Reafferent principle

from Wikipedia, the free encyclopedia

The principle of reactivity is a model in the field of movement control - the movements of living things. It is a control principle that enables the central nervous system to block out expected stimuli. This principle can be used to explain, for example, why the environment is perceived as immobile when the eye moves , although the movements of the environment are depicted on the retina .

The principle of reactivity was discovered, named and described by Erich von Holst (together with Horst Mittelstaedt ) around the middle of the 20th century and first published in his eponymous article in the journal Naturwissenschaften .

The principle of reactivity

description

The basic idea of ​​the reactivity principle is that a copy ( efference copy , EK) is stored (+) of the commands ( efferents ) to the execution systems (for example the muscles) in the lowest, but still central, center of the nervous system (Z 1 ) . The commands are then passed on to the effectors (EFF) (E) and executed there. From the effectors (reafferences) and the environment (exafferences), reports of execution (A, - ) are passed back to the lowest central center and offset against the stored efference copy. If the efference copy and the feedback are the same - there is no remainder - the process is complete.

However, if there is a remainder, this is reported to the next higher central center, and an attempt is made from there to delete the remnants of the efference copy by means of new commands that trigger new reafferences and exaffences . If this does not succeed, the next higher center is tried (M). This process is carried out until the efferent copy is deleted . If the "error message" reaches the highest center in the cortex (Z n ), we become aware of the process of the necessary correction.

An important consequence of the reactivity principle is that the organism can distinguish between signals that are caused by its own movement (proprioceptive signals) and the signals that originate as exafferations from the environment.

2 examples (reported by von Holst in his article on the principle of reactivity): The classic example of von Holst for the principle of reactivity is that of the fly Eristalis in the stripe cylinder:

If you put a fly in a cylinder that has vertical light-dark stripes inside and, while the fly is sitting quietly in the cylinder, slowly turns the cylinder, the fly begins to turn its body in the same direction as it is the cylinder also rotates. This happens because of the optokinetic reflex, in which the fly tries to capture its image because it does not move of its own accord . (It is assumed that when the fly moves spontaneously, this reflex is inhibited, meaning that it can move freely). If you turn her head by 180 ° in the longitudinal axis of her body and stick it to her chest, she perceives a rotation of the cylinder in the opposite direction to the direction in which it actually takes place. If it tries again - following the optokinetic reflex - to hold the image, it moves in the opposite direction to the rotation of the cylinder. If the environment were now neutral, the optokinetic reflex would be inhibited and the fly could move freely. In this structured environment, however, the fly begins to turn in the opposite direction, then back again in ever tighter circles until it stops as if frozen. When the head is brought back into its natural position with the eyes, the fly moves as before when the stripe cylinder is turned.

Von Holst interprets this in the sense that he assumes that the fly expects a certain shift in the field of vision on the retina when it moves spontaneously , which when it occurs during the movement is somehow neutralized. This happens with a natural head position when the cylinder is rotated. If the eyes are swapped , however, the retinal shift does not correspond to the expectation - the movement commands cannot be neutralized. The fly then changes its direction of rotation in order to achieve the expected visual impressions. Again, that doesn't happen. The fly changes its direction of rotation and so on until it freezes, exhausted.

Another example reported by Holst is this: Centipedes are known to crawl around in regular circles to the intact side when one half of their brain is removed. If you then cut off a few segments from the back of their body, the more segments cut off, the narrower the circles. This can be explained with the principle of reafference. Because the efferent signals for the "normal" circle are stored in the efferent copy. The feedback on the actual movement can only come from the segments that are still available. Therefore, remnants of the efferent copy remain. In order to ensure that these are also deleted, the worm has to contract the existing segments until the response copy has been processed . This causes the worm to crawl in a tighter circle.

Holst, describes in his article also how the in psychology (Gestalt and Cognitive Psychology ) significant constancy phenomena (eg room or size constancy) using can explain the Reafferenzprinzips.

General version

Generally speaking, centers are:

From the efference E for a movement sequence that originates from a superordinate nervous center Z n , an efference copy EK is created in certain subordinate centers .

In interaction with this and the reactivity A of the effector EFF , we perceive a movement success. This allows movement sequences that are influenced by other superordinate centers or from outside to be controlled and regulated.

The efference is the target state, the position of the effector of the actual state, a difference between both the Exafferenz M .

Origin and development of the theory of the principle of reafference

Reflexes, reflex chains, reflexology

Scientific research into the nervous system began at the beginning of the 20th century. Two of the outstanding researchers were Charles Scott Sherrington (1857–1952) and Iwan Petrovich Pawlow (1849–1936). Both dealt with the reflexes. But while Sherrington mainly examined the anatomy of the individual nerve cells and their function, Pavlov developed important models of the reflexes (stimulus-response theory).

Sherrington, for example, discovered that the nervous system is not a coherent structure (syncytium), but rather that it consists of individual nerve cells that are physiologically separated from one another, but can exchange signals with one another at certain points.

Pavlov developed, among other things, the theory of the conditioned reflex (see Pavlov's dog ), which also became the basis of behaviorism . Together with Vladimir Mikhailovich Bechterew (1857-1927) he developed mechanistically oriented psychology. This describes the behavior as a sequence of reflexes (stimulus-reaction scheme: the reaction (muscle contraction) to a stimulus is the triggering stimulus (stimulus) for the next muscle contraction). More complex, differentiated (e.g. instinctive) behaviors are viewed by them as reflex chains and as the basis of learned behavior (see learning theories - conditioning ). These reflex chains formed a very rigid system - imaginable like a coin-operated machine - through which movements are triggered and carried out. The individual elements of complex movements are then completely dependent on one another. (This theory is also known as reflexology ).

For a long time it was imagined that such reflex chains would create complex movements and also be controlled, for example the crawling of a worm or the flapping of the wings of a bird. The movement of humans was also imagined (in behaviorism) to come about in this way.

Development of the principle of reactivity

However, there were also doubts about this very rigid interpretation of the origin of movement and control, because in many animals, disturbances in their normal mode of locomotion resulted in other, but thoroughly controlled, movement sequences. One of the researchers who investigated these phenomena was the German physiologist Erich von Holst (1908–1962), who did his doctorate in 1932 on the function of the central nervous system in earthworms . In his investigations into the locomotion of living things - worms, fish, flies and others - he observed that their limbs (for example their fins) not only move very evenly in the same rhythm, but sometimes also completely independently of one another. He referred to the latter as relative coordination, as opposed to even movement ( absolute coordination ) . After further systematic investigations and reflections on these phenomena, he developed the model of the principle of reactivity.

He was aware that this model fit into the consciousness of biologists and physiologists, which was developing at the time, that numerous regulatory processes were taking place in the organism in the function of living beings. He only mentions this in passing, possibly because the reactivity principle is not a simple feedback system, but a feedforward system. These were not often described at the time.

Slow spread

In 1950 his reflections were published in the essay The Reafferenzprinzip in the journal Naturwissenschaften , which he wrote together with Horst Mittelstaedt. Four years later, an English short version by him about the principle of reactivity appears in the British Journal of Animal Behavior.

In the following years, the reactivity principle was occasionally mentioned in connection with the control loops in the organism for movements (for example in Volume 14 Sensomotorik of the series Physiologie des Menschen , published in 1976 by OH Gauer, K. Kramer and R. Jung). However, its full importance was not recognized for a long time. In the English / American discussion about human motor skills, the principle of reactivity hardly played a role for a long time - there was no complete translation and therefore too little was known. It has often been equated with the similar concept of corollary discharge .

It was not until the 80s of the last century, when the importance of motor skills research grew and there was more international collaboration, that some studies on the principle of reactivity could be found. For example with Charles Gallistel 1980. Heidrun Schewe also described the reactivity principle and its significance for movement and movement learning in 1988 in her book The Movement of Man . But especially around the turn of the millennium, when mathematicians became more involved, the principle of reafference gained new importance.

Neuroscience and the principle of reactivity

Neuroscience had also developed and it was known that there are different types of control of processes in the living organism. Above all Daniel M. Wolpert (motorist and mathematician; Cambridge, England,) and his group are concerned with the verification of von Holst's ideas and his further research. For example, in 1995 he showed in an article that the reactivity principle can be described using the mode of action of a Kalman filter . From this he develops the idea, which also emerges from von Holt's description, that there are several control centers in the organism. First of all, a forward model works in this imagination, which calculates the flow of the process by predicting the next state from the current state and the motor commands that are stored in the efficiency copy, and an inverse model, which is based on the current Changes in state the motor commands are estimated and compared with the efference copy. With the forward model for the internal feedback, the result of the motor commands can be compared and corrected before the sensory feedback can arrive. The authors assume that there are a large number of such control pairs in the organism. This leads to a modular structure of movement control and is of great importance for movement learning .

The ability to distinguish between proprioceptive and exafferent signals can lead to self-perception being dampened. This means that signals coming from outside can be perceived more precisely if required.

Of great interest to the researchers was also where in the brain the activities take place that make these processes possible. This was made possible by the newer imaging methods such as positron emission tomography (PET) or functional magnetic resonance tomography (fMRI).

It has been known for a long time that an intended movement sequence is reported from the association cortex to the motor cortex, which can activate the muscles directly (cortico-spinal tract), for example, so that they perform the required movements ( motor command ). The proprioceptive sensors in the muscles and tendons report the results back to the central nervous system (CNS). However, this path (transcortical loop) takes too long (40–60 ms) to be able to regulate a movement sequence during its course . To enable faster control , the cerebellum (spinocerebellum: vermis and middle part of the hemispheres) and the magnocellular part of the red nucleus ( nucleus ruber ) receive the efference copy of the motor commands (feedforward). In these parts of the brain, a model of the muscle system was built (through learning) . This model is able to make a rough prediction of the actual process from the information it receives. The resulting possible error in relation to the actual implementation is transmitted via the rubrospinal tract to both the motor cortex and the muscles. This path is faster than the transcortical loop at 10-20 ms. ( Forward model )

The inverse model is formed by the cerebrocerebellum (lateral parts of the cerebellar hemispheres ) and the small-cell part of the nucleus ruber . These parts of the brain do not receive information from the periphery of the body, but from many parts of the cortex. This enables them to observe and compare both the intended sequence of movements and that predicted by the forward model . A discrepancy between these two can be reduced due to the plasticity of the nervous system (learning, experience), so that the following commands issued by the forward model can lead to better execution. The model is called inverse because, in contrast to the forward model, which has the intended sequence of movements as input and the corresponding motor commands as output, in the inverse model the sequence of movements output by the commands serves as input, the improved commands as output.

This description dates back to 1987. In the meantime, knowledge of movement control has developed further - while maintaining the notion of a forward and an inverse model.

The development is determined by the knowledge that it is not individual structures that are responsible for executing and controlling the movements, that rather the complex performance of a movement sequence comes about through the participation and networking of many parts of the brain.

Further examples

The concept of efference copy is also suitable to explain further effects:

  • Most people cannot tickle themselves because the body expects their own hand to be touched.
  • Animals with active sensors (e.g. bat echolocation ) have the problem that their sensors can actually only perceive their own signal when it is emitted. However, the reactivity principle makes it possible to “subtract” the expected response from the sensors from the efference copy and thus only perceive the actual response from the environment. Curtis Bell demonstrated this very impressively in 1982 with the electrical perception of the elephant-trunk fish .
  • Crickets can “hide” their own singing neuronally from their own perception.

literature

  • Erich von Holst, Horst Mittelstaedt: The Reafferenzprinzip. In: Natural Sciences. 37 (1950) pp. 464-476.
  • Erich von Holst: Relation between the central nervous system and the peripheral organs. In: British Journal of animal behavior. 2, 1954, pp. 89-94.
  • David W. Franklin, Daniel M. Wolpert: Computational Mechanisms of Sensorimotor Control. In: Neuron. 72 (2011) pp. 425-442.
  • Erich von Holst: On behavioral physiology in animals and humans. collected treatises. Volumes I and II. Piper & Co Verlag, Munich 1969.
  • R. Jung : Introduction to movement physiology. In: OH Gauer, K. Kramer, R. Jung (Hrsg.): Physiologie des Menschen. Volume 14: Sensorimotor Skills. Urban & Schwarzenberg Verlag, Munich 1976, pp. 1-98.
  • Heidrun H. Schewe: The human movement. Thieme Verlag, Stuttgart 1988, pp. 147-150.
  • Charles R. Gallistel: The Organization of Action: a new Synthesis. Erlbaum, Hilsdale 1980.
  • Richard A. Schmidt. Motor Control and Learning, A behavioral Emphasis . Human Kinetics Publishers, Champaign Illinois 1982.
  • Daniel M. Wolpert, Zoubin Gharahmani, Michael I. Jordan: An Internal Model for Sensorimotor Integration. In: Science. (269) 1995, pp. 1880-1882.
  • Daniel M. Wolpert, M. Kawato. Multiple paired forward and inverse models for motor control. In: Neural Networks. 11 (1998), pp. 1317-1329.
  • Sukhwinder S. Shergill, Thomas P. White, Daniel W. Joyce, Paul M. Bays, Daniel M. Wolpert, Chris D. Frith: Modulation of somatosensory processing by action. In: NeuroImage. 70 (2013) pp. 356-362.

Individual evidence

  1. ↑ Natural Sciences. 37, (1950) pp. 464-476.
  2. ↑ Erich von Holst: On behavioral physiology in animals and humans. collected treatises. Volume I and II, Piper & Co Verlag, Munich 1969, Volume IS 37f
  3. ↑ Erich von Holst: On behavioral physiology in animals and humans. collected treatises. Volume I and II, Piper & Co Verlag, Munich 1969, Volume IS 160.
  4. ↑ Erich von Holst: On behavioral physiology in animals and humans. Collected Treatises. Volume I, Piper & Co Verlag, Munich 1969, pp. 1-132.
  5. ↑ Erich von Holst, Horst Mittelstaedt: The Reafferenzprinzip. In: Natural Sciences. 37 (1950) pp. 464-476.
  6. ^ Erich von Holst: Relation between the central nervous system and the peripheral organs. In: British Journal of Animal Behavior. 2, 1954, pp. 89-94.
  7. ↑ R. Jung: Introduction to movement physiology. In: OH Gauer, K. Kramer, R. Jung (Hrsg.): Physiologie des Menschen. Volume 14: Sensorimotor Skills. Urban & Schwarzenberg Verlag, Munich 1976, pp. 1-98.
  8. ^ For example in Richard A. Schmidt: Motor Control and Learning, A behavioral Emphasis . Human Kinetics Publishers, Champaign Illinois 1982, pp. 227-235.
  9. ^ Charles R. Gallistel: The Organization of Action: a new Synthesis. Erlbaum, Hilsdale 1980 (he has an almost complete bibliography of von Holst's work)
  10. ↑ Heidrun H. Schewe: The movement of people. Thieme Verlag, Stuttgart 1988, pp. 147-150.
  11. ↑ for example: David W. Franklin, Daniel M. Wolpert: Computational Mechanisms of Sensorimotor Control. In: Neuron. 72 2011, pp. 430/431.
  12. ↑ Daniel M. Wolpert, Zoubin Gharahmani, Michael I. Jordan: An Internal Model for Sensorimotor Integration. In: Science. (269) 1995, pp. 1880-1882.
  13. ^ Daniel M. Wolpert, M. Kawato: Multiple paired forward and inverse models for motor control. In: Neural Networks. 11 (1998), pp. 1317-1329.
  14. Jump up ↑ Sukhwinder S. Shergill, Thomas P. White, Daniel W. Joyce, Paul M. Bays, Daniel M. Wolpert, Chris D. Frith: Modulation of somatosensory processing by action. In: NeuroImage. 70 (2013) pp. 356-362.
  15. ↑ M. Kawato, Kazunori Furukawa, R. Suzuki; A Hierarchical Neural Network Model for Control and Learning of Voluntary Movement . In: Biological Cybernetics 57, 1987. pp. 169-185.
  16. ^ Daniel Wolpert, Chris Frith, Sarah-Jayne Blakemore: Why can't you tickle yourself? In: NeuroReport . tape 11 , no. August 11 , 2000 ( online [PDF; accessed February 18, 2012]).
  17. ^ Curtis Bell: Properties of a modifiable efference copy in an electric fish. In: J Neurophysiol . 47: 1043-1056 (1982).
  18. ^ Poulet & Hedwig: A corollary discharge maintains auditory sensitivity during sound production. In: Nature . 418 (2002), pp. 872-876.