Thinking Clearly About Concepts in Behavioral Science

Roger K. Thomas

 � 2001 All rights are reserved.

Background Information

 

External observable antecedents to behavior (EOAs) are environmental sources of physical energy (heat, light, sound, etc.) that affect an organism's senses.  EOAs may be associated with objects as emitted, reflected, and transduced energy that enables an organism to observe (identify) such entities which humans might specify, for example, as being a red wooden block, a gray granite rock,a tree, a bird, a person, a ship, a cloud, the song of a bird, the smell of an apple, etc.   EOAs are often referred to as stimuli.  With respect to an organism, some EOAs may be causes or correlates of some external observable consequents (EOCs; see next paragraph)) manifested by that organism, namely, behavioral or physiological responses.  In an experimental context, the manipulated EOAs are known as independent variables.  EOAs are observable and, in principle, measurable. [Note: Because an organism is always "moving through time" the EOCs of a given moment may become part of the next moment's EOAs.]  

 

External observable consequents (EOCs) are also manifested (that is, detectable) as environmental sources of energy such as those described above.  EOCs are the potentially measurable, physical movements, sound, etc. that an organism makes that we usually refer to as behavior.  EOCs may also include directly observable physiological reactions such as but not limited to vasodilatation (e.g., blushing), vasoconstriction (becoming pale), and sweating. EOCs are also known as responses or, in some cases, effects or correlates of EOAs.   In an experiment, the most relevant EOCs are the dependent variables.

 

Internal observable antecedents (IOAs) are direct extensions of EOAs that occur as physical events within the organism.  They are the potentially observable (aided by instruments) and measurable manifestations of the EOAs that arise in the organism's sense organs and nervous system.  IOAs are typically measured as electrical (e.g., action potentials, evoked potentials, graded potentials) or chemical changes within the sense organs and the nervous system.  Essentially, the external environmental energy that impinged upon an organism is transduced into chemical and electrical energy within the organism that can be observed and measured.  By definition here, IOA's are on the sensation and perception side of the manifestations of the environmental energy that stimulated the organism.

 

Internal observable consequents (IOCs) are the precursors of EOCs.  IOCs are the electrical and chemical changes within an organism that mark the beginning of a behavioral or physiological response.  For example, before an action such as writing (the act of writing, itself, is an  EOC) occurs, it began as an action in the brain.  Physiological responses that are interpreted as behavior, e.g., blushing, begin as neurophysiological actions in the brain (e.g., actions associated with awareness that you have something to blush about).

In the diagram above, /??/ indicates that there is uncertainty in our present state of knowledge about where EOAs and IOAs end and where IOCs and EOCs begin.  Where is the interface?  Are there structures or systems of structures in the brain that provide the interface (e.g., "association cortex" postulated in the 19th century) or is the interface at the cellular level?  

 

Additionally complicating the diagram above are two factors that must be added.  To prevent the above diagram from becoming overwhelmingly complicated "at first sight," these two additional factors are represented in the diagram below.  For a fuller understanding, one must "conceptually merge" the diagram above and the diagram below.  The first of the additional factors is that the organism is always "moving through time."  Hence, it often is likely to be the case that some consequents at a given moment may become part of the antecedents of the following moment which affect (or effect) subsequent consequents.  Second, it is likely in almost every situation that the antecedents and internal consequents of a moment will activate physical-chemical events ("memories") that are specific to the ongoing situation and activities.  Therefore, an interactive cycle of physical-chemical activities within the brain that underlie sensory, effector, and memory processing will occur. 

Most, if not all, concepts in behavioral science are the Linking Concepts indicated in first diagram and referred to as Conceptual Correlates in the second diagram, especially the Intervening Variables.  These conceptual correlates which will now be discussed are not real in the sense of being independent, isomorphic sources of environmental energy.  Therefore, they are neither observable nor measurable and cannot be either antecedents or consequents (including, of course, that they cannot be causes or effects).

 

Conceptual Terms in Psychology and Behavioral Neuroscience

 

 Most psychological concepts or terms are not EOAs, IOAs, IOCs, or EOCs. Most concepts in psychology represent something that is not observable or measurable via sense perception or with the use of special instruments.  Rather most concepts in psychology are terms that summarize, abbreviate, or link observables represented among the EOAs, IOAs, IOCs, and EOCs.  For example, "fear" cannot be observed, because fear does not exist as an isomorphic source of observable environmental energy.  What we typically observe are certain EOAs that we have learned to associate with certain EOCs, and we have learned to label the observed EOA/EOC combination as "fear."  Of course, we don't necessarily have to experience the frightening EOAs directly, we may have learned and, therefore, remembered what the EOCs associated with such stimuli might be.  For example, we do not have to have been bitten by a growling dog to fear a growling dog.  We can learn that from hearing or reading about others' experiences, seeing motion pictures, etc.

 

 In psychology, two different kinds of concepts were distinguished historically, intervening variables and hypothetical constructs.  The most important reference to this is:

 

 

 We use intervening variables when we have no reason to believe that the concept has an independent, isomorphic existence as a source of environmental energy; that is, we likely believe from the outset that the concept is fictitious in this sense.  For example, we never expect to observe directly an isomorphic source of physical energy called "fear."  We expect only to use fear as a term to summarize, abbreviate, or link certain observable antecedents and consequents that we are linked with the term fear. We invent intervening variables (or, most commonly, borrow terms from traditional language) as opposed to inferring the existence of physical sources of energy in conjunction with that which we name as an intervening variable.

 

We use hypothetical constructs when we infer that something physical is being named by the concept, but we do not yet know what it is.  An historical example is the gene.  When the term, gene, was proposed by Wilhelm Johannsen in 1909 to represent the hypothetical Mendelian elements, it was long before a gene (i.e., a specifiable segment of DNA had been shown to exist). Perhaps, the best potential example of a hypothetical construct in psychology is the "engram,"a term proposed by Karl Lashley for the hypothetical "memory unit."  Some believe it is possible that someday a memory molecule or class of memory molecules (e.g., analogous to the gene/DNA relationship) will be observed to be isomorphic with specific remembered events.  Can you think of another possible example of a hypothetical construct in psychology?� I cannot. 

 

Levels of Concepts

 

Level I. The most fundamental or basic level of a psychological concept is illustrated with the following example, where the concept "fear" may be used to describe the psychological state attributed to the rat in the example.   

 

 

What we observe are the EOA(s) and EOC(s).  With appropriate methods, we might also observe within the rat's sense organs and nervous system IOAs initiated by the buzzer and the IOCs that are direct precursors of the EOCs.  However, we do not observe or expect to be able to observe an isomorphic source of energy that can be detected as fear.  The concept of fear was invented (or borrowed from traditional language) to abbreviate, summarize, or link the antecedents and consequents that we did observe.  This is, of course, similar to being an operational definition of fear. 

 

At this lowest level, and by McCorquodale's and Meehl's strict definition, the meaning of the concept, fear, is limited to and should not be generalized beyond the specifically observed antecedents and consequents.

 

"Joy," "anger," and "short-term memory" are other examples of concepts or processes whose meaning at level I is limited to the specific observable antecedents and consequents that are being used to define the concept. 

 

Higher Levels Representing More General or Abstract Concepts

 

Level II.  At level II, we acknowledge that a concept like fear, for example, may be defined in other contexts which have sufficient similarities among many of the antecedents and consequents across the individual definitions of fear to make it useful to acknowledge that fear can have meaning that generalizes beyond such individual definitions.  Thus, we might find it useful to accept and use a more general definition of fear that acknowledges these similarities.  In day-to-day practice, we are unlikely to compile a list of the similarities except, perhaps, to express a general or informal acknowledgment that the antecedents associated with "fear" are usually potentially harmful or lethal and that the consequents usually involve responses associated with a greater-than-baseline activation of the autonomic nervous system and neuroendocrine system. We might refer to this new, more general definition of fear as being a superordinate or higher-level definition of fear.

 

In principle, a superordinate (or Level II) definition of fear is firmly based on level I definitions of fear.

Level III.  At an even more general or abstract level, we recognize that some concepts that are seemingly different at levels I and II, such as fear and joy, nevertheless, have similarities that link them conceptually, such as, significant involvement of the autonomic nervous system and the endocrine system.  To acknowledge the similar or overlapping relationships that such concepts might have, we find yet higher order category concepts such as "emotion" to be useful to organize them into a common conceptual framework.

Another example of a Level III concept might be perception.  For example, we may begin to see commonality in identifying an object by sight and by touch.  We may also associate objects with particular sounds, such as the songs of a bird, or particular smells, such as the odor of an apple.  Therefore, the perception of an object may involve the integration of various sensory experiences.   Other examples of higher order concepts are motivation, general memory, learning, conceptualization, etc.  Concepts at level III are used to integrate or categorize processes that may differ significantly at lower levels of analysis but may be meaningfully related at a higher level of definition and analysis.

 

As it was with Level II concepts, concepts at this Level III, in principle, must be reducible and ultimately definable in terms of EOAs, IOAs, IOCs, and EOCs.  

 

Level IV and beyond?  Examples here might be concepts like cognition, knowing, and thinking which, neuroscientifically speaking, refer to neural processes that represent combinations of other processes, such as, attention, learning, memory, perception, etc.   Also relevant at all levels are the neural mechanisms that enable sequencing and coordination of the constituent processes in the nervous system, etc.  

 

Historically, the concept of consciousness seems to have been considered to be the ultimate psychological and neuroscientific concept.  Consciousness, defined from a neuroscientific standpoint, is the summation of processes that might be best observed, measured, described, and defined at the lower levels (see above).

 

One eminent neuroscientist, Karl Pribram, has discussed different kinds of consciousness.  First, he discussed objective consciousness, episodic consciousness, and narrative consciousness which appear to be at parallel levels.  Pribram also discussed transcendental consciousness which he seems to regard as being beyond the first three.  Pribram's definitions of the first three seem to be anchored in observable and measurable events, although it is not clear (to me) whether Pribram believes that they have emergent properties. 

 

The issue of emergent properties is scientifically troublesome.  Please see my arguments regarding the Hazards of "Emergentism" in Psychology in History & Theory of Psychology Eprint Archive,