Menu Table of Contents Previous

III. The Paradigmatic Context of Psychology Within Science

We ask "why" of change. The challenge is to generate a valid paradigmatic answer. We must see change within a coherent framework. The following section presents three dimensions within which activities trying to honestly understand change can be positioned. The terminology will be "equilibrium of adaptation" to emphasize that the change in a phenomenon can best be seen as the new equilibrium resulting from a change in the environment, and under new conditions the equilibrium can change.

The first dimension of the structure with which to understand psychology is the goal of that scientific activity. In the last chapter we saw that the types of questions which are asked and the types of answers which are accepted vary as a function of the goals of the researcher. Generally basic research, or research to understand nature, will be used as the illustration, but it is not the only possible goal.

The second dimension of the structure is the molarity of the paradigmatic context. Essentially the same figure can be used to characterize each of what has come to be known as the scientific disciplines. Only the molarity (and as we will subsequently see, the time scale) of the figure changes. If we represent some arbitrary change in the environment with a heavy line, then

would show the environment changing from a baseline (one set of relationships, contingencies or rules) to an altered state or some different set of contingencies, then subsequently returning to the original baseline state.

We can add the change in some arbitrary dependent variable to this diagram.

The dependent variable is to be seen as initially in some equilibrium with the environment. The environment then changes and the dependent variable re-equilibriates to the changed environmental state. As can be seen, this adaptation can be reversed. In fact to establish the existence of a causal relationship a reversal (or some control procedure) must be implemented.

The various paradigmatic contexts of scientific investigation can be grouped in terms of the molarity of the subject matter. Each context is distinct because its measures are distinctly different. The measures simply do not exist at levels higher or lower in molarity. (The phenomena obviously always exist; it's that each of our measures don't isolate or react to every change at every level of molarity.

This involves the adaptation of existence itself. The basic forces in the universe adapt as a function of interacting. (The remaining basic forces are the "environment" for the one under consideration.)

This involves the adaptation of atoms. For example, when brought into conjunction under the right conditions the atoms of sodium and chlorine adapt thereby forming salt.

This involves the adaptation of cells to changes in the environment. This adaptation can be seen across a variety of time scales. A cell may adapt over the short term to various environmental influences by secreting a substance (functioning); a cell may adapt over time (maturation), or a cell (more accurately, a DNA pool) may also adapt over a very long time span by changing into a cell with other characteristics (evolving).

This involves the adaptation of the behavior of a whole life form, (not the adaptation of a cell or the adaptation of an anatomical structure) to changes in the environment. If a measure of behavior is altered as a function of changes in the environmental conditions then the behavior is said to have adapted. A human coming to fish in a particular spot is an obvious example. Elaboration of the various time scales of adaptation for this level of molarity will be covered in the next section.

This involves the alteration in the proportion of a population responding to events in the environment (an exposure) as the result of some change in the contingencies established by that environment. This is a purely statistical property of a group and is not the behavior of a particular individual. The dependent measure could be, for example, that 12% of the population bought a product following an ad, but not that Harry or Mary bought the product. An example would be that a report of a syringe in a Pepsi can would extract a behavior Z from xx% of the American population in 1950, while in 1993 following exposure to changed contingencies in the culture, the same event extracts Z behavior from yy% of the population. Time scale groupings are applicable. Some cultural practices once established reverse only with a new culture, such as after a major social disruption. This is much like personality, once established it reverses only following relatively substantial disruption or only across progeny.

This adaptation is the adaptation of a system of groups, each containing homogeneous elements such as a group of humans and a group of trees, etc. The prototypical system is an ecosystem contained within a sealed glass sphere. A characteristic of a system is that it is virtually closed in that little or no input occurs to the whole system.

It is important to keep in mind that these conceptual categories do not exist in isolation. A specific member of a group which participates, is in fact, a behaving individual which is made up of cells which are in turn made up of atoms which are themselves made up of forces. All levels exist and function simultaneously much like the levels of a television set. The television receives signals and presents a picture. It changes with changes in the broadcast signal (stimulus-response relationships). It also changes as the result of changes made to its control knobs (reinforcement history). But none of these deny the fact that the television functions within a particular standard such as NTSC or PAL (culture), and is also made up of transistors and diodes (cells). And that the transistors are in turn are made up of semiconductors (chemicals), and that most basically the semiconductors are themselves made up of forces. (Note that at the most basic level a television and a human function for the same reasons; the fundamental forces are the same.) The fact is that the behavior of either a person or a TV is a combination of factors operating at both more molar and more molecular levels.

Alterations in the internal components or the molar context of either the television set or of the pigeon will have effects on the operation of the whole. But in both cases we are most typically only interested in the input/output relationships of the whole system. What does the pigeon, as a whole, do when the light is turned on? What does the TV set, as a whole, do when the channel 6 broadcast signal contains a red and blue cross hatch?

Successively more molecular, or reductionistic explanations could be viewed as the "inner" causative forces for the emergent properties of more molar phenomena. But that is only one meaning for cause. It is in fact more appropriate to see cause at the same level of analysis (how does what comes out change as a function of what knobs or contingencies we change). Recall the discussion on the difference between a reductionistic and a correlative explanation given in Chapter 1.

The following figures illustrate the successively more molecular and more molar organization for nature. All exist simultaneously.

This figure presents the various scientific activities as a function of the level of molarity of that paradigm and the goals of that paradigm.






Participational Adaptation

Common term






To Under-

why existence

why substances

why life

why behavior

why participation

To Solve

Applied Research

atomic weapons

fusion research




clinical research



cultural research

To Dispense


bomber pilot

chemical sales-man, gas station

country agricul-
tural agent,

clinical psycho-

law maker


Selection Process






The third dimension of the structure is the time scale of the effect of interest. If we use behavioral adaptation as the example, we would point out that not only does life exist – and not only do life forms behave – but they behave in different ways as the result of experience with the environment. This is what behavioral adaptation means.

Sensation, learning, developmental, and animal behavior are four seemingly distinct, nearly autonomous areas of inquiry involved in the analysis of behavior change. In point of fact, there is a fundamental continuity underlying these approaches. The appropriate perspective points out that these areas of inquiry are inextricably interdependent and vary along a closed continuum. This emphasis along with its implications provide a comfortable integration of available data, the production of fruitful research, and an understanding of behavior change.

The continuity underlying the seemingly disparate approaches to the analysis of behavior change are made more clear by viewing them in terms of the reversibility of their functional relationships. Time scale of adaptation varies from (1) behavior change which reverses almost instantaneously, such as the reporting of the presence and then the absence of a stimulus when a light comes on and then goes off (traditionally referred to as a "reflex" or a "sensation") or a behavioral output to a specific stimulus such as pecking while a key is green (traditionally referred to as a learned response or the performance of a discrimination). This class of adaptation is hereafter referred to as instantaneous adaptation; to (2) behavior change which can be reversed only after some training, but that can be reversed many times within the life of the organism (traditionally referred to as "learning," as in coming to peck the green light but not the red light). It is the acquisition and loss of the relationship pecking to green and not to red that is of interest and which defines this class of behavior. This class, hereafter, will be referred to as short-term adaptation); to (3) relationships which are virtually permanent within an individual, once established, but that are reversible across progeny, such as personality or intelligence (traditionally referred to as "developmental changes." Again, it is the time scale of the acquisition and loss of this relationship which is of interest and the defining characteristic of this class of behavior. This class, hereafter, will be referred to as medium-term adaptation). And finally to (4) long-term functional relationships which are reversible only across many generations (traditionally studied as species typical behavior by animal behaviorists). (Note that long term changes follow from variation and selection and do not in any way require the heritability of acquired characteristics). As before, it is the time scale of the acquisition and loss of this behavior which defines this behavior class.

If you give a person a piece of candy and they smile and salivate, what is the cause of the smile and salivation? At first glance, it is certainly the stimulus of being given the candy; but, further thought adds the realization that the person learned over the course of a few experiences that those things that are wrapped in the bright blue wrapper are sour and taste good. Further, you would accept that it was necessary for the person to develop a preference for extremely sour candy over the years. Finally, it is obvious that animals developed the tendency to salivate to acids millions of years ago over the course of many many generations.

The interdependence of these areas is obvious, but it is often overlooked. The fact that a functional relationship can be confounded comes as no surprise to anyone. However, functional relationships are often not conceptualized with respect to all of the conditions under which they were obtained. The presumed "failures" with general process learning theory in the 1970s are a good example of an insufficiently comprehensive paradigm. No behavior is the result of variables operating in only one time scale exclusive of all others. Behavior does not exist apart from perceptual, learning, developmental, and phylogenetic factors, and these must be held constant if their variation alters a functional relationship. If experience across other time scales is a source of confounding, then it is simply an interaction to be understood. It is not a failure of anything and should surprise no one.

The following integrating perspective provides the most useful context to view obtained data and provides the questions which most effectively advance our ability to predict behavior. This continuum also provides the necessary frame of reference for explanations of functional relationships which invoke mechanisms from other time scales, such as the suggestion that ethological or developmental variables account for an important portion of the variability obtained in a learning task.

The various types of behavioral adaptations studied by psychology can be categorized in terms of their time scale of adaptation. All behavioral adaptation can be illustrated with the same diagram -- only the time scale changes.

Because our topic is behavioral adaptation, the various time scale illustrations will all come from the behavioral adaptation level of molarity, but in principle any level of molarity could be used. Returning to the basic depiction of behavioral adaptation introduced in the previous section, the initial state of behavior could be said to be in equilibrium with the initial state of the environment.

When the environment changes, the behavior does not change at exactly the same time as the environment. There is a time lag between the environmental change and the behavioral shift. This lag is called hysteresis. Subsequently, the behavior comes into equilibrium with the environment. When the environment returns to its original state, the behavior re-equilibriates.

The above figure can illustrate the independent variable / dependent variable relationship in any area of psychology, only the time scale changes. It is important to note that the following four groupings refer to the time course of the change in a functional relationship, not the state of behavior before and after the independent variable changes. For example, in the following illustrations it is not pecking (peck / no peck) that changes but rather pecking the green key (peck green, don't peck red / peck red, don't peck green) that changes.

The measured behavior is a behavior difference to a stimulus difference. For example, the light is off and the person is quiet; the light goes on and the person says "I see it."
  1. hysteresis in milliseconds - seconds range
  2. a stimulus change immediately controls a response
  3. previously known as a sensation, a learned reaction, a reflex, an instinctive response, etc.
  4. a typical functional relationship - recognition
  5. a typical research topic - signal detection

These adaptations occur immediately following the stimulus presentation, as soon as the organism “experiences” them. For example, presenting a green light (following key peck training) is followed by pecks to the key. It is a behavior change as soon as the stimulus changes. The appropriate response to the stimulus has been selected by the past consequences of that behavior (ontogenetic or phylogenetic).

Descriptive unit of analysis:
Explanatory framework:
  1. Why did the organism respond? Because the stimulus changed.
  2. The empirical theory of signal detection.

The measured behavior is a different instantaneous adaptation to a stimulus change as the result of exposure to a contingency change.
  1. hysteresis in seconds - days range
  2. different contingencies come to control different behaviors
  3. previously known as learning
  4. a typical functional relationship - discrimination
  5. a typical research topic - matching

These adaptations take time to occur. It is a change from a specific behavior difference to a stimulus change to some new behavior difference to the same stimulus difference. For example, a bell but not white noise could control salivation. Subsequently, white noise but not a bell could elicit it. This speed of adaptation (seconds to days) is optimal when the demands of the environment change many times within the life of the individual.

Descriptive unit of analysis:
Explanatory framework:
Note that the reductionistic system which extracts invariant “relationships” from the environment (learning system) is essentially the same as the one which extracts invariant “properties” from the environment (perceptual system).

The measured behavior is a change in the equilibrium established by short-term adaptation as the result of correlations which extend across or are shared by multiple contingencies.
  1. hysteresis in days - years range
  2. the ability of a stimulus to control a response is altered for
    the individual for virtually the rest of the individual's life
  3. previously known as developmental psychology
  4. a typical functional relationship - disposition
  5. a typical research topic - personality or memory

Descriptive unit of analysis:

Explanatory framework:

It is important to realize that designs used to prove causation in short term (recover baseline within same subject) may not be appropriate for medium term adaptation. Proving causation for this time scale may require substantively different research designs. For example if we wish to study the acquisition of bicycle riding we may never be able to recover baseline within that same individual. We may never be able to make that subject naive again, but rather will only be able to recover baseline across progeny. The factors which control this time scale of adaptation are not well understood.

The measured behavior is a change in the behavior difference to a stimulus difference as the result of differential reproductive success rather than ontogenetic experience.
  1. hysteresis in years - millennia range (actually generations)
  2. the ability of a stimulus difference to control a response difference is altered for the species
  3. previously known as animal behavior or comparative psychology
  4. a typical functional relationship - instinct
  5. a typical research topic - migration

Descriptive unit of analysis:
Explanatory framework:
Darwin (1859) suggested that the diversity of life forms could be accounted for by genetic variation and differential reproductive success caused by natural contigencies. If a more extreme aspect of some trait reproduces more than alternative forms, then the better reproducing form will come to predominate even to the extent of creating a new species. Alternatively stated, given variation in functional relationships involving some behavior to the environment, if more and more extreme versions of one of those variants provides relative reproductive success, then the eventual result is that all individuals of that species will possess that type of “adaptive strategy” to the environmental event.

There are two important implications of the evolution of behavior. First, the functional relationship can change as the result of the selection of individuals--for example the time or target of migration. Secondly, various behavioral adaptations to environmental contingencies "seek their most appropriate time scale." The reaction to light is an instantaneous eye blink, the reaction to key pecks to red being followed by food is an increase in the tendency to peck red. This only makes sense. Very little flexibility is needed when "deciding" what to do to bright light in the eye, whereas substantial ontogenetic information is necessary to "decide" what to do when a red light comes on. But little can be “pre-learned” before the individual is born or in terms of "common knowledge," with respect to what to do when the red light comes on. Evolution can, therefore, be seen as: (a) as a process whereby long-term functional relationships can be altered and reversed, (b) a potential confound when comparing functional relationships across species (e.g., a moth and a roach react to light differently), and (c) an explanation for how time scales of adaptation came about.

Note that this refers to a behavior not a structure. In this sense, pigs (as a source of DNA) can be taught to fly (a behavior) and then baseline recovered. Of course, many changes will take place in the structure of the organisms along the way but presumably physical changes also take place when we learn to play the piano. If we play now and didn't before, something had to change. If it is not in nature, then what and where did the change take place? We need not actually discover the exact body change in order to teach someone how to play the piano or to determine what variables increase or decrease the time it takes to teach someone to play the piano. In fact, it would be unlikely that we would even care what the body change was knowing the body change is certainly unnecessary for an explanation of the behavior. But nevertheless something in nature must have changed. This doesn't say however that it in principle could not help to know what structures were changing or how. It may be that neural network theory and pet scans will dramatically help us in some way to develop the set of functional relationships which describe learning. Knowing what made the pig offspring lighter weight may help us to teach it to fly by helping us to quickly shape the behavior.

Learning speed may reflect underlying genetic predisposition much like sensitivity to heat shock revealed underlying genetic pre-adaptation (Waddington). See also Staddon (1983, page 11).

Because evolutionary selection must be through genetic selection (genotype), we must have a reliable way to select. We must find a way to select genotype based on something we can detect. Speed of learning, if it is related to genotype, provides the needed bridge.

A second important issue is that after recovering the non-flying pig baseline we may not have a pig which will breed with the original pig species. Unless we exert special control to shape the DNA along that line also (needless to say recovering a true breeding pig baseline would be more difficult than simply generating a large land mammal good for bacon), but that is structure not behavior.

All of these “speeds” of adaptation contribute to the adaptation of the organism. In addition, they interact and through that interaction, the time scale across which an adaptation occurs to a particular class of problems can change. From the genetically provided complement of functional relationships (long-term), the behaving organism can learn (short-term) to behave differently in some way. Speed of learning could function to isolate a subset of the population for reproductive success. If that learned behavior provides reproductive success, then faster variations in learning “ability” will be selected and given that learning speed is related to genetic predisposition, the genetically provided complement of learning speed will be modified. When what was previously a short-term functional relationship is in the gene pool as an instinct, then speed of learning can again be “used” to select even more (or less) extreme versions of that characteristic or even other details of the adaptation altogether. As a result, stable problems which required the same learned solutions can come to be instincts. On the other hand, problems which can no longer be solved with the same adaptation will no longer have instinctual solutions but rather will require learned solutions.

This ability to select intermediate forms of time scale of adaptation dramatically simplifies the development of the optimum time scale of adaptation for each problem posed by the environment. Enduring stable problems in the environment produces the correct adaptation process through a process like shaping. In the same way you cannot easily establish an FR 100 on the first attempt in short-term adaptation, but you can easily establish it if you’re able to shape successive approximations of the behavior. So does learning provide for successive approximations or the selection of partial genetic solutions which are closer to the desired goal. In both ontogenetic and phylogenetic shaping cases the mean of the distribution is incrementally shifted. This provides for the straightforward emergence and selection of more extreme instances in the future.

The following depiction of the net effect of all the time scales acting together uses complex waves and Fourier analysis as its metaphor. Fourier's theorem states that any wave form can be expressed as a sum of sinusoidal components. In the same way as a complex sound can be seen as a combination of various frequencies, the complex behavior of an individual can be seen as the result of contingencies operating at a variety of time scales.

An advantage of this metaphor is that it makes it clear that we cannot necessarily attribute a particular increment on the y axis to a particular component frequency. Rather it is the net change of all the factors together which determine behavior. Some factors could be increasing and some could be decreasing at any point in the behavior stream.

A detailed analysis of the time scales of behavioral adaptation is presented on the following pages. As can be seen, it compares the various subspecialties of Psychology. Each is a subparadigm. Given that where Psychology fits within science is understood, and given that where the subspecialty fits within Psychology is understood, then little appeal need ever be made to other levels of molarity (e.g., biology or sociology) or time scales (e.g., perception or developmental for learning).

The subspecialties within the disciplines one level of molarity above and below Psychology are provided in the figure directly below for perspective.

pathway responsive- ness is controlled by activity
behavior is a function
of the environment
a proportion of the members of a group will respond to an event

Descriptive unit of analysis

I - IV are time scales
change in pathway responsiveness
associated with
I. the occurrence of
an event
II. the contingencies
III. commonalities in
IV. genetic selection
a change in behavior as the result of events in the environment
I. the occurrence of
an event
II. the contingencies
III. commonalities in
IV. genetic selection
a change in the pro-
portion of a group
reacting as a result of
I. the occurrence of
an event
II. the contingencies
III. commonalities in
IV. genetic selection

Explanatory Perspective
why did the nervous system react that

because the environ-mental history was
sufficient to modify synaptic transmission

(reductionistic would be chemical explanation)
why did behavior occur

because the environ-mental history was sufficient to modify the behavior

(reductionistic would be neural/hormonal explanation)
why did that proportion of the group do that?

because the environ-mental history was sufficient to extract the behavior

(reductionistic would be behavioral explanation)
Of what is "why?" asked
why does the pathway change? Of what is the pathway a
function of?
why does behavior adapt? Of what is behavior a function of?
why does some pro- portion participate? Of what is participation a function of?
which factors change connections? how and by how much?
which factors change adaptation? how and by how much?
which factors change participation? how and by how much?

Behavioral Adaptation

milli - sec
sec - days
days - years
years - eons


an external event can change behavior

the behavioral repertoire can be changed by environmental contingencies

enduring character-
istic ways of
responding can be established by exposure to common-
alities in correlations

"rules" memory organi
equiva- zation of
llences behavior

genetic selection can establish a characteristic behavior to an environmental event

Descriptive Unit of Analysis

must be input
output relation-

a change in behavior associated with changes in the environment

(reverses with stimulus)

a change in behavior repertoire associated with exposure to some nonrandom relationship in the

(reverses with some contrary contingen- cy)

a change in the characteristic way of responding attribu- table to commonalities in correlations

predisposition enduring

(virtually life-long but does not affect offspring)

a change in behavior attributable to genetic selection

(breeds in and breeds out)

Explanatory Perspective

why did organism respond?

because the stimulus

why did organism respond?

because of its rein-
forcement history

why did organism respond?

because it was ex-
posed to commonal-
ities in correlations

why did organism respond?

because its ancestors that did, obtained differential reproductive success

Of What is "Why" Asked

why does an organism react to a stimulus?

of what is
a function of?

why does an organism respond differently following some contingencies?

of what is
a function of?

why does the organism consistent- ly respond that way?
why does exposure to commonalities in correlations result in characteristic ways of responding?

of what is
a function of?

why does an organism respond in "species-typical" ways?
why does genetic selection produce different behavior?

of what is
a function of?


which factors change reception?; how, and by how much?

which factors change learning?; how, and by how much?

which factors change dispositions?; how, and by how much?

which factors change instincts?; how, and by how much?

The two major concepts of adaptation (level of molarity and time scale of adaptation) can be represented with the two "axes" of a sphere.

Measures can be more or less molar. Distance from the center of a sphere could conveniently represent molarity of measurement.

More physiological variables toward the center and more molar variables more distant from the center. Successive layers of the "onion" represent molarity of paradigmatic context. The sphere depicted in the initial figure is actually one concentric layer of a solid. The molarity dimension depicts that physiological variables underlie all behavior and that ultimately physiological (or chemical or existential) mechanisms mediate all behavior change. Additionally, the behavior change of an individual is how only one individual of a group behaves.

A reductionistic explanation, or an explanation which appeals to a lower paradigmatic context would be seen as appealing to an inner sphere for explanation with this spherical model. With this model, correlative explanations would be seen as appealing to factors on the same surface at the same level as the original question. A point on the surface of the sphere could then represent the time scale of reversibility for a behavior at a particular level of molarity.

The behavior change of interest can be across any of a number of time scales. Points on the surface of the sphere could represent the time scale dimension. Any specific time scale of adaptation within some particular paradigmatic context could be represented as a point on the surface of the sphere, at some distance from the center (some particular layer of the onion).

For example, the surface of the sphere at the level of behavioral adaptation could depict time scale of reversibility of behavioral phenomena.

The surface could be partitioned into four equal quadrants, like the four sides of a pyramid mapped onto a sphere (imagine a pyramid tin can exploded out). The left figure shows the covering of the pyramid flattened out.) The spherical representation provides all time scales as contiguous or interacting. For example a functional relationship such as discrimination would be depicted on the surface labeled "short" and could be drastically changed by changing any of the other three time scales, just as topologically changing Cartesian space changes a form on that space. A figure drawn on a rubber sheet changes as the sides of the sheet are distorted or changed by pulling. An advantage of this conceptualization is that it provides orthogonal axes all of which could be continuous rather than discrete and any of which could affect a relationship on any other.

It is again important to keep in mind however, that these categories do not exist in isolation. A behaving pigeon is alive, is chemicals, is matter, and has an evolutionary, a developmental, a reinforcement, and a perceptual "historical" context. These multiple coexisting determinants are analogous to the determinants of a point in space. A point does not exist in the x-axis alone. It is simultaneously in the x, the y, the z-axis, and the temporal plane.

With this spatial model several important ramifications of characterizing a particular functional relationship in terms of its time scale of reversibility become more clear. First, the continuum is closed rather than being linear. It is not appropriate to view instantaneous relationships at one end and long term relationships at the other extreme, related only through all intermediate stages. All are equally interdependent with one another. No one time scale exists in isolation from any of the others. Second, the meaning of a functional relationship in any time scale is with respect to the context or boundaries established by the other class of variables. Clearly a demonstrated relationship at any one time scale can be changed by changing variables in any one of the other three time scales. Third, category boundaries are actually a matter of convenience. There is no sharp demarcation between classes of functional relationships. There is no real reason to maintain a hard distinction between enduring perceptual variables and brief learning variables. On the contrary, there is every reason to examine manipulations typically reserved for one time scale by other research areas.

Traditionally each area of investigation has had its own premises, descriptive units of analysis, and explanatory framework. These have been reworded in the context of the new paradigm to emphasize their continuity. This will provide for psychology what Maxwell's equation did for electromagnetic radiation.
A three-dimensional structure within which to view nature results if we combine the three previous dimensions. In this case, the "goals" dimension has been rotated 90 degrees to the z axis. Because we are primarily dealing with basic research, we need not emphasize goals other than those of basic research.

Menu Table of Contents Next

Send comments/criticisms/speculations to

Date Last Reviewed: November 17, 2002