Behavioral Disciplines and Activities

Abstract

Behavioral Disciplines and Activities is the umbrella category under which the U.S. National Library of Medicine's Medical Subject Headings (MeSH) file the sciences of behavior together with the activities those sciences support. It is not a theory but an indexing header (tree number F04) that groups the disciplines that study behavior — the behavioral sciences and their psychological techniques — alongside the applied practices that change it, of which psychotherapy is the largest. The grouping encodes a distinction cognitive psychology has argued over since its founding: behavior as an object of cumulative science versus behavior as something a practitioner intervenes upon. This article treats the header as a lens on that science-and-application divide, tracing the term's origins, Cronbach's two disciplines, the scientist-practitioner model meant to bridge them, and the modern debate over whether the behavioral sciences cohere as a science.

Keywords: behavioral sciences, two disciplines, scientist-practitioner model

What Behavioral Disciplines and Activities Is

Behavioral Disciplines and Activities is one of the sixteen top-level branches of the MeSH tree, occupying the whole of category F04. MeSH is a controlled indexing vocabulary, so the heading is a filing decision rather than a claim about nature: it is the drawer into which the National Library of Medicine sorts every heading that names either a discipline devoted to behavior or an activity a behavioral professional performs. Its immediate children run from Behavior Control through Behavioral Sciences, Mental Health Services, Personality Assessment, Psychotherapy, and Schizophrenic Psychology — a list that mixes fields of study with clinical and assessment practices in a single tier.

That mixture is the point. The heading yokes together two things cognitive psychology treats as conceptually distinct: a discipline, which is an organized body of knowledge with its own methods and standards of evidence, and an activity, which is something a person trained in that discipline does to or with another person. The behavioral sciences aim at general, cumulative knowledge of how organisms perceive, learn, decide, and act; the activities — counseling, testing, therapy, commitment — apply fragments of that knowledge to a particular case whose outcome is judged by whether the person improved, not by whether a theory advanced. The category's usefulness for a cognitive psychologist is that it makes the science-and-application boundary visible in a single place, and the tensions along that boundary are the subject of this article.

The heading sits above, and should not be confused with, the narrower descriptor Behavioral Sciences (its own child, F04.096). The umbrella is broader than any one field: it is the administrative parent of the disciplines and their practices, whereas behavioral sciences names only the family of studying disciplines. Because F04 is a root of the tree, it has no parent heading of its own; nothing in MeSH classifies it as a kind of something more general.

Types of Behavioral Disciplines and Activities

MeSH divides Behavioral Disciplines and Activities into ten direct children, listed in Table 1. They are not a taxonomy in the biological sense — the children are grouped for retrieval, not because they share a defining essence — and they cut across the discipline/activity distinction rather than respecting it: some name a field (Behavioral Sciences), some a class of instrument (Psychological Tests), and some a practice (Psychotherapy). These axes are largely orthogonal: a single case can be filed under a testing activity, a therapeutic activity, and the science that grounds both. Only the children that already have a dedicated article on this site are linked; the remainder are named as MeSH lists them. Because the header is an indexing classification and not a claim about how behavior is really carved up, the presence of a child here reflects the Library of Medicine's filing conventions, not a settled scientific ontology.

Table 1
MeSH child descriptors of Behavioral Disciplines and Activities (F04)

Each descriptor below is a direct child of the F04 header. Decide whether it names a discipline that studies behavior or an activity that changes it. The header deliberately mixes both.

The classification is not always sharp — personality assessment is both a measurement science and a clinical activity — which is exactly why MeSH files all ten under one umbrella rather than splitting science from application.

Origins of the Term

Behavioral science is a mid-twentieth-century coinage, and the umbrella heading inherits its ambiguity. James Grier Miller is generally credited with giving the term its modern currency around 1949-1955, seeking a label for an interdisciplinary science of behavior that would sidestep the funding and political connotations of social science during the early Cold War (#ref-miller-1955). The Ford Foundation institutionalized the phrase: its Behavioral Sciences Program, directed by Bernard Berelson, spent much of the 1950s funding work under the new banner and, in doing so, fixed the term in the vocabulary of American research administration (#ref-solovey-2013). Berelson later co-authored a vast inventory of what the young science claimed to know, an attempt to show that behavioral science named a real, cumulative body of findings rather than a funding convenience (#ref-berelson-1964).

The political history matters because it explains the header's shape. The behavioral sciences were assembled less by intellectual convergence than by patronage: a foundation and, later, the National Science Foundation needed a category, and the category then acquired disciplines. Mark Solovey's histories trace how the search for public funding repeatedly forced the behavioral and social sciences to present themselves as hard, value-neutral science to survive in a hostile appropriations climate (#ref-solovey-2020). MeSH's F04 grouping is a late administrative echo of that settlement: it files the disciplines and their activities together because that is how the institutions that paid for them came to think of them.

The Two Disciplines

The deepest fault line inside the header is the one Lee J. Cronbach named in his 1957 presidential address to the American Psychological Association: scientific psychology, he argued, had split into two disciplines that barely spoke to each other (#ref-cronbach-1957). The experimental discipline manipulates a variable, holds everything else constant, and asks whether the average response differs between conditions; individual differences are noise to be minimized. The correlational discipline treats those individual differences as the signal, studying how people who vary naturally on one trait vary on another. One tradition builds knowledge by creating variance; the other by measuring variance that is already there.

The division is not merely methodological — it partitions the whole enterprise. The experimental tradition gave rise to the study of learning, perception, and cognition; the correlational tradition to psychometrics, personality, and the psychological tests that the F04 header also files. Cronbach's plea was that the two disciplines be reunited by studying how experimental manipulations interact with individual differences — an aptitude-by-treatment interaction — but the split he described still organizes graduate training, journals, and the sub-fields the umbrella heading collects. The same empirical effect can be expressed in either currency, and the worked example below shows how a single dataset yields both an experimental effect size and a correlational coefficient.

The experimental discipline reports a standardized mean difference (Cohen's d); the correlational discipline reports the same effect as a point-biserial correlation (r). Move the controls and watch both currencies update from the identical numbers.

At the default values (108, 100, SD 15) the effect is d = 0.53 or, equivalently, r = 0.26. Two traditions, two numbers, one finding — Cronbach's point that the disciplines describe the same reality in different vocabularies.

From Science to Practice

The activities half of the header — therapy, assessment, counseling, commitment — raises a second question: how should a science of behavior connect to a profession that practices on it? The dominant answer in clinical psychology is the scientist-practitioner (or Boulder) model, adopted at a 1949 training conference in Boulder, Colorado, which held that clinicians should be trained both to consume and to produce research, so that practice is disciplined by evidence and research is informed by the clinic (#ref-baker-2000). The model is an explicit attempt to keep the disciplines and the activities of F04 in contact rather than letting practice drift free of science.

Whether it succeeds is contested. The largest applied activity under the header, psychotherapy, has a robust evidence base showing that it works, yet decades of comparative trials find that the specific techniques distinguishing one therapy from another explain far less of the outcome than the factors all therapies share — the alliance, the expectation of help, the therapist (#ref-wampold-2015). That finding is awkward for a strict scientist-practitioner picture in which the practitioner delivers a scientifically validated technique: it suggests the science and the activity are joined more loosely than the model assumes, and that the effective ingredients of an activity may not be the ones its founding discipline named.

Outcome research partitions a psychotherapy's benefit into common factors shared by all therapies and the specific technique that names the brand. Drag the split and see how the picture changes; comparative trials tend to place most of the effect on the common-factors side.

The scientist-practitioner model assumes the practitioner delivers a validated technique. If the common factors dominate, the science and the activity are joined more loosely than that model supposes — the effective ingredient may not be the one the founding discipline named.

Worked Example

Consider a single study that both of Cronbach's disciplines could claim. A researcher randomizes participants to a training condition or a control condition and measures a cognitive score. The training group scores a mean of 108, the control group 100, with a pooled standard deviation of 15.

The experimental discipline reports a standardized mean difference — Cohen's d:

d = (108 − 100) / 15 = 8 / 15 = 0.53.

The correlational discipline treats the same data as an association between a grouping variable and the score, reporting a point-biserial correlation. For two equal groups, r = d / √(d² + 4):

r = 0.53 / √(0.53² + 4) = 0.53 / √4.284 = 0.53 / 2.070 = 0.26.

The two numbers describe one and the same effect. The correlational figure, squared, says the grouping accounts for r² ≈ 0.066, or about 6.6%, of the variance in scores — a translation that lets an effect discovered in the experimental tradition be compared with effects reported as correlations in the correlational tradition. Neither number is more true than the other; they are two dialects for one finding, which is exactly Cronbach's point that the disciplines describe overlapping reality in non-overlapping vocabularies. The demo above lets these means and the standard deviation be varied so the d-and-r pair can be read off any configuration.

Discussion

Grouping the behavioral disciplines with their activities under one header is administratively convenient and conceptually loaded. It presumes that the sciences of behavior and the practices they license belong together, and much of the twentieth-century history of psychology is a quarrel over that presumption. Sigmund Koch argued that psychology had never been a single science and never would be — that psychology names a loose federation of psychological studies held together by an institutional label rather than a shared subject matter or method (#ref-koch-1993). On Koch's view the F04 header is exactly the kind of false unity he warned against: a filing category mistaken for a discipline.

The opposing impulse is unificationist. E. O. Wilson's consilience thesis held that the behavioral and social sciences would eventually be woven into a single explanatory fabric reaching down to biology (#ref-wilson-1998), and within psychology, Gregg Henriques proposed a Tree of Knowledge system meant to give the field the theoretical spine it lacks (#ref-henriques-2003). Between Koch's pluralism and Wilson's consilience sits the everyday reality the header encodes: a set of disciplines and activities that share funding streams, journals, and training pipelines without yet sharing a theory. Whether that shared infrastructure will ever become a shared science is the question the next section takes up.

Current Directions

The most active contemporary debate about the behavioral disciplines is whether they have the cumulative, formal theory a mature science needs. Michael Muthukrishna and Joseph Henrich argue that psychology's replication troubles are downstream of a deeper theory problem: without an overarching theoretical framework, findings accumulate as a disconnected pile rather than a structure, and there is no principled way to say which results should generalize (#ref-muthukrishna-2019). Markus Eronen and Laura Bringmann sharpen the diagnosis into a theory crisis, contending that the discipline's difficulty building good theories, not merely its difficulty replicating effects, is the primary obstacle to progress (#ref-eronen-2021).

A parallel critique targets generalizability directly. Tal Yarkoni argues that the statistical models behavioral scientists routinely use license verbal conclusions far broader than the narrow designs that produced them can support — a mismatch he calls the generalizability crisis (#ref-yarkoni-2022). Taken together these arguments recast Cronbach's sixty-year-old worry for the present: the behavioral disciplines remain a collection of powerful methods in search of the integrating theory that would turn a filing category into a science. The debate is unusually self-aware and quantitative, and it is where the umbrella heading's oldest tension — many disciplines, one label — is being fought out today.

Common Misconceptions

Behavioral science is just a synonym for behaviorism.

It is not. Behaviorism is a specific theoretical program that restricts psychology to observable behavior; behavioral science is a mid-century umbrella for any discipline studying behavior, cognitive and social sciences included. The term was coined in part to be broader and more politically neutral than existing labels (#ref-miller-1955).

The MeSH header names a real scientific discipline.

It names an indexing category. MeSH is a controlled vocabulary for retrieval; F04 groups headings for filing, and Koch argued at length that treating such an institutional label as evidence of a unified discipline mistakes bookkeeping for science (#ref-koch-1993).

A validated therapy works because of its signature technique.

Comparative trials find that the common factors shared across therapies account for much more of the outcome than the specific techniques that distinguish them, so a therapy's brand-name method is often not the main reason it helps (#ref-wampold-2015).

Psychology's main problem is that experiments fail to replicate.

A growing body of argument holds that unreliable replication is a symptom, and that the deeper problem is a shortage of formal, cumulative theory to say what should replicate and generalize in the first place (#ref-muthukrishna-2019; #ref-eronen-2021).

Glossary

Aptitude-by-treatment interaction.

A pattern in which the effect of an experimental treatment depends on a person's standing on some individual-difference trait; Cronbach's proposed bridge between the two disciplines.

Behavioral sciences.

The family of disciplines — psychology foremost — that study the actions and reactions of organisms; a child of, and narrower than, the F04 umbrella.

Cohen's d.

A standardized mean difference, the gap between two group means divided by a pooled standard deviation; the experimental discipline's characteristic effect size.

Common factors.

The ingredients shared across psychotherapies — alliance, expectation, the therapist — that account for a large share of outcome independent of any specific technique.

Consilience.

E. O. Wilson's thesis that the sciences, including the behavioral and social ones, can in principle be unified into a single explanatory framework grounded in biology.

Correlational discipline.

Cronbach's term for the tradition that studies naturally occurring individual differences and their associations, rather than manipulating a variable experimentally.

Experimental discipline.

Cronbach's term for the tradition that manipulates a variable and compares condition means, treating individual differences as error to be controlled.

Generalizability crisis.

Yarkoni's argument that behavioral scientists draw broad verbal conclusions their narrow statistical models and designs cannot actually license.

MeSH.

Medical Subject Headings, the U.S. National Library of Medicine's controlled vocabulary for indexing the biomedical literature; the source of the F04 classification.

Point-biserial correlation.

A correlation between a two-category grouping variable and a continuous score; the correlational discipline's way of expressing a two-group effect.

Psychological techniques.

The methods and procedures used to investigate and to influence mental processes and behavior; a child descriptor of the F04 header.

Psychotherapy.

The treatment of mental and behavioral disorders by psychological rather than physical means; the largest applied activity filed under the umbrella.

Scientist-practitioner model.

The Boulder-conference training model holding that clinical psychologists should be trained both to conduct research and to deliver evidence-based practice.

Theory crisis.

Eronen and Bringmann's contention that psychology's central obstacle is its difficulty in constructing good, cumulative theories, of which the replication crisis is a symptom.

Two disciplines.

Cronbach's diagnosis that scientific psychology had divided into an experimental and a correlational stream that seldom communicate.

Key Researchers

Bernard Berelson (1912-1979). Directed the Ford Foundation's Behavioral Sciences Program, which funded and helped name the mid-century behavioral sciences, and co-authored a landmark inventory of scientific findings on human behavior. Wikipedia - Wikidata

Lee J. Cronbach (1916-2001). His 1957 APA presidential address named the two disciplines of scientific psychology, framing the experimental-correlational split that still organizes the field. Wikipedia - Wikidata

Markus I. Eronen (living). Philosopher of science at the University of Groningen working on theory formation and explanation in psychology; co-author of the theory-crisis argument. ORCID - Google Scholar - Faculty Page

James Grier Miller (1916-2002). Credited with giving behavioral science its modern currency around 1955 and with founding living systems theory; worked at the University of Michigan and led the University of Louisville. Wikipedia - Wikidata

Michael Muthukrishna (living). Professor of economic psychology at the London School of Economics; co-author of the argument that the behavioral sciences need cumulative theory rather than more data. ORCID - Google Scholar - Faculty Page

Mark Solovey (living). Historian of the behavioral and social sciences at the University of Toronto whose work traces how Cold War patronage shaped and named the field. Google Scholar - Faculty Page

Frequently Asked Questions

Is Behavioral Disciplines and Activities a scientific field? No. It is a MeSH indexing header (tree number F04) that groups the sciences of behavior with the activities they support, so that biomedical literature can be filed and retrieved. It is a classification convenience, not a discipline with its own methods.

How is it different from Behavioral Sciences? Behavioral Sciences is a narrower child of this header, naming only the family of disciplines that study behavior. The umbrella is broader, covering both those disciplines and the applied activities, such as psychotherapy and assessment, that put them to use.

Who coined the term behavioral science? James Grier Miller is generally credited with giving the term its modern currency in the early 1950s, and the Ford Foundation's Behavioral Sciences Program, led by Bernard Berelson, spread it through American research administration.

What are Cronbach's two disciplines? In a 1957 address, Lee Cronbach described scientific psychology as split into an experimental stream, which manipulates variables and compares group means, and a correlational stream, which studies naturally occurring individual differences. He argued the two seldom communicate.

What is the scientist-practitioner model? Adopted at a 1949 conference in Boulder, it holds that clinical psychologists should be trained both to produce and to consume research, so that practice stays disciplined by evidence and research stays informed by the clinic.

Why do critics say the technique of a therapy may not matter most? Comparative trials repeatedly find that factors shared across therapies, such as the alliance and the client's expectation of help, explain more of the outcome than the specific technique that gives a therapy its name.

What is the theory crisis in psychology? It is the argument, associated with Eronen and Bringmann and with Muthukrishna and Henrich, that psychology's deepest problem is its difficulty building cumulative formal theory, of which unreliable replication is a downstream symptom.

Does the umbrella heading have a parent in MeSH? No. F04 is one of the top-level roots of the MeSH tree, so nothing in the vocabulary classifies it as a kind of something more general. Its structure runs downward, into ten child descriptors.

References