|J. A. Scott Kelso (2008), Scholarpedia, 3(10):1611.||doi:10.4249/scholarpedia.1611||revision #50371 [link to/cite this article]|
John von Neumann, the father of the digital computer, once remarked that the problem of understanding the central nervous system is not how it effects one particular thing, but rather how it does all the things that it does in their full complexity. What, he asked, are the principles of organization in complex biological systems that contain enormous numbers of different components? Notwithstanding a nervous system that contains ~10 12 neurons and neuronal connections, the human body alone consists of over 790 muscles and 100 joints. Thus, any ordinary human activity requires the cooperation among very many structurally diverse elements. An attractive hypothesis is that in such complex living systems the elements are organized into synergies (also known as coordinative structures) defined as functional groupings of structural elements (e.g. neurons, muscles, joints) that are temporarily constrained to act as a single coherent unit. Just as new states of matter form under certain conditions when a group of atoms behaves as a single particle (the Bose-Einstein condensate) so a new state of biological function emerges when ensembles of different elements form a synergy. The synergy hypothesis (Kelso, 1995; in press) is therefore a hypothesis about Nature’s way to handle biological complexity. Synergies appear in many contexts on many levels of biological organization, from the genetic to the social.
Although the concept of synergy has a long and cherished history in physiology and medicine (e.g., Bernstein, 1967; Edelman, 1987; Easton, 1972; Gelfand, et al., 1971; Gelfand & Latash, 1998; Kelso, 1995; Latash & Turvey, 1996; Sherrington, 1906; Turvey, 1977), the significance of synergies has been underappreciated for a number of reasons. These include: their various and uneven interpretation and meaning in different fields; their unfortunate association with disavowed notions of “wholism” and “vitalism”; the lack of direct scientific evidence for synergies; and the absence of scientific explanations for how they come about. A further complicating feature is that synergies, though implemented by biological structures, are not essentially defined by their anatomy. When potentially independent elements come together to form a synergy, the cooperativity that emerges is dictated primarily by functional, not (or not only) anatomical considerations (Turvey, 1977). Depending on context, synergies may accomplish different functions using some of the same anatomical components (e.g., speaking and chewing) and the same function using different components (e.g. writing one’s name with a pen attached to the big toe). Another way of saying this is that biological systems exhibit tremendous degeneracy at all levels of observation (Edelman & Gally, 2001).
Evidence for synergies
The hallmark of a synergy is that during the course of ordinary activity a perturbation to any part of the synergy is immediately compensated for by remotely linked elements in such a way as to preserve the functional integrity of the system. Thus, natural variations (“errors”) that occur in the synergy’s components are compensated by adjustments (“covariations”) in other members of the synergy in such a way as to maintain a given function stable. A further property of synergies is that the relations between interacting components are preserved stably in time despite quantitative variation in measures of component parts (Kelso, et al., 1979). Again, such variations may arise naturally from the needs of the organism and the demands of the environment or be induced experimentally through parametric manipulations.
Some of the most direct evidence of synergies comes from seemingly quite unrelated sources: speech production and posture. This is as it should be, since in principle synergies are a ubiquitous feature of complexity at all levels of biological organization. In the case of human speech, a difficult system to study because it is not easily accessible to measurement, Kelso et al (1982; 1984) showed that an unexpected load applied to the jaw during upward motion for the final /b/ closure in the syllable “bab” produced near immediate changes (15-30 thousandths of a second) in upper and lower lip muscles. Since the upper lip is not mechanically linked to the jaw, this rapid and adaptive compensation to unexpected perturbations at sites remote from the locus of the perturbation qualifies as necessary, but not sufficient evidence for a synergy. Similar results were reported anecdotally, for example, by Abbs and Gracco (1983). The key test of a synergy, however, is its functional-, task- and context-sensitivity. Kelso et al (1982; 1984) demonstrated that when the same perturbation was applied to the jaw during the syllable “baz” (as in the first syllable of a well-known swear word) the lips did not respond at all. Rather tongue muscle activity (the genioglossus) showed a rapid increase in order to carry the tongue to the roof of the mouth to produce the /z/ sound (called a fricative). Although perturbing the jaw represented a threat to both utterances, no perceptual distortion of speech occurred. That a challenge to one member of a group of potentially independent components (here the speech articulators and their neuromuscular control elements) was met—on the very first perturbation experience—by remotely linked members of the group, provides direct evidence for synergies or coordinative structures.
Analogous results come from studies of the human postural system during voluntary movement. In response to a perturbation to the thumb (!) which subjects were using to perform a tracking task, Marsden and colleagues (1983) observed reactions in muscles remote from the prime mover, e.g., the pectoralis major of the same limb and the triceps of the opposite limb. Similar to human speech, these distant reactions are much faster (around 40 milliseconds) than typical reaction times and operate at delays comparable to autogenic reflexes. In line with the synergy hypothesis, the observed responses were both functional and context sensitive. For example, postural responses in triceps disappeared if the non- tracking hand was not exerting a firm grip on a supporting object. If instead of holding onto a table top, the non-tracking hand held a cup of tea, the triceps response reversed—which is exactly what has to happen to prevent the tea from spilling. Marsden et al (1983) concluded that these rapid remote responses to perturbations “constitute a distinct and apparently new class of motor reaction” (p.645). But it seems likely that they were observing a synergy. Bernstein (1967) referring to his early work in the 1920s published in Russian remarks, in line with the synergy hypothesis, that “movements react to changes in one single detail with a whole series of others which are sometimes very far removed from the former, both in space and in time” (p.67).
Some FAQ about synergies
Is everything a synergy?
Not everything is a synergy. If a living thing or a part of a living thing is studied in isolation outside of its role in a specific task or function, it cannot display the properties of a synergy. In such cases, the response of a single isolated component is limited to that component, and does not reflect a synergy. In a true synergy, each component is context-sensitive and carries the potential to couple with other members of the putative synergy. This distinguishes synergies from other units of neural and behavioral organization.
Where do synergies come from?
Fundamentally, synergies are the unique expression of two mechanisms heretofore conceived of as independent: self-organization and natural selection. Self-organization is the basic mechanism that Nature uses to form spatial, temporal and functional patterns in systems open to exchange of energy, matter and information with their environments (Haken, 1983; Nicolis & Prigogine, 1977). Signatures of self-organization come from experimental (Kelso, 1984) and theoretical studies (Haken, Kelso & Bunz, 1985) demonstrating that the recruitment and dissolution of synergies is a dynamic process that depends crucially on stability: a synergy may be stable over a range of naturally occurring variation (or parameter values). However, as conditions become critical or reach threshold values, one synergy may become unstable and switch spontaneously to another. Thus, synergies are best seen as spontaneous forms of organization and reorganization in complex, living things that have been adapted or “tuned” by the forces of natural selection. It seems reasonable to propose that natural selection exploited self-organization as a means to create synergies, since coordination and control are crucial to adaptation and survival. This essential confluence of two basic, complementary mechanisms—self-organization and natural selection-- assures both the stability and flexibility of complex living systems as they evolve in time on multiple levels.
Are synergies learned or innate?
Though natural to ask, this may not be the best question. Some synergies appear to be innate, such as breathing, and others appear to be learned such as speech. Yet breathing requires suitable environmental conditions and speaking relies on a vocal tract capable of producing sounds that are unique and specific to human communicative function. Synergies are best seen as spontaneous forms of organization and reorganization in complex, living things that have been adapted or “tuned” by the forces of natural selection. Development and learning, of course, influence the expression of synergies in significant ways, for example by tailoring existing synergies and creating new synergies to accomplish novel tasks both in and out of the laboratory.
Is there a mathematical basis for synergies?
Several main lines of enquiry are under development. One, the Center Manifold Theorem (CMT) is the essence of the well-known “slaving principle of synergetics” proposed by Hermann Haken (1983). Very briefly, near an instability where complex systems form new patterns the evolving structure or pattern can be described by one or a few unstable modes, the so-called order parameters. Intuitively a separation of timescales occurs: all the stable modes have fast timescales and are ‘enslaved’ to the slowly varying order parameter. Through this mathematical mechanism an enormous reduction of degrees of freedom is possible. Another, lesser-known approach is the so-called Uncontrolled Manifold Hypothesis (UCM) developed by Scholz and Schoner (1999). The idea is that a functional task is associated with selecting a performance variable that is stabilized with respect to perturbations. Individual elements of a putative synergy are allowed to change their states as long as they remain within the manifold but not if they leave it. Hence, the individual elements are said to be less controlled within the manifold than outside it. The operational upshot is that the variance in the selected variable is less that the summed variance of the individual components. The two approaches are complementary: CMT may be said to constitute an approach to the coordination problem (the compression of degrees of freedom near critical points); UCM is an approach to the control problem (i.e., which performance variables are controlled and which are not for a given function). A third line of theoretical development, related to and perhaps encompassing the previous two is coordination dynamics, a conceptual framework for understanding how the parts and processes of living things work together, i.e., synergize. Coordination dynamics views coordination on all levels in terms of informationally coupled dynamical systems (e.g., Kelso, 1995; Sternad & Turvey, 1996; Turvey & Carello, 1996). A major plus of coordination dynamics is that it explicitly takes into account both the intrinsic properties of the component elements and the nonlinear coupling between them. For example, coordination in the brain has been hypothesized to arise as a result of changes in the dynamic balance between the coupling among neural ensembles (mediated, typically by reciprocal pathways) and the expression of each individual neural ensemble's intrinsic properties (typically heterogeneous in nature). All three approaches emphasize the compression of degrees of freedom/reduction of dimensionality typical of synergies and the consequent controllability that such a design confers.
Outlook 1: Brain Synergies
Synergies are the proposed units of organization for living things at all levels. As a complex system composed of billions of cells which in turn is capable of displaying a vast repertoire of behaviors, the brain is likely to be highly synergized both on its own and with the environment. This would mean that the neural circuitry, though supported by anatomy, is essentially configured flexibly in a function- or task-specific manner. To identify synergies in the brain it would be necessary to perturb one member of the synergy (e.g., a piece of cortical tissue known to be engaged for a given task or function) and observe remote compensation by other putatively linked brain areas. A modern tool such as Transcranial Magnetic Stimulation when combined with sophisticated imaging technologies to record remote effects may be a useful means to test whether the organization of the brain is based on task- and function-specific synergies.
Outlook 2: The genotype~phenotype relation
Up to now, scientists’ attempts to relate genes, brains and behavior have been frustrated in part because of inadequate or overly global descriptions of behavior. Despite the availability of new tools to track thousands of genes at once, the gap between genes and human behavior may be too big to bridge. The identification of the synergy as a naturally selected chunk of self-organized behavior could alleviate this problem significantly, and renders the synergy an obvious target for enhancing the molecular understanding of brain and behavioral function. Synergies are the proposed relevant units underlying behavior, not the entire behavior itself. They constitute nature’s way of compressing information in systems of enormous complexity: synergies are nature’s way of ‘cracking’ the complex into the simple. Yet, as a kind of grammar, they make complex behavior on all levels possible.
- Abbs, J. H., & Gracco, V.L. (1983) Sensorimotor actions in the control of multimovement speech gestures. Trends in Neuroscience, 6, 393-395.
- Bernstein, N. (1967) The coordination and regulation of movement. Pergamon, London.
- Easton, T.A. (1972) On the normal use of reflexes. American Scientist, 60, 591-599.
- Edelman, G.E. (1987) Neural Darwinism: the theory of neuronal group selection. Basic Books, New York.
- Edelman, G.E. & Gally, J.A. (2001) Degeneracy and complexity in biological systems. Proceedings of the National Academy of Science, (USA) 98, 13763-13768.
- Gelfand, I. M., & Latash, M.L. (1998) On the problem of adequate language in motor control. Motor Control, 2, 306-313.
- Gelfand, I. M., Gurfinkel, V.S., Fomin, S.V. & Tstelin, M.L., (Eds). (1971) Models of the structural-functional organization of certain biological systems. The MIT Press, Cambridge, MA.
- Haken, H. (1977/83). Synergetics, an introduction: Nonequilibrium phase transitions and self-organization in physics, chemistry and biology. Berlin: Springer.
- Haken, H., Kelso, J.A.S., & Bunz, H. (1985). A theoretical model of phase transitions in human hand movements. Biological Cybernetics, 51, 347 356.
- Kelso, J.A.S. (1984). Phase transitions and critical behavior in human bimanual coordination. American Journal of Physiology: Regulatory, Integrative and Comparative, 15, R1000-R1004.
- Kelso, J.A.S. (1995). Dynamic Patterns: The Self Organization of Brain and Behavior. Cambridge: MIT Press. [Paperback edition, 1997].
- Kelso, J.A.S. (2008) Synergies: Atoms of brain and behavior. In D. Sternad (Ed) A multidisciplinary approach to motor control. Springer, Heidelberg.
- Kelso, J.A.S., Southard, D., & Goodman, D. (1979). On the nature of human interlimb coordination. Science, 203, 1029-1031.
- Kelso, J.A.S., Tuller, B. & Fowler, C.A. (1982). On the functional specificity of articulatory control and coordination. Journal of the Acoustical Society of America, 72, S103.
- Kelso, J.A.S., Tuller, B., Bateson, E. V., & Fowler, C.A. (1984). Functionally specific articulatory cooperation following jaw perturbations during speech: Evidence for coordinative structures. Journal of Experimental Psychology: Human Perception and Performance, 10, 812-832.
- Latash, M., & Turvey, M. T. (Eds. (1996). Dexterity and its development. Hillsdale, NJ: Erlbaum.
- Marsden, CD., Merton, P.A., Morton, H.B. (1983). Rapid postural reactions to mechanical displacement of the hand in man. In Desmedt., J.E., (ed) Motor control mechanisms in health and disease. Raven, New York, pp. 645-659.
- Nicolis, G., & Prigogine, I. (1977). Self-organization in nonequilibrium systems. Wiley, New York.
- Scholz, J.P., & Schöner, G. (1999). The uncontrolled manifold concept: identifying control variables for a functional task. Experimental Brain Research, 126, 289-306.
- Sherrington, C.S. (1906) The integrative action of the nervous system. Constable, London.
- Sternad, D., & Turvey, M. T. (1996). Control parameters, equilibria, and coordination dynamics. The Behavioral and Brain Sciences, 18, 780-783.
- Turvey, M.T. (1977) Preliminaries to a theory of action with reference to vision. In Shaw, R.E., & Bransford, J. (eds) Perceiving, acting and knowing: toward an ecological psychology. Erlbaum: Hillsdale, NJ.
- Turvey, M. T., & Carello, C. (1996). Dynamics of Bernstein's level of synergies. In M. Latash. & M. T. Turvey (Eds.), Dexterity and its development (pp. 339-376). Hillsdale, NJ: Erlbaum.
- Valentino Braitenberg (2007) Brain. Scholarpedia, 2(11):2918.
- Jack Carr (2006) Center manifold. Scholarpedia, 1(12):1826.
- Olaf Sporns (2007) Complexity. Scholarpedia, 2(10):1623.
- Gregoire Nicolis and Catherine Rouvas-Nicolis (2007) Complex systems. Scholarpedia, 2(11):1473.
- James Meiss (2007) Dynamical systems. Scholarpedia, 2(2):1629.
- Rodolfo Llinas (2008) Neuron. Scholarpedia, 3(8):1490.
- Hermann Haken (2008) Self-organization. Scholarpedia, 3(8):1401.
- Philip Holmes and Eric T. Shea-Brown (2006) Stability. Scholarpedia, 1(10):1838.
- Hermann Haken (2007) Synergetics. Scholarpedia, 2(1):1400.
- Anthony T. Barker and Ian Freeston (2007) Transcranial magnetic stimulation. Scholarpedia, 2(10):2936.