Adaptive Individuals in Evolving Populations:
Models and Algorithms
Glossary
|
||
|
|
|
|
|
|
||
Selected Keywords
As the biologists, psychologists, and computer scientists contributing to this volume have attempted to understand one another, it has become clear that much of the confusion can be traced back to a number of "keywords" that are central to our conversation.In many cases, these terms have been quite precisely defined within one discipline or another but also retain more colloquial, "folk" connotations from general parlance.The challenge is to retain the hard-won precision of one field's technical term while extending its use to the interdisciplinary phenomena experienced by others.Our current effort to provide working definitions for some of the most critical keywords is modeled after Fox Keller and Lloyd's (1992) Keywords In Evolutionary Biology.Fox Keller and Lloyd make an eloquent argument for the importance of such a vocabulary to scientific progress, and for the insight these words can give into the scientific process they mediate.(The introduction to this volume discusses this connection in more depth.)
In our case, the following "definitions" have come from our authors and workshop participants, sources recommend by them, a group of students at the University of California at San Diego who were early readers of this manuscript, dictionaries, and encyclopedias.We are most grateful to all these contributors for their willingness to participate in this unusual intellectual amalgam.The terms are often defined in the context of readings in this volume.In some cases this large collection of authors has generated multiple, sometimes inconsistent and sometimes fully contradictory definitions of the same term.Rather than attempting to reconcile the various definitions into a single, authoritative one (as a logical positivist would recommend), we present the variations to the reader as an indication of the rich texture and scientific questions remaining to be answered.
accommodation - Adaptation on the individual level, via interaction with the environment; nonhereditary adaptation. See ASSIMILATION
* According to Piaget (Chapter 19), the infant's process of changing an existing schema to fit an external stimulus.
action - A primitive unit of behavior. In simulations, often one of a discrete set of choices the individual is faced with at each (discrete) moment in time.
adaptation - To make fit (as for a specific or new use or situation) often by modification; modification of an organism or its parts that makes it more fit for existence under the conditions of its environment (On-line Webster program). See also Fox Keller and Lloyd (1992).
* The process of changing the parameters of a system from their initial values to the values needed to solve a particular problem optimally.
* Genotypic change which results in an individual with fitness superior to that of its parent.
* Adaptation is also a description of a feature, or state, of adaptedness. For example, the gill is an adaptation to the marine environment.
adaptationist - The tendency to believe that the evolutionary process unfailingly optimizes species to perfectly fit their niche.
* A school of thought in evolutionary theory that views most or all organismal traits as adaptive, and, more specifically, as having arisen as a direct result of natural selection acting on the trait in question.
adaptation, co- - Mutually adapted, especially by natural selection (On-line Webster program).
* Used by Hinton and Nowlan (Chapter 25) to describe the specific situation in which a set of genetic alleles is not adaptive unless they occur in conjunction.
adaptations, pre- - Genetic changes which result in no increase in fitness, but which have been analyzed post hoc as having laid the groundwork for subsequent adaptive changes.
* A trait or traits that lay the groundwork for future adaptive changes. These traits may be random with respect the subsequent adaptive changes, or they may have come into being through selection pressures which are irrelevant to the subsequent ones. For example, it has been speculated that the birds' proto-wings evolved first as "sweepers" by which running bipedal reptiles swept their prey into their mouths. The trait of swinging feathered arms while running at top speed may have been a preadaptation for flight.
adaptive computation - See COMPUTATION, ADAPTIVE
allele - One of a group of genes that occur alternatively at a given locus (On-line Webster program). See GENE
* In genetic algorithms terminology (e.g., as used by Hinton and Nowlan, Chapter 25), it is one of the forms that a gene can take (e.g., fixed present, fixed absent, or modifiable).
artificial intelligence - The capability of a machine to imitate intelligent human behavior (On-line Webster program).
* The capacity of a digital computer or computer-controlled robotic device to perform tasks commonly associated with the higher intellectual processes characteristic of humans, such as the ability to reason, discover meanings, generalize, or learn from past experience. The term is also frequently applied to that branch of computer science concerned with the development of systems endowed with such capabilities ("artificial intelligence", Britannica Online [13 November 1995]).
artificial life A field of study devoted to understanding life by attempting to abstract the fundamental dynamical principles underlying biological phenomena, and recreating these dynamics in other physical media--such as computers--making them accessible to new kinds of experimental manipulation and testing (Langton et al., 1992).
assimilation - According to Piaget, the infant's process of incorporating external stimuli to an existing schema. See ACCOMMODATION
assimilation, genetic - The "genetic fixation" (via selective processes) of "acquired" characters (phenotypes). Prior to fixation, these characters seem to appear solely as a response to specific environmental stimuli, i.e., they are modifications whose range of variability is genetically controlled; after fixation (assimilation) they are also produced in the absence of the particular environmental condition ("pseudo-exogenous adaptations").
* The phenotypes assumed to be somatically "acquired," stabilized, and genetically assimilated may in fact be a consequence of "threshold effects" (Stern, 1958). In the case of "threshold selection," the special environmental conditions may reveal which individuals (from among a number) already carry polygenes and modifiers of the phenotype in question. The character differences may be subthreshold and thus not discernible in the original environment and above threshold in the new environment where they are endowed with positive selective value. Under these circumstances genotypes whose action is above threshold in both the original and the new environment may arise by crossing (Reiger, Michaelis, and Green, 1991).
* Almost 50 years after Baldwin and his contemporaries, Waddington (1942) proposed a similar but more plausible and specific mechanism that has been called "genetic assimilation." Waddington reasoned that certain sweeping environmental changes require phenotypic adaptations that are not necessary in a normal environment. If organisms are subjected to such environmental changes, they can sometimes adapt during their lifetimes because of their inherent plasticity, thereby acquiring new physical or behavioral traits. If the genes for these traits are already in the population, although not expressed or frequent in normal environments, they can fairly quickly be expressed in the changed environments, especially if the acquired (learned) phenotypic adaptations have kept the species from dying off. The previously acquired traits can thus become genetically expressed, and these genes will spread in the population. Waddington demonstrated that this had indeed happened in several experiments on fruit flies (Mitchell, 1996).
* Waddington's version of the Baldwin effect, but with his perspective of mutation as a back-filling process, versus Baldwin's exploratory one.
* Schull (Chapter 3) sees genetic assimilation as a discipline-specific special case of Bateson's formulation. It takes the Baldwinian scenario and transposes it from the level of intelligent behavior to the level of physiological adaptation.
attainability - The problem of providing a model that is not only accurate as far as fitting data, but also biologically or cognitively plausible. For Shafir and Roughgarden (Chapter 12), attainability motivates a solution to optimal foraging that is based on abstract operations available to the individual (lizard).
Baldwin effect - The tendency of organisms that learn or acquire useful characteristics to be successful, leading to a higher probability of their reproduction and a fixation of their useful characteristics in the population despite the absence of direct inheritance of these characteristics.
* The tendency for ontogenetically acquired adaptations (e.g., acquired through learning) to produce evolutionarily mediated changes in a species--population-genetic adaptation in response to selection pressures produced in part by the prior ontogenetic adaptations. (N.B., This definition avoids the concepts of genes and population genetics, which post-date Baldwin's formulation.)
* According to Schull (Chapter 3), the "Baldwin effect" was a term Baldwin managed to attach to a family of processes about which he and several of his contemporaries (Lloyd Morgan and Poulton) theorized. Baldwin battled for the credit, and in doing so managed to narrow the scope and to obscure the generality of the concept. In his scenario, which was intrinsically self-limiting, the effect of intelligent plasticity was to usher in nonintelligent, nonplastic instincts, thereby causing evolution to cease.
* According to Hart and Belew (Chapter 27), the increase of evolutionary performance (e.g., average population fitness) in non-Lamarckian evolution when the frequency of learning is sufficiently small.
* Definitions vary in the scope of evolutionary situations that they encompass, but most seem to involve the evolution of genetically determined characteristics similar in function to those that arise through phenotypic plasticity. Todd (Chapter 21) suggests an even broader definition. In his simulation, when the sexual imprinting gene is on, the organism makes no reference to its mate preference genes. So, it seems that there would be little pressure for learned mate preferences to become innate. However, individual learning affects mate choice, which certainly affects the genetic traits of offspring. The article seems to say that the term "Baldwin effect" could encompass this.
behavior - Actions taken by an organism that change the environment and/or the organism's relationship to the environment.
* Phenotypic traits exhibited in the interaction between an organism and its environment.
behavior, optimal - In a simulation, the behavior of an individual with the largest possible fitness score.
behaviorism - A school of psychology that takes the objective evidence of behavior (as measured responses to stimuli) as the only concern of its research and the only basis of its theory without reference to conscious experience (On-line Webster program).
* A school of psychology which posited that behavioral data and a behavioral theory of psychology must be restricted to measurable stimuli and measurable responses. Conscious experiences, purposive intentions, or any other psychological or mental processes were specifically considered beyond any science of behavior.
* According to Schull (Chapter 3), materialistic and behavioralistic interpretations of biology (e.g., those that rejected Lamarckism not only for its account of inheritance but also for its granting of a causal role to the individual organisms, needs, wants, and intentions) created a climate that was antipathetic to the broader implications of organic selection, even though they were actually quite compatible with them. Skinner (Chapter 18) saw that learned behavior could change selection pressures and thereby influence the course of biological evolution. Skinner's theories received little attention within biology, however.
benefit - Selective advantage. Not to be confused with "utility"; traits useful to an individual may still not confer a reproductive advantage, and vice versa.
biology - 1: A branch of knowledge that deals with living organisms and vital processes; 2a: the plant and animal life of a region or environment; b: the life processes of an organism or group; broadly: See ECOLOGY (On-line Webster program).
* According to Schull (Chapter 3), Lamarck was among the first to use this word. Lamarck thought it would be an appropriate new discipline that his philosophy of science could nurture.
biosphere - Living beings together with their environment (On-line Webster program).
* What James referred to as the "earth" and "environment." (See Schull, Chapter 17.)
* What many now refer to as Gaia (Lovelock, 1987).
caloric - The invisible fluid that was believed to cause heat. Throughout Lamarck's article (Chapter 4) fluids play an important causal role in his explanations. For example, Larmarck suggests that animals that butt heads together cause a flow of fluids to their heads which eventually causes horns to form (over many generations).
canalization - 1a: To provide with a canal or channel; b: to make into or similar to a canal; 2: to provide with an outlet; esp.: to direct into preferred channels (On-line Webster program).
* Waddington demonstrated that plasticity (in morphological development) could influence selection of genes which facilitate, elaborate or "canalize" the particular adaptive response. Over several generations adaptive response could become so reliable that plasticity seems irrelevant even though it played a crucial role at the outset.
* According to Waddington (Chapter 7), canalization forces evolution on a particular path, on which the course of evolution will remain regardless of minor variations. His most explicit invocation of the metaphor is in his description of a ball rolling on the developmental landscape; the grooves worn are the canals.
* Reduced variance; tendency of a maturational system to remain constant in the face of (external, environmental) perturbation.
Central Dogma of biology - Information flows from DNA to RNA to proteins.
complexity - An imprecise term used to describe characteristics of systems including a large number of parts and interactions between parts, nonlinearity of interactions, and difficulty of analysis and prediction of behavior. The term is used precisely in some contexts, e.g., "computational complexity," "environmental complexity," etc.
complexity, environmental - A more complex environment has a larger number of more varied survival strategies. Complexity is a function of the "chemistry," "physics," the topology of the environment, the modes of interaction available to the simulated environment, and similar considerations.
computation, adaptive - Roughgarden et al. (Chapter 2) refer to an unbreak attributed quote, "all reality is computation," and describe Adaptive Computation as an agent-oriented (or individual-oriented) method of computing (and maximizing) the fitness of discrete individuals engaging in discrete interactions with their environments. It is contrasted with differential equations, which treat the world as being continuous. The implication is that the discrete method is better at describing the complexity of small populations than the continuous method. See INDIVIDUAL-BASED MODELING
* The theory and practice of building computational systems that are able to adapt--i.e., improve their performance in response to their environment, continue to perform satisfactorily in changing environments, and cope well with wholly new environments. Adaptive computation researchers often take inspiration from natural adaptive systems in developing adaptive computer systems. Such nature-inspired systems include genetic algorithms, neural networks, economic approaches to resource allocation, and immune-system-based pattern recognition and computer security systems. Adaptive computation is also often taken to include computational modeling of natural adaptive systems, such as that done by Hinton and Nowlan (Chapter 25) and Menczer and Belew (Chapter 13).
computer science - The theory and practice of computation, typically taken to refer to computation on human-constructed electronic digital or analog devices. In this book the computer scientists focus primarily on using such devices to simulate natural systems and to perform experiments on these simulations. The word "computation" can also refer to information-processing activities in natural systems, such as learning in organisms. In this context, computer scientists can study the computational properties of these behaviors as a way of understanding aspects of nature.
contingency - A condition under which a behavior and its consequences occur.
correlation, dynamic - Two surfaces are dynamically correlated in a region in which movement that results in increased values on one surface in that region results in increased values on the other surface.
correlation, statical - Two surfaces are statically correlated if they tend to have peaks and valleys in similar places, or, at the least, if those areas with high values for one surface have high values for the other, and those areas with low values for one surface have low values for the other.
cost - Selective disadvantage. As with "benefit," this does not refer to a hardship or inconvenience but rather to a trait that decreases the organism's reproductive success.
development - All changes occurring to an individual during its lifetime, whether due to genetic causes or interaction with the environment. "Maturation" is used to refer to changes associated primarily with the former, and "learning" with the latter. See MATURATION, LEARNING, ONTOGENY
ecological - "...given in terms of the relationship between the animal and its environment, rather than in terms of the animal alone" (Johnston, Chapter 20). Johnston wants to distinguish ecological learning--that to which the cost/benefit analysis is applicable--from "surplus" (selectively irrelevant) learning.
ecology of mind - Bateson's (Chapter 9) evocative term (and book title) for a science which accommodates mental phenomena in a science spanning both psychology and biology. In these matters, Bateson drew inspiration from Lamarck (a pre-Darwinian), Samuel Butler (an anti-Darwinian), and Alfred North Whitehead (an admirer of the trans-Darwinian William James). Schull's chapters are a post-Batesonian attempt to show where the ecology of mind comes from, to show where it might be leading, and to show how it is consistent with, but broader than, traditional Darwinism. Bateson's general formulation of the Baldwin effect is one example of the kind of discipline-spanning evolutionary theory that Bateson advocated. See SELECTION, ORGANIC
ecology - 1: A branch of science concerned with the interrelationship of organisms and their environments; 2: The totality or pattern of relationships between organisms and their environment (On-line Webster program).
* Roughgarden et al. use the term in a pretty standard way, differing from Webster's only in that they choose to emphasize the role of the environment in determining individual development.
energy - A measure of capability to perform work. Energy can be stored (potential energy) but in a closed system, according to the Second Law of Thermodynamics, it tends to be transformed into motion (kinetic energy). Energy in a closed system (potential plus kinetic) is conserved.
energy, latent - If performing work W upon a system enables it to perform work W´>W, we say that the system stores latent energy W´-W. From an ethological point of view, this means that behavior must supply work in order to realize (use or store) energy latent in an environment.
environment - Used in reference to the physical environment but also seems to include habits. See also Fox Keller and Lloyd (1992).
* The complex of physical, chemical, and biotic factors that acts upon an organism or an ecological community and that ultimately determines its form and survival.
* Everything outside the structure (skin) of the individual organism.
* In Zhivotovsky et al.'s model (Chapter 10) a set of "external factors" that can be described by an mth-order Markovian process. The transition probabilities are unchanging over time. The environment is assumed to undergo a very large number of transitions in the course of an individual's lifetime. The authors seem to have in mind something like whether it is raining, or where the individual happens to be today. This is in contrast to a slowly changing environment, such as whether the organism is living in a rain forest or a desert.
* In Shafir and Roughgarden's model (Chapter 12), environment is completely characterized by the abundance of prey per unit area-time. In the simulations this is a constant parameter. This greatly simplified model of environment seems appropriate for an "idea" model in which fitting actual data is not a primary concern, compared to tractability.
environmental information - For Zhivotovsky et al. (Chapter 10), this is a priori information available to individual about an environment prior to its experience of it. It is encoded as a function giving the probability of the next environmental state, based on the last l states. This information may be complete (if l = m, the order of environmental process itself) or incomplete. This information is assumed to be absolutely correct. The case where it is not correct is not considered. This seems to be an unrealistic notion of information. Organisms probably have many ways of implicitly predicting future environmental states, but they almost certainly lack perfect information about conditional probabilities.
epigenesis - The concept that an organism develops by the new appearance of structures and functions, as opposed to the hypothesis that an organism develops by the unfolding and growth of entities already present in the egg at the beginning of development (preformation) (King, 1990).
* Development in which differentiation of an individual's cells into organs and systems arises primarily through their interaction with the environment and each other.
epistasis - A form of gene interaction whereby one gene interferes with the phenotypic expression of another nonallelic gene (or genes), so that the phenotype is determined effectively by the former and not by the latter when both genes occur together in the genotype.... In population genetics and quantitative genetics, the term epistasis is sometimes used to refer to all nonallelic gene interactions (Reiger, Michaelis, and Green, 1991, p. 170; see also Fox Keller and Lloyd, 1992).
* Nonlinear interactions among genes due to their linkage on the chromosome under recombination events.
* The phenomenon of a phenotypic trait being affected by nonadditive interactions among multiple genes.
equilibration - The process whereby an organism adapts to its environment, forming a correspondence. The process by which internal and external variables interact to produce a certain set of behaviors.
evolution - A process of change in a certain direction; a process of continuous change from a lower, simpler, or worse state to a higher, more complex, or better state; a theory that the various types of animals and plants have their origin in other preexisting types and that the distinguishable differences are due to modifications in successive generations (On-line Webster program). See also Fox Keller and Lloyd (1992).
* Evolution is both process and result. The process of evolution includes all mechanisms of genetic change that occur in organisms through time, with special emphasis on those mechanisms that promote the adaptation of organisms to their environment or that lead to the formation of new reproductively isolated species. The result of evolution is observed in the evolutionary history of life on Earth and in the genetic relationships among species that exist today (Hartl, 1988, p. 143).
* Time course of any system. Physicists often describe dynamical systems in this way.
* The process of change from a lower (simpler) to a higher (more complex) form of biological organization through accumulation of variations and selective reproduction. However, few would disagree that parasites typically "devolve" from higher forms, and yet are simpler in significant ways from their progenitors.
* Changes in the genome that alter inborn behavior; in contrast to "learning."
exaptation - Selectively neutral traits that are subsequently adapted towards some function of selective consequence. See PREADAPTATION
feedback signal - Any information available to an adaptive process that might alter its state. In a learning framework, input that an individual can use to improve itself relative to some objective. Some examples include supervisory signals, reinforcement signals, unsupervised signals, and heuristic signals.
fitness - The relative success an individual achieves at reproduction, relative to other members of the same generation. See also Fox Keller and Lloyd (1992).
* The relative ability of an organism to survive and transmit its genes to the next generation (King, 1990).
* In population genetics, a quantitative measure of reproductive success of a given genotype, i.e., the average number of progeny left by this genotype as compared to the average number of progeny of other, competing genotypes (a.k.a. adaptive value, selective value) (Reiger, Michaelis, and Green, 1991).
* In the genetic optimization framework, a measure of how well an individual achieves the goal set by the designer of the simulation (e.g., predicting an environment); that which the adaptive system optimizes.
* The relationship between a reproducing pattern (gene, genotype, or phenotype) and its environment which determines the differential reproductive success of that entity relative to competing alternatives.
* The way to determine the reproductive success of an individual (candidate solution) with respect to the rest of the population. It can be exogenous (defined a priori) or endogenous (defined by interactions between individuals and an environment that may not be completely specified a priori).
* To be distinguished from "adaptedness." For example, imagine a hypothetical individual that produces ten times as many offspring as others in the population, but imagine also that all the offspring are sterile. The adult parent is more fit, but not well adapted.
fitness, inclusive - Going beyond the fitness of an individual to consider the collective fitness of an individual and all genetically close relatives.
* The sum of an individual's fitness, quantified as the reproductive success of the individual and its relatives, with the relatives devalued in proportion to their genetic distance (Reiger, Michaelis, and Green, 1991).
function - Baldwin (Chapter 5) speaks of "the functions which an organism performs in the course of his life history," "new or modified functions," "normal congenital functions," etc. Baldwin's central definition of "Organic Selection" involves the word "function": "the organism's behavior in acquiring new modes or modifications of adaptive functions with its influence of structure." Function is also, according to Baldwin, a central part of the Lamarckian argument: "the evidence drawn from function, `use and disuse,' is discredited."
* In mathematics, a function is a relation between values in a set D and values in a set R such that for all values d in D, there is associated a unique value r=f(d) in R.The value r can be viewed as a consequence (or result) of the value d.For example, a "fitness function" maps individuals to fitness scores (real numbers).
gene - The basic unit of Mendelian inheritance which represents a contiguous region of DNA (or RNA in some viruses) corresponding to usually one (less often two or more) transcription unit or transcription. The particular sequence of nucleotides along the nucleic acid molecule represents a function unit of inheritance (defined operationally by the cis-trans test), a cistron.... Genes can be divided into those that code for polypeptides (structural genes), those that are transcribed into RNA...but not translated into proteins, and possibly those in which functional significance does not demand that they are transcribed at all (DNA sequences that may govern the punctuation or regulation of genetic transcription).
A gene consists of a linear array of potentially mutable units...between which intragenic genetic recombination can occur and permits the genetic mapping of mutational sites....
As a result of gene mutation, alternative forms, referred to as alleles, of a particular gene are produced. The existence of a gene is extrapolated from these alleles which generally influence the same phenotypic character or trait (being the product of gene expression)....
Genes are the basis of both continuous (qualitative) and discontinuous (quantitative) characters and usually produce effects on a wide variety of biochemical and morphological characters. So-called "pleiotropic" "effects" [sic] are the result of a variety of effects originating from a single primary action, i.e., coding for a definite polypeptide ("one cistron-one polypeptide model").
The collection of genes contained within one chromosome in linear sequence (genetic map) and in definition positions constitute a linkage group. There is evidence, in some microorganisms at least, that genes which control a series of related biochemical reactions are often adjacent to each other in the linkage structure (gene cluster). The same is true for the case of the mutational sites within one gene. Sites with similar properties show a tendency to cluster and are not distributed randomly within the nucleotide sequence defined functionally as a gene (Reiger, Michaelis, and Green, 1991, pp. 189-191; see also Fox Keller and Lloyd, 1992).
genetic algorithm (GA) - An evolutionary model that performs global adaptive search with respect to an objective function. A population of candidate solutions are encoded as gene vectors and evaluated relative to the objective function. Some vectors are selected to differentially reproduce and are subjected to genetic variation (e.g., mutation, crossover). The resulting set of gene vectors forms the next generation, and the cycle repeats itself until some termination criterion is reached.
genome - See GENOTYPE
genotype - 1: The genetic constitution of an individual or group; 2: a class or group of individuals sharing a specified genetic makeup (On-line Webster program; see also Fox Keller and Lloyd, 1992). See PHENOTYPE
habits - The meaning of habits seems to be more general in usage than the modern use of the word and maps onto our use of the word "behavior."
immune system - The system responsible for recognizing, and defending against, pathogens, toxins, and other foreign molecules.
individual - 1a: A particular being or thing as distinguished from a class, species, or collection; b(1): a single human being as contrasted with a social group or institution; (2): a single organism as distinguished from a group; c: a particular person; 2: an indivisible entity; 3: the reference of a name or variable of the lowest logical type in a calculus (On-line Webster program).
individual-based modeling - See MODELING, INDIVIDUAL-BASED Lamarckian inheritance - Inheritance in which characteristics acquired during the lives of one generation are passed on to the next.
* This notion of inheritance was required by (but not unique to) Lamarck's theory of biological evolution. Darwin (as well as Lamarck) implicitly assumed that acquired characteristics, like somatic characteristics, were transmitted physiologically through the germ line, an assumption that was refuted by August Weismann. learning - Knowledge or skill acquired by instruction or study. Modification of a behavioral tendency by experience. Coming to know (On-line Webster program).
* "...any process in which, during normal, species-typical ontogeny, the organization of an animal's behavior is in part determined by some specific prior experience." (Johnston, Chapter 20.) Learning, for Johnston, excludes both conditioning per se and "nonspecific" experience such as inadequate nutrition.
* Often used synonymously with modification, independent of effect on resulting performance at some skill or knowledge.
* Refers to types of within-lifetime developmental growth that are primarily due to the individual's interaction with the environment. See MATURATION
learning algorithm - A set of rules that determine how an individual's experiences affect its future behavior.
mastery effect - The diminution of evolutionary improvement when learning alone is capable of attaining maximal performance at some task. Especially dependent on the non-Lamarckian character of a given simulation, as well as a lack of a fitness cost based on the amount of learning performed by a given individual. Both of these dependencies are met by the GA used by Hightower et al. (Chapter 11).
maturation - 1a: The process of becoming mature; b: the emergence of personal characteristics and behavioral phenomena through growth processes; c: the final stages of differentiation of cells, tissues, or organs (On-line Webster program).
* Refers to types of within-lifetime developmental change that are primarily due to genetically determined growth processes. See LEARNING
meme - A hypothesized analog in systems of cultural change to the gene in biological evolution.
memory - The retention and retrieval in the human mind of past experiences.... The function of remembering and its converse, forgetting, are normally adaptive. Learning, thought, and reasoning could not occur without remembering. On the other hand, forgetting has many functions, including time orientation by virtue of the tendency of memories to fade over time; adaptation to new learning by the loss or suppression of old patterns; and relief from the anxiety of painful experiences ("memory", Britannica Online [13 November 1995]).
* In Zhivotovsky et al. (Chapter 10), a record of the last k environmental states, which is used in making a decision about the next state. Nothing else counts as memory, in particular no information about states before the last k. It is not necessary that the organism `recall' these states, only that they have an effect on the decision.
* Storage of past events. In Shafir and Roughgarden (Chapter 12), memory refers to the accumulation of prey appearances across a temporal window.
model - A miniature representation of something; an example for imitation or emulation; a description or analogy used to help visualize something (as an atom) that cannot be directly observed; a system of postulates, data, and inferences presented as a mathematical description of an entity or state of affairs. (See On-line Webster program.) Roughgarden et al. (Chapter 2) follow the last definition very closely.
* A set of assumptions, identifying the essential features of a system, presented as a mathematical or algorithmic description of the system. This simplified description can be used as a theoretical laboratory in order to test hypotheses about the system.
model, individual-based - Rather than assuming all individuals in a population are the same, or that average values of individuals in a population can be used to describe population phenomena,...individual-based models have been developed by ecologists and evolutionary biologists who want to emphasize individual differences and local interactions.
There are two kinds of individual-based models: (1) distribution and (2) configuration. Distribution models consider individual differences by lumping together individuals with common characters. Configuration models track all individuals in a population to incorporate both individual differences and local interaction. In a broad sense, individual-based configuration models can be regarded as artificial life (Kawata and Toquenaga, 1994).
model, minimal idea - A model that explores some idea or concept without the inconvenience of specifying any particular species, environment, or much of anything else. The premise is that some phenomenon of interest--say, sexual reproduction--is a computational entity whose properties will be pretty much the same across a wide range of possible universes. See MODEL, MINIMAL SYSTEM
model, minimal system - An exploration of the dynamics of some greatly simplified subset of the features of a real environment. Basically a minimal idea model with some contact to the real world. See MODEL, MINIMAL IDEA
model, synthetic system - Also called a systems model, a synthetic model is supposed to be an expansion of a minimal system model in which (ideally) all the assumptions are treated formally. The implication is that it is so huge that it could only be the product of large-scale collaboration; interestingly, Roughgarden et al. (Chapter 2) make no mention of synthetic models in their discussion of how computation might be useful to biology.
modification - Changes to the "body structure" of a phenotype.
mutation - Any genetic difference between parent and offspring.
* Random modifications to the genome. Note that crossover (the model of sexual recombination) in the genetic algorithm is distinguished from mutation under this definition, while included under the first definition.
* In modern biology, mutation usually refers to a change in the genotype that may or may not result in a phenotypic change. Lamarck used this term to refer to a morphological change in the phenotype. Neo-Darwinian - A theory that holds natural selection to be the chief factor in evolution and specifically denies the possibility of inheriting acquired characters (On-line Webster program).
* The "Neo-" qualifies the integration of Darwin's theory about the process of evolution with genetic models developed following the rediscovery of Mendel's work and subsequent mathematical models of population genetics near the turn of the twentieth century. neural network, artificial - A computational model, usually simulated on a general-purpose computer, that maps vectors of numbers to different vectors of numbers through a sequence of mathematical transformations. It is intended to behave in a manner analogous to biological neurons and, hence, consists of components termed "connections," "weights," and "units." The transformations are usually chosen to be simple thresholdlike operations. The mapping implemented by a neural net can be altered to more closely match a given pattern using any of a number of "learning" mechanisms, such as "error back-propagation."
ontogeny - The trajectory of changes occurring to an individual during its lifetime. See PHYLOGENY
* Ernst Haeckel's "biogenetic law" held that an individual's ontogeny recapitulates the phylogeny of its species, but this facile connection between the two forms of change is now discredited (Gould, 1977). phenocopy - A phenotypic variation that is caused by unusual environmental conditions and resembles the normal expression of a genotype other than its own (On-line Webster program).
* Schull (Chapter 3) sees Piaget's use (Chapter 19) of this term as a discipline-specific special case of organic selection. Piaget gave credit to both Baldwin and Waddington, and applied his term to the more general case in which genetic adaptedness comes to supplant or replace phenotypic adaptability, with little change in the observable phenotype. Plasticity would persist or would be enhanced due to its effect upon natural selection. Piaget thought of behavior as the "motor of evolution."
* A product of the convergence between a phenotypic variation and a genotypic variation which comes to take its place. The idea is that changes in the behavior of a particular organism create changes in the genetic makeup of that organism's descendants.
phenotype - 1: The detectable expression of the interaction of genotype and environment constituting the visible characters of an organism; 2: a group of organisms sharing a particular phenotype (On-line Webster program). See also Fox Keller and Lloyd (1992).
phylogeny - 1: The racial history of a kind of organism; 2: the evolution of a genetically related group of organisms as distinguished from the development of the individual organism; 3: the history or course of the development of something (On-line Webster program). See ONTOGENY
plasticity - 1: The quality or state of being plastic; esp.: capacity for being molded or altered; 2: the ability to retain a shape attained by pressure deformation; 3: the capacity of organisms with the same genotype to vary in developmental pattern, in phenotype, or in behavior according to varying environmental conditions (On-line Webster program). See also Fox Keller and Lloyd (1992).
* According to Schull (Chapter 3), plastic individuals are often purposive, goal-driven, and intelligent, and plasticity is a prerequisite for ontogenetic adaptation and learning. Further, if plastic individuals play a causal role in evolutionary processes, then the question arises whether the evolutionary process, itself is purposive, goal-driven, and/or intelligent.
population - A group of individuals of a species living in a certain area. Because they are the same species they can interbreed, although these breeding patterns might show local structure.
* A local population or "Mendelian" population is the local panmictic unit, for example, in "island" models of structured populations. A collection of local populations makes up the species.
* 1: The whole number of people or inhabitants in a country or region; 2: the act or process of populating; 3a: a body of persons having a quality or characteristic in common; b(1): the organisms inhabiting a particular area or biotope; (2): a group of interbreeding biotypes that represents the level of organization at which speciation begins; 4: a group of individual persons, objects, or items from which samples are taken for statistical measurement (On-line Webster program).
preadaptation - Traits initially evolved in response to one selective pressure but then adapted to a second. See EXAPTATION
preformation - Alternative (now discredited) account of development. See EPIGENESIS
psychology - 1: The science of mind and behavior; 2a: the mental or behavioral characteristics of an individual or group; b: the study of mind and behavior in relation to a particular field of knowledge or activity; 3: a treatise on psychology (On-line Webster program).
punishment - Something "bad" that happens to an organism after a behavior has occurred. Punishment cannot be avoided as it occurs after the behavior has already happened. See REINFORCEMENT, NEGATIVE
purposiveness - 1: Serving or effecting a useful function though not as a result of planning; 2: having or tending to fulfill a conscious purpose or design; purposeful (On-line Webster program).
* According to Schull (Chapter 3), Darwin's theory demonstrated that purely mechanistic processes (mutation and differential reproduction) could explain evolutionary adaptation without necessarily invoking "purposiveness." In this light, organismic behavior and psychological phenomena were seen as results rather than causes of evolutionary progress. reinforcement, negative - Any stimulus that, when removed following a response, increases the probability of the response (Hilgard, Atkinson, and Atkinson, 1979). See PUNISHMENT
* The modification of behavior based upon the omission of a stimulus (typically a punishment). reinforcement, positive - Any stimulus that, when removed following a response, decreases the probability of the response (Hilgard, Atkinson, and Atkinson, 1979). See REWARD
* The modification of behavior based upon the presentation of a stimulus (typically a reward). reward - Something "good" that happens to an organism after a behavior has occurred. Reward cannot be avoided as it occurs after the behavior has already happened. See REINFORCEMENT, POSITIVE
search, global - An algorithm that estimates extreme values of a function over a specified domain.
search, local - An algorithm that refines an initial estimate of a solution to a function. Specifically, an algorithm that identifies a local optimum near the initial estimate. Examples include the back-propagation method for neural networks (also called stochastic approximation); the conjugate gradient method, which exploits gradient information in some local neighborhood; and Solis-Wets, which does not require explicit gradient information.
selection - The process that chooses among variants, and that is itself shaped by those it selects for and against; in James' view, selection operates at nearly all levels, within and across individuals.
selection, intra- - Plasticity of an individual organism, changes within an individual in response to environment.
selection, organic - Baldwin's own term for any mechanism based on natural selection by which the adaptive achievements of individuals could influence the direction of evolution. Ontogenetic adaptation to local circumstances could prolong the race long enough for natural selection to accumulate genetic variations that would support the behaviors in question. Ontogenetic adjustments would be mediated by learning. The originally learned behavior would become innate. Credit for the origin of adaptations could thus be given to creative individuals, not to the heritable adaptations. A way of reconciling the Darwinian mechanism of evolution (natural selection) with a Lamarckian characterization of evolution (as purposive and individual-driven). See BALDWIN EFFECT, ECOLOGY OF MIND
* An individual's ability to adjust to its local environment in its own lifetime (ontogenetic adaptation) is valuable but limited: an organism's investment in adapting to one domain will tend to limit its capacity to adapt to other domains or in other directions. Since the time scale for ontogenetic adaptation is so much shorter than that of natural selection, adaptive strategies will often be "invented" first through individual plasticity rather than genetic mutation and population-genetic evolution. But once invented, the persistence of these adaptations may affect selection pressures and allow time for mutations such that population-genetic evolution occurs. Thus, the role of population-genetic evolution is often to "ratify" rather than to generate adaptations which set the course of evolution. Schull (Chapter 3) suggests that this general formulation developed by Bateson (Chapter 9) can subsume a number of discipline-specific special cases, such as "genetic assimilation," "canalization," "phenocopy," and the "Baldwin effect."
* A term originally associated (by Baldwin and his contemporaries) with what is now known as the "Baldwin effect." For example, Simpson (Chapter 8) essentially equates this term with his own characterization of "Baldwin effect."
selection, sexual - A particular form of selection. While natural selection refers to any differential reproductive success due to differences in survival ability, sexual selection refers to those effects due to differences in mate choice. See also Fox Keller and Lloyd (1992).
* Sexual selection occurs when females (typically) of a particular species tend to select mates according to some criterion (e.g., who has the biggest, most elaborate plumage or antlers), so males having those traits are more likely to be chosen by females as mates. The offspring of such matings tend to inherit the genes encoding the sexually selected trait and those encoding the preference for the sexually selected trait. The former will be expressed only in males, and the latter only in females. Fisher (1930) proposed that this process could result in a feedback loop between (typically) females' preference for a certain trait and the strength and frequency of that trait in males. As the frequency of females that prefer the trait increases, it becomes increasingly sexually advantageous for males to have it, which then causes the preference genes to increase further because of increased mating between females with the preference and males with the trait. Fisher (1930) termed this "runaway sexual selection." (Mitchell, 1996.)
* Some sociobiologists distinguish between "sexual selection" and "natural selection," as disjoint types. "We view sexual selection as arising from variance in mating success and natural selection as arising from variance in other components of fitness" (Arnold and Wade, 1984).
simulation - 1: The act or process of simulating; 2: a sham object: counterfeit; 3a: the imitative representation of the functioning of one system or process by means of the functioning of another ; b: examination of a problem often not subject to direct experimentation by means of a simulating device (Merriam-Webster's Collegiate Dictionary).
soma - All of an organism except the germ cells (On-line Webster program). See WEISMANN'S DOCTRINE
variation - Changes with a "germinal origin"; changes to the genotype.
*The existence of variants; the range over which natural selection can select. Weismann's Doctrine - The German biologist August Weismann distinguished between two substances that make up an organism: the soma, which comprises most body parts and organs, and the germ plasm, which contains the cells that give rise to the gametes and hence to progeny. Early in the development of an egg, the germ plasm becomes segregated from the soma, that is, from the cells that give rise to the rest of the body. This notion of a radical separation between germ and soma prompted Weismann to assert that inheritance of acquired characteristics was impossible, and it opened the way for his championship of natural selection as the only major process that would account for biological evolution.
References
Arnold, Stevan, and M. Wade. Evolution 38 (1984): 720-734.
Britannica On-line: http://www.eb.com/.
Fisher, R. A. The Genetical Theory of Natural Selection. Clarendon, 1930.
Fox Keller, E., and E. A. Lloyd, eds.
Keywords in Evolutionary Biology. Cambridge, MA: Harvard University Press, 1992.
Gould, S. J. Ontogeny and Phylogeny. Cambridge, MA: Harvard University Press, 1977.
Hartl, D. L. A Primer of Population Genetics. 2nd ed. Sinauer Associates, 1988.
Hilgard, E. R., R. L. Atkinson, and R. C. Atkinson. Introduction to Psychology. 7th ed. New York: Harcourt Brace Jovanovich, 1979.
Kawata, M., and Y. Toquenaga "From Artificial Individuals to Global Patterns." Trends in Ecology and Evolution 9(11) (1994): 417-421.
King, R. C. A Dictionary of Genetics. 4th ed. Oxford University Press, 1990.
Langton, C. G., C. Taylor, J. D. Farmer, and S. Rasmussen, eds. Artificial Life II. Santa Fe Institute Studies in the Sciences of Complexity, Proc. Vol. XIII. Reading, MA: Addison-Wesley, 1992.
Lovelock, J. E. Gaia: A New Look at Life on Earth. Oxford University Press, 1987.
Merriam-Webster's Collegiate Dictionary. 10th ed. Springfield, MA: Merriam-Webster, Inc., 1993.
Mitchell, M. An Introduction to Genetic Algorithms. Cambridge, MA: MIT Press, 1996.
On-line Webster program. Based on Webster's Seventh Dictionary. Springfield, MA: Merriam-Webster Inc., 1965.
Reiger, R., A. Michaelis, and M. M. Green. Glossary of Genetics. 5th ed. Berlin: Springer-Verlag, 1991.
Stern, C. "Selection for Subthreshold Differences and the Origin of Pseudo Exogenous Adaptations." Amer. Nat. 92 (1958): 313.
Waddington, C. H. "Canalization of Development and the Inheritance of Acquired Characters." Nature 150 (1942): 563-565.
Original text version © 1996 by Addison-Wesley Publishing Company; reprinted by permission.