Abstract
The invention of behavioral function is often a punctuated event, whereas the development of function is a gradual process. The chapter includes examples from the evolutionary robotics tracking experiment introduced in the previous chapter and shows how both gradual and discontinuous improvement relate to the discovery of function. Attractor landscapes are a conceptual tool that show how invariants appear as agents converge to function. In nature, we find analogies in behavioral function across phyla and taxa, which also exhibit analogous solutions. Analogies in the form and behavior of organisms derive from the ideal implementations of function to which evolution may converge. Convergent evolution towards behavioral function underlies the appearance of instincts and analogous behavior of different organisms. Convergence and divergence also collaborate to resolve an old controversy about punctuated equilibria. In the interplay between the two, an answer can be given to Stephen Jay Gould’s question of what would happen if the evolutionary tape were replayed. The chapter ends with a list of the sources of constancy and variability in behavior.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The structure of the nervous system is doubtlessly fundamental to function. Nonetheless, its importance is but instrumental: any structure endowed with the same potentialities for subserving behavioral function is, from the perspective of an organism’s viability, equivalent. A priori, for functional behavior it is not even necessary that there should be neurons. Any element with the same potentialities may do the same job. Of course, the extraordinary breadth of potentialities of the neuron is not easy to imitate in full, and perhaps only neurons will be able to subserve all breadth of behavior of complex organisms.
- 2.
An interesting issue that arises here is that of historical contingencies in evolution. Neutrality causes organisms’ designs to drift across fitness landscapes, in roughly horizontal paths of equivalent viability. But even neutral mutations can eventually be instrumental for the evolution of function. That is, an organism comes to be dependent on its history of structural modifications, even those that apparently had no role as of its appearance. The contingent history of neutral evolution is relevant for the appearance of function. Sometimes a neutral modification of now becomes a fundamental step to a functional modification in the future. An experimental proof of this in bacteria is found in [1].
- 3.
Fitness is abstraction over viability, where the abilities of organisms can be somehow quantified, and organisms themselves compared.
Fitness is hard, viability is soft. Fitness always produces specific orderings of the quantified agents, totally ordered sets. Conversely, viability’s orderings are subjective, contingent, and context-specific – when they are at all possible. In such orderings, relations of transitivity, reflexivity, or equivalence would only hold in isolated cases. Most of the others are neutral or simply incomparable. Even so, when orderings exist, through a selection procedure they can become motors of convergence. In an artificial evolution, orderings and selection mechanisms are the basis for the appearance of interesting evolutionary phenomena. Evolutionary robotics and the artificial evolution of recurrent neural networks (RNNs) for robot control operate according to idealized orderings of fitness, and unlike nature, arrange individuals according to very regular lattices (totally ordered sets). Incidentally, this is the basis for the distinction I have adhered to so far between fitness and viability. I talk about fitness when there exists a well-defined function for ordering of individuals, whereas “viability” refers to natural and irregular orderings, where relations of transitivity and reflexivity are not well defined. Fitness is essentially an idealization of viability.
- 4.
The networks in this chapter can be found on the accompanying online resources, at http://www.irp.oist.jp/mnegrello/home.html.
- 5.
Behavior is potential in interaction. Lest I move on too swiftly on and forget this important disclaimer, it should be noted that there are more sources of novel behavioral function than structural mutation. An organism couples with its environment, and function appears from this interaction. Changes in environmental circumstance, such as migration or niche alteration, may lead to a new set of possible ways to couple with the environment functionally. But as we focus on the study of the isolated agent, we might be led to overlook the fundamental contributions of the environment to behavioral functions.
- 6.
Yakov Sinai was a Russian mathematician who proved that the motion of a spherical particle in such a setup is not only chaotic, but also ergodic. To add to this complexity, the initial state of the ball was randomized. The cylinder protects the tracking head while being invisible to it, meaning the rays do not see the cylinder.
- 7.
The log files of the evolution can be found on the online resources, at http://www.irp.oist.jp/mnegrello/home.html.
- 8.
In the modular version, evolution of modularity was guided by a procedure which would cut connections between the modules. The arbitrariness of this procedure impairs the ability of analogizing with evolution properly.
- 9.
For a more detailed analysis of the ability to control the wheels with proprioceptor motors, see Sect. 10.3.3.1.
- 10.
See also the footnote on page 16 for the difference between viability and fitness.
References
Blount Z, Borland C, Lenski R (2008) Inaugural article: Historical contingency and the evolution of a key innovation in an experimental population of Escherichia coli. Proc Natl Acad Sci 105(23):7899
Di Paolo E, Noble J, Bullock S (2000) Simulation models as opaque thought experiments. In: Artificial life VII: Proceedings of the seventh international conference on artificial life. MIT, Cambridge, CA, pp 497–506
Eldridge N, Gould S (1972) Punctuated equilibria: An alternative to phyletic gradualism. In: Models in paleobiology. Freeman Cooper & Company, San Francisco, California
von Foerster H, von Glaserfeld E (2005) Einführung in den Konstruktivismus, 9th edn. Piper Press, Munich
Gould SJ (1990) Wonderful life: The burgess shale and the nature of history. WW Norton & Company, New York
Gould SJ, Lewontin RD (1979) The spandrels of San Marcos and the Panglossian paradigm: A critic of the adaptationist programme. Proc R Soc Lond 205:581–598
Hochner B, Shomrat T, Fiorito G (2006) The octopus: A model for a comparative analysis of the evolution of learning and memory mechanisms. Biol Bull 210(3):308
Mahn B (2003) Entwicklung von Neurokontrollern für eine holonome Roboterplatform. Diplomarbeit, Fachhochschule Oldenburg / Ostfriesland / Wilhelmshaven
Mayr E (1954) Change of genetic environment and evolution. In: Evolution as a Process. Allen and Unwin, London, pp 157–180
Mayr E (1976) Evolution and the diversity of life. Harvard University Press, Harvard
Morris SC (2003) Life’s solution: Inevitable humans in a lonely universe. Cambridge University Press, Cambridge, UK
Reidys C, Stadler P, Schuster P (1997) Generic properties of combinatory maps: neutral networks of RNA secondary structures. Bull Math Biol 59(2):339–397
Schuster P (1997) Landscapes and molecular evolution. Physica D 107(2–4):351–365
Simpson G (1953) The baldwin effect. Evolution 7:110–117
Sporns O (2002) Graph theory methods for the analysis of neural connectivity patterns. In: Neuroscience databases A practical guide, pp 169–83
Varela F (1979) Principles of biological autonomy. North Holland, New York
Varela F, Maturana H (1987, 1998) The tree of knowledge, 1st edn. Shambala, Boston, MA
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2011 Springer Science+Business Media, LLC
About this chapter
Cite this chapter
Negrello, M. (2011). Convergent Evolution of Behavioral Function. In: Invariants of Behavior. Springer Series in Cognitive and Neural Systems, vol 1. Springer, New York, NY. https://doi.org/10.1007/978-1-4419-8804-1_9
Download citation
DOI: https://doi.org/10.1007/978-1-4419-8804-1_9
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4419-8803-4
Online ISBN: 978-1-4419-8804-1
eBook Packages: Biomedical and Life SciencesBiomedical and Life Sciences (R0)