral and Brain Sciences 17, 293-355.
• White H. E. (1940) Classical and Modern Physics, New York: Doran Nostrand Co., Inc.