Viktig studie som hinter til den virkelige oppgaven til nervesystemet, som ikke er sende info inn til eller ut fra hjernen… Om vitenskapens forstelse av nervesystemet opp igjennom tidene og hvordan alt vi trodde om nervesystemet er feil.
http://adb.sagepub.com/content/21/2/67.abstract
Hele studien på dropbox: https://dl.dropboxusercontent.com/u/17457302/Forskning%20mappe%20for%20terapi/Keizer%202013%20What%20nervous%20systems%20do-%20early%20evolution%2C%20input%C2%96output%2C%20and%20the%20skin%20brain%20thesis.pdf

We hold that the fundamental problem here was not so much to act intelligentlya problem that had already been solved in various ways without a nervous system (Section 3.3)but to act as a single multicellular unit.

Nervous systems arose as a source and coordinator of patterned activity across extensive areas of contractile tissue in a way that was only loosely constrained by sensor activity.

In this view, the central direction of nervous system connections runs transverseat right anglesto the through-conducting stream that runs between sensors and effectors: early nervous systems evolved as connec- tions across a contractile tissue and in close connection to the animal epithelium or skin.

Adopting the phrase skin brain introduced by Holland (2003), we will refer to this idea as the skin brain thesis, or SBT.

Although the inputoutput view is deeply entrenched, there are issues involving nervous system functioning that are highly puzzling or awkward when the input output view is taken as a fundamental account of ner- vous systems.

The current inputoutput interpretation of nervous sys- tems is closely linked to a computational information- processing interpretation. This linkage is intrinsic to the classic neuron doctrine, according to which neurons are individual entities that receive and send electrical signals to one another through synapses in an all-or-none fashion that is basically similar to electrical switches. Consistent with the neuron doctrines one-way flow of information, nervous systems could be interpreted as electronic circui- try, which may be far more complex than artificial circui- try, but not intrinsically different.

The problem with this input-output interpretation is that the neuron doctrine on which it is based has been seriously undermined (e.g., Bullock et al., 2005; Guillery, 2007; Kruger & Otis, 2007) since it was first advanced by Ramon y Cajal in the late 19th century. Famously, Cajal formulated what came to be called the neuron doctrine explicitly in opposition to the then-current idea that ner- vous systems are reticular organizations of nerve cells directly connected to one another, through which electri- cal activity flows diffusely in all directions (Guillery, 2007; Kruger & Otis, 2007).

The neu- ron doctrine can not plausibly explain the diversity of neuromodulatory substances, such as amines and neu- ropeptides, that remodel neuron behavior and circuitry within minutes and hours instead of the standard milli- second time scale (Bullock et al., 2005). Many of these neuromodulatory molecules are not recent evolutionary developments but have a deep genomic history. More recently, immune system elements, such as cytokines, have been shown to play critical roles in modulating neural plasticity under normal as well as challenged conditions (McAfoose and Baune, 2009; Yirmiya and Goshen, 2011), and these associations are also very old (Maier and Watkins, 1998). The neuron doctrine cannot explain these associations either. Moreover, in many neurons, action potentials can travel backward from the axon and cell body to the dendrites.

Clue 2: The detailed operation of neurons and nervous sys- tems is much more complex and diverse than can be readily accounted for by the inputoutput view.

Clue 3: The reflex arc organization may very well be a sec- ondary optimization of nervous systems.

The inputoutput interpretation stresses that nervous systems function as information processing devices. However, in recent years serious claims concerning the complexity, and even cognitive, nature of the behavior of single-celled organisms have come to the fore. For example, John Allman (1999) discusses how the most fundamental features of brains such as sensory integra- tion, memory, decision-making, and the control of behavior, can already be found in simple organisms such as bacteria (pp. 56).

While this is presumably true of complex ner- vous systems, the point does not seem to apply to basic forms. When one systematically compares organisms with basic nervous systems, they do not show more complex behavior than creatures without a nervous sys- tem.

According to Jennings, the possession of a nervous system brings with it no observable essential changes in the nature of behavior. We have found no important additional features in the behavior when the nervous system is added (p. 263).

Clue 4: Basic nervous systems do not lead to more complex behavior than is often present in organisms without a nervous system.

Clue 5: Many of the biomolecular characteristics of neurons are already present in non-neural precursor contexts.

Clue 6: Understanding what nervous systems do is a question that requires an answer at the level of the whole animal.

Clue 7: The main animal effector consists of muscle tissue that requires spatiotemporal coordination.

Clue 8: Coordinating extensive areas of muscle tissue requires endogenous activity.

Nowadays, the picture has changed again. While Mackies scenario for the origins of nervous systems is still influential (e.g., Arendt, 2008; Je kely, 2011; Miller, 2009), it faces important difficulties. A key problem is that nervous systems are found more widely among animal phyla and classes than electri- cally coupled conductive epithelia. Notably, while all four major cnidarian classes have a nervous system, there is substantial evidence that only the Hydrozoa have functional gap junctions (Mackie, Anderson, & Singla, 1984; Satterlie, 2011).

Clue 9: Chemical transmission between adjacent cells can have provided the basis for primitive conductive epithelia that formed a half-way station to nerve nets.

Clue 10: Chemically transmitting conductive (myo)epithelia can have provided a basic form of muscle coordination.

Clue 11: Specialized axodendritic connections can have sub- sequently evolved to broaden the existing possibilities for muscle coordination.

Under this interpretation, the core business of such nerve nets consisted of organizing and integrating activity across contractile effector surfaces (e.g., mus- cle) spread out beneath an external epithelium. Such a task would involve parallel organization and coordina- tion requiring signaling across a surface rather than a through-conducting, sequential organization based on a set of pre-existing sensors and effectors. No stimulus can specify by itself the behaviorally relevant contrac- tion patterns across such a surface. Patterns that workthat is, patterns that lead to movements that are appropriate under the circumstancesare a func- tion of the particular effector surface that is present in the animals rather than of any triggering stimulus. Also, based on what we know about organisms today, movement is likely to have been self-induced, while external stimuli acted rather as modulating factors on continuous effector activity.

While modern nervous sys- tems have various other functions, it is evident that enabling an organism to move and manipulate its envi- ronment in specific ways is the prime reason for the huge investment in these metabolically expensive organs (Allman, 1999).

Such cellular con- tractions must be coordinated with respect to one another, however. Uncoordinated contractions by indi- vidual cells would not result in whole-body motility. This, we believe, is where nervous systems come in. Nerve nets are intrinsically tied up with muscle surfaces.

The SBT can now be formulated as the proposition that early nerve nets evolved when some conducting cellseither within or connected to the myoepithelium evolved elongated processes and synaptic connections in a way that modified and enhanced the patterning capabil- ities of a pre-existing myoepithelium. Rather than pro- viding specific connections from sensors to effectors, the proper function of such nerve nets was to control, modify and extend the available self-organized pattern- ing across a Pantin surface. The key adaptation pro- vided by early nerve nets was the way in which they added to the generic self-organizing properties of pre- existing epithelial and muscular tissues.

To summarize, the SBT claims that nerve nets origi- nated as a new mechanism by which Pantin surfaces could be more intricately and flexibly patterned to accommodate efficient motility at larger bodily scales. At a fundamental level nerve nets are fitted to spatial patterning and to accommodating spatially patterned feedback.

The SBT offers a genuinely new conceptual approach for understanding nervous systems at a whole systems level. Starting with the most primitive neural organiza- tionsproto-neural myoepithelia and nerve netswe argue that both are characterized by connections trans- verse to the standard sensor-effector direction and evolved their characteristics to bind the many cellular units of muscle sheets together into a unitary system. Nervous systems are in this view not organized aroundor rather betweensensors and effectors. They are themselves a precondition for both extended con- tractile effectors as well as multicellular sensory arrays.

We have stressed from the beginning that the SBT provides a conceptual reinterpretation of nervous system functioning.

The skin brain pro- posal casts animal behavior as a dynamical phenotype, necessarily tied to the species or class of animals under consideration. Sherrington once observed that posture follows movement like a shadow (Stuart, 2005). We would like to stress that dynamically changing body pos- ture is a precondition for all task-oriented animal beha- vior. Animal behavior is a part of animal organization.