|
| Behavioral Evolution and Selection by Consequences |
| Monday, May 30, 2016 |
| 9:00 AM–10:50 AM |
| Alpine, Swissotel |
| Area: TPC/EAB; Domain: Translational |
| Chair: Jose E. Burgos (University of Guadalajara) |
| Discussant: Jose E. Burgos (University of Guadalajara) |
| Abstract: Darwin’s theory of evolution may be summed up in the phrase “selection by consequences.” Seen as a general process, selection by consequences applies to genetic evolution, cultural evolution, and behavioral evolution. Genetic evolution concerns change in populations of organisms across generations. Cultural evolution concerns change in behavioral patterns within groups across generations or lesser time periods. Behavioral evolution concerns change in the behavior of an individual organism within its lifetime. This symposium will illustrate a variety of approaches to understanding behavioral evolution as selection by consequences: by analysis, by modeling, and by applying known theory and data. Catania will focus on verbal behavior as a cultural phenomenon. McDowell will discuss how behavioral evolution may be implemented in artificial life forms. Smith will discuss the conceptual benefits of selection by consequences to scientific thinking about behavior. Baum will talk about the analytical power of an equation derived by George Price when applied to behavioral dynamics. These various approaches to behavioral evolution may bring behavior analysis closer to biology and restate or replace the law of effect. |
| Keyword(s): Behavioral Evolution, Selection, Variation |
| |
| Improving on the Meme: Cultural Selection and the Shaping of Verbal Behavior |
| A. CHARLES CATANIA (University of Maryland, Baltimore County) |
| Abstract: Skinner has discussed three varieties of selection: in phylogeny, as in Darwin’s natural selection; in ontogeny, as in the shaping of operant behavior; and in culture, as behavior is passed on from some individuals to others in what he called cultural selection. Dawkins introduced the meme as a unit passed on from some individuals to others, but it was not well-defined; also, the transmission of memes received more attention than their evolution. If we regard memes as units of behavior, however, they become examples of Skinner’s cultural selection, and for Skinner the most crucial example of cultural selection was verbal behavior. Phonemes provide good examples of such culturally selected or memetic units: infant vocalizations are shaped by automatic reinforcing consequences, as they come more and more closely to resemble those heard in the verbal environments created by their caregivers (cf. Skinner, Risley, Palmer). Echoic behavior, a product of this shaping, is defined by correspondences of phonetic rather than physical units. Such selectionist accounts of verbal behavior can be extended to other properties of verbal behavior, such as verbal governance, and they have implications for theories of language evolution. |
| |
| Two Versions of Variation and Selection |
| TERRY SMITH (Edinboro University of Pennsylvania) |
| Abstract: If one focuses upon behavior that adapts to a changing environment in the interest of the organism, then a framework of variation and selection offers a promising approach to the analysis of behavior. Two contemporary behavioral scientists that exemplify this approach are John Staddon and Jack McDowell. In both cases, behavioral adaptation during the lifetime of the individual is analyzed as a process of evolutionary interaction between behavior and the environment. In quite different ways, these two scientists offer alternatives to mentalistic, cognitive theorizing. Staddon’s formalization addresses the impulse to posit beliefs and desires; McDowell’s the impulse to posit information processing structures. Together, they provide a broadly unified approach to doing without beliefs, desires, and information processing structures. Both formulations, however, are theories of a kind that B. F. Skinner suggested were unnecessary in his 1950 article, “Are Theories of Learning Necessary?” They nevertheless exemplify behaviorism and have demonstrated significant advantages over Skinner’s non-theoretical behaviorism. |
| |
| Algorithmic Behavioral Evolution as Artificial Intelligence |
| JACK J. MCDOWELL (Emory University) |
| Abstract: Artificial Life is a branch of Artificial Intelligence that seeks to animate artificial agents using algorithms that mimic the functioning of biological organisms. An algorithmic evolutionary theory of adaptive behavior dynamics has been studied extensively over the past decade and has been shown to produce behavior in virtual (software) agents that is qualitatively and quantitatively indistinguishable from live organism behavior in a variety of environments. Because these agents reproduce the behavior of biological organisms, they are artificial life forms. It is also possible to create artificial life in mechanical agents. To illustrate, a specific mechanical agent, a robot spider, can be animated by the evolutionary theory and placed in a 2-dimensional grid world where prey items are made available at various locations and times. Surface navigation of the grid world requires the addition of stimulus control to the evolutionary theory. The robot spider’s foraging behavior is completely determined by the evolutionary theory, which means that it behaves autonomously in the grid world. The evolutionary algorithm causes the spider’s foraging behavior and prey capture to adapt to the spatial and temporal pattern of prey availability, and to readapt when the pattern of prey availability changes. |
| |
| Behavioral Evolution and the Price Equation |
| WILLIAM M. BAUM (University of California, Davis) |
| Abstract: Price’s equation describes evolution across time in simple mathematical terms. Although it is not a theory, but a derived identity, it is useful as an analytical tool. It affords lucid descriptions of genetic evolution, cultural evolution, and behavioral evolution (often called “selection by consequences”) at different levels (e.g., individual versus group) and at different time scales (local and extended). The importance of the Price equation for behavior analysis lies in its ability to precisely restate behavioral selection by consequences, thereby restating, or even replacing, the law of effect. When applied to dynamics in operant performances, the equation affords a way to evaluate methods and measures and to compare among measures, because deviations from it indicate deficiencies either in assumptions or in measures. Price’s equation may also serve as a base for theory. For example, applying the equation to the dynamics within stable performance on a variable-interval schedule, with a few assumptions, results in a complete explanation for the moderate response rates that occur. |
|
| |