Verbal Behavior: A Clinician's Guide

Most clinicians were taught about language as a thing the mind contains — a vocabulary, a grammar, a store of meanings the speaker reaches into. Skinner’s Verbal Behavior makes a deliberately different move: it treats talking as operant behavior, analyzing speech not by its form or its meaning but by the environmental variables that control it ²⁵. The same word, “water,” is a fundamentally different behavior when a thirsty child requests it than when a child labels it on seeing a lake, because the controlling variables differ; and that distinction, trivial to a linguist, turns out to be the whole game for a clinician teaching a minimally verbal child to communicate ³⁵. Understood this way, Verbal Behavior is less a theory of grammar than a functional taxonomy that tells you what to teach, in what order, and how to know which “language” a person actually has ³LLM.

Type & Discipline

Verbal Behavior is a theory — specifically a functional analysis of language — within the discipline of behavior analysis, sitting at the boundary where behavior analysis meets psycholinguistics ²LLM. It is not a therapy or a technique in itself; it is the conceptual framework from which a family of language-teaching procedures is derived ³LLM. Its defining commitment is functional rather than formal analysis: instead of describing language by its structure (phonemes, syntax, word classes), Skinner analyzed “behavior itself…with reference to the functional relationships of the behavior in the environment in which it occurs” ². Form alone is insufficient precisely because the same form carries different functions — the shout “Fire!” means something different at a shooting range, in a theater, and in a fireplace, depending entirely on the controlling circumstances ².

The unit of this analysis is the verbal operant: a class of verbal responses defined jointly by its antecedent controlling variable and its consequence, not by what it sounds like ³⁵. Because the framework treats speaking, signing, typing, and exchanging picture cards as functionally equivalent verbal behavior so long as they are reinforced through the mediation of a listener, it generalizes across communication modalities in a way that a form-based account cannot ³LLM. This is what makes it a construct-level theory with direct downstream procedures rather than a procedure itself LLM.

Creators & Lineage

The framework originates with B. F. Skinner, whose book Verbal Behavior was published in 1957 after decades of development ². The work extended Skinner’s operant analysis from the lever-pressing of the laboratory to the most distinctively human behavior, language, and it did so almost entirely theoretically: the book itself is “almost entirely theoretical, involving little experimental research,” offering an interpretive framework rather than a body of data ². That theoretical character is essential to understanding the lineage, because the empirical engine came later and from other hands ²⁴LLM.

The decisive turn from theory to clinical practice runs through Jack Michael and Mark Sundberg, whose work translated Skinner’s operants into a teaching and assessment technology for children with autism and developmental disabilities ³⁵. Sundberg’s contributions include early texts on teaching language to children with autism and a widely cited catalog of research topics drawn from Skinner’s book, and Sundberg and Michael’s foundational review is repeatedly cited as the work that organized the applied field ³⁵. The Verbal Behavior Milestones Assessment and Placement Program (VB-MAPP), developed by Sundberg, became the standard instrument for assessing a learner’s repertoire across the verbal operants and placing intervention targets, grounding modern language assessment for autism directly in Skinner’s taxonomy LLM.

Two intellectual descendants complete the picture. Applied behavior analysis absorbed the verbal operants as its language-teaching backbone, and Relational Frame Theory emerged later as a post-Skinnerian behavioral account of language and cognition that addresses generativity in a way Skinner’s original treatment did not, the same gap Skinner’s critics had pressed ²LLM.

Core Principles

First, language is behavior controlled by the environment, not a window into meaning. A verbal response is defined by the variables that occasion and maintain it, so the analysis always asks “what controls this response” rather than “what does this word mean” ²⁵.

Second, verbal behavior is operant behavior reinforced through a listener. What distinguishes verbal from other behavior is that its reinforcement is mediated by another person — the speaker’s “water” produces water only because a listener is trained to deliver it — which is why the same physical act can be verbal in one context and not another ²LLM.

Third, the same form can be different operants. Because function, not topography, defines the operant, “water” as a request (mand) and “water” as a label (tact) are different verbal operants that must, in teaching, be established separately rather than assumed to transfer automatically ³⁵. This is the single principle with the largest clinical payoff ³LLM.

Fourth, the elementary operants are distinguished by their controlling variables. A mand is controlled by motivation (deprivation or aversive stimulation), a tact by a nonverbal stimulus, an intraverbal by other verbal behavior, and an echoic by a verbal stimulus with point-to-point correspondence — and these distinctions are not academic, because each implies a different teaching arrangement ³⁵.

Fifth, higher-order verbal behavior modifies primary verbal behavior. The autoclitic is “a form of verbal behavior which modifies the functions of other forms,” as when “I think” in “I think it is raining” tempers the strength of the assertion, providing Skinner’s account of grammar, qualification, and self-editing without invoking an internal grammar module ².

Interventions & Techniques

The framework’s core clinical instruction is to teach language operant by operant, by function, rather than by vocabulary. Because a request and a label are different operants, intervention establishes each verbal operant under its own controlling variable rather than teaching a word once and assuming it will appear across functions ³⁵.

The most developed procedure is mand training, which teaches a learner to request preferred items by manipulating the relevant motivation — the establishing operation — so that the response is genuinely controlled by wanting the item rather than by a prompt ³. Mand training has been delivered through manual sign training, the Picture Exchange Communication System, discrete-trial instruction, video modeling, and differential reinforcement of vocal approximations, and the reviews emphasize that effective protocols ensure “EOs controlled the targeted mands” so the child is truly requesting and not merely complying ³. Mand training is the most-studied application by a wide margin, accounting for the majority of the empirical literature ³⁴.

A clinically pivotal application is functional communication training (FCT): when severe problem behavior is maintained by the same consequence a mand would produce (attention, escape, a tangible), teaching a functionally equivalent request can replace the problem behavior ³⁴. The reviews describe mand-training in the form of FCT as “a highly effective treatment for severe problem behavior in individuals with developmental disabilities,” a striking instance of a language procedure resolving a behavior problem ⁴.

Beyond manding, intervention systematically builds the other operants — tacts (labeling and commenting), echoics (vocal imitation), intraverbals (conversational responding with no point-to-point correspondence, such as answering “How are you?”), textual behavior (reading aloud), and listener responding (following instructions) — and exploits the emergence of untrained responses, arranging conditions so that teaching one operant produces others without direct training ³⁵.

LLM-generated illustrative example (not a guideline): A five-year-old says “bubbles” only when an adult holds up the bubble jar and asks “What do you want?” — a prompted response that disappears when the prompt does. A verbal-operants analysis recognizes this as a partially prompted mand under instructional control, not motivational control. The team withholds the verbal prompt, lets the child notice the desirable bubbles (arranging the establishing operation), and reinforces any approximation of “bubbles” with immediate bubbles. Separately, they teach the tact “bubbles” by having the child label bubbles in a book for praise. The same word is built twice, under two different controlling variables ³⁵LLM.

Evidence Base

The honest evidence picture requires splitting the applied framework from the general theory, because they have very different standing ²³LLM.

The applied verbal-operants approach is established. A systematic review of interventions for children with autism identified 172 empirical studies published between 2001 and 2017 across 22 journals, involving 493 children, with high interrater reliability (96% agreement on the third of studies double-coded) ³. Within that corpus the mand is by far the best supported — 91 of 172 studies (52.9%) targeted mands — followed by tacts (56 studies, 32.6%) and intraverbals (40 studies, 23.3%), while echoics were least studied (4 studies, 2.3%) ³. An earlier review found a comparable concentration, with roughly 72% of the human empirical literature involving a mand repertoire, and documented near-threefold growth in studies over the preceding decades ⁴. The applied literature reports basic mand acquisition and generalization, reduction of destructive behavior through FCT, emergence of untrained responses, and collateral gains such as a decrease in stereotypy following tact acquisition ³.

Two honest qualifications follow. First, the evidence is concentrated and uneven: it rests heavily on single-subject designs in children with autism and developmental disabilities, is dominated by manding, and thins out sharply for complex repertoires — autoclitics are essentially unstudied empirically (“no new studies solely on autoclitics” in decades), and storytelling, problem-solving, and intraverbal complexity remain underdeveloped, with the field’s own reviewers stating that “considerably more research is needed” ³⁴. Second, the framework historically suffered from a gap between citation and test: of over 800 studies citing Verbal Behavior, only 31 had explicitly examined one of Skinner’s operants, meaning the theory was long invoked far more than it was empirically interrogated ⁴.

The general theory remains contested. Skinner’s book was almost entirely interpretive, and Noam Chomsky’s 1959 review famously argued that children acquire language without explicit behaviorist teaching and routinely produce novel sentences they have never heard — generativity Skinner’s operant account struggled to explain ². MacCorquodale’s 1970 reply countered that Chomsky misunderstood behavioral psychology and quoted Skinner out of context, and the dispute was never empirically settled so much as it bifurcated the field ². The defensible clinical position is that the applied technology is well supported even where the theory’s adequacy as a complete account of language is not ²³LLM.

Populations & Indications

The framework’s clearest indications are populations who are not acquiring language typically and for whom function-by-function teaching is necessary ³⁴LLM. The empirical core is children with autism spectrum disorder, the population of essentially the entire modern applied literature, with reviewed participants ranging from birth to age 12 ³. Closely adjacent are individuals with developmental and intellectual disabilities, where the framework’s recent studies have concentrated on “clinical interventions for individuals with little-to-no communicative behavior” ⁴.

Within those groups, minimally verbal and early language learners are the prototypical candidates, because the operant-by-operant approach lets a clinician build communication from a near-absent repertoire and assess precisely which functions exist ³⁵. People with severe problem behavior maintained by communication deficits are a distinct indication via functional communication training, where the relevant “language” target is a replacement request rather than vocabulary as such ³⁴. The framework has been applied far less to typically developing children and to adults with “adequate or exceptional language repertoires,” which is both a limit of the evidence and a reflection of where function-by-function teaching is most needed ⁴.

Problems-for-Work

The verbal-operants lens converts a vague “language delay” into a specific, addressable question: which operants does this person have, under which controlling variables, and which are missing ³⁵?

• Limited functional communication is reframed as an absent or weak mand repertoire; the work is mand training under genuine motivational control so the learner can request rather than escalate ³⁴.
• Severe problem behavior maintained by communication deficits is addressed by functional communication training, teaching a functionally equivalent request that produces the same reinforcer the behavior was producing ³⁴.
• Requesting and mand deficits specifically are the most-supported target, approached by arranging establishing operations and reinforcing approximations across modalities including sign and picture exchange ³.
• Restricted conversational and intraverbal repertoires — a child who can label objects but cannot answer “What do you do when you’re tired?” — are worked as intraverbal deficits, taught as verbal-controls-verbal responses distinct from tacts ³⁵.
• Expressive and receptive language delay is parsed across operants, distinguishing speaker behavior (mand, tact, echoic) from listener responding (following instructions), so intervention targets the specific weak function rather than “language” generically ³⁵.
• Lack of generalization of language skills — a word that appears only under one prompt or one function — is treated by deliberately establishing the response across operants and arranging conditions for untrained responses to emerge ³⁵.

LLM-generated illustrative example (not a guideline): A nonverbal eight-year-old hits and bites when worksheets appear, and the behavior reliably ends the task. A functional assessment indicates escape maintains the behavior. Rather than treating “aggression” directly, the team teaches a functionally equivalent mand — handing a “break” card — and ensures it produces an immediate brief escape, the same consequence the aggression produced. As the break-request becomes the more efficient route to escape, the aggression has nothing left to do ³⁴LLM.

Contraindications, Cautions & Cultural Humility

The framework is a way of analyzing language, so the cautions concern how it is applied rather than a contraindicated population LLM. The first caution is not mistaking the applied technology’s strength for the theory’s completeness: the operants are an excellent map for building a repertoire, but Skinner’s account does not fully explain the generativity Chomsky pressed, so a clinician should not assume that operant-by-operant teaching alone will produce the open-ended, novel language a child ultimately needs ²LLM. Relational Frame Theory exists in part because that generativity gap is real LLM.

The second caution is empirical humility about complex repertoires: the evidence is robust for mands and reasonable for tacts, but thin for intraverbals and essentially absent for autoclitics, so plans that lean on the framework for advanced conversation, narrative, or perspective-taking are extrapolating beyond the strong data ³⁴LLM. The third caution is procedural — manding only counts as manding when motivation, not a prompt, controls it, and a “request” that is really prompted compliance has not been taught at all, a failure mode the reviews flag explicitly ³.

Cultural humility enters through the recognition that what is reinforcing, and what counts as appropriate communication, is not universal. The items a learner is motivated to mand for, the conversational forms valued by a family, and the very goals of “more talking” are shaped by culture, language background, and family priorities, so targets and reinforcers must be selected with the family rather than from a generic menu ³LLM. The neurodiversity-informed clinician also holds the goal honestly: the aim is functional communication the person can use to get their needs met, not the suppression of natural communication or the imposition of speech for its own sake LLM.

Treatment-Plan Suggestions & SMART Objectives

Common Misconceptions

The most consequential misconception is that a word, once taught, transfers across functions — that a child who can label “cookie” can therefore request it. The framework insists the opposite: a tact and a mand are different operants under different controlling variables, and each must be established in its own right ³⁵. A second misconception is that verbal behavior means vocal behavior; because the operants are defined functionally, signing, exchanging pictures, and typing are all verbal behavior when reinforced through a listener, which is why the framework spans communication modalities ²³. A third is that “language is just imitation”; echoics (imitation) are only one operant among several, and the framework is built precisely to teach the non-imitative functions that imitation cannot reach ³⁵. A fourth is that Skinner’s theory is settled science; the original work was almost entirely interpretive, Chomsky’s 1959 critique remains influential, and the theory’s adequacy as a complete account of language is genuinely contested even while the applied technology is well supported ²⁴. A fifth is that the autoclitic and grammar are well established empirically; in fact autoclitics are essentially unstudied in the applied literature, so confidence there rests on theory, not data ⁴LLM.

Training & Certification

There is no certification in “verbal behavior” as such; competence in applying it lives within behavior-analytic training and credentialing, and the operants are a core part of that curriculum ³LLM. Practical competence rides on two layers: the conceptual fluency to reframe a presenting language problem as a question about which operants exist under which controlling variables, which any clinician can adopt; and the technical skill to run mand training, functional communication training, and the differential-reinforcement and prompting procedures that build the operants, which sits within formal behavior-analytic education and supervised practice ³⁵LLM.

Assessment fluency matters as much as teaching fluency, because the framework’s payoff is diagnostic before it is prescriptive — knowing which operant is weak is what makes the intervention specific rather than generic ³LLM. In practice this means familiarity with operant-by-operant assessment and placement tools and with the reviews that document what is and is not empirically supported, so that a clinician extends the framework into well-evidenced territory (manding) with confidence and into thinly studied territory (complex intraverbals, autoclitics) with appropriate caution ³⁴LLM.

Key Terms

• Verbal operant: a class of verbal responses defined by its controlling antecedent and its consequence, not by its form ³⁵.
• Mand: verbal behavior under the control of motivation (deprivation or aversive stimulation) and reinforced by the specific item or action requested; a request or demand ²³.
• Tact: verbal behavior under the control of a nonverbal stimulus, reinforced by generalized conditioned reinforcement; a label or comment ²³.
• Intraverbal: verbal behavior under the control of other verbal behavior with no point-to-point correspondence; conversational responding such as answering a question ²³.
• Echoic: verbal behavior under the control of a verbal stimulus with which it shares point-to-point correspondence; vocal imitation ²³.
• Textual: a vocal response controlled by a written verbal stimulus; reading aloud, with no implication of comprehension ²⁵.
• Autoclitic: a higher-order verbal operant that modifies the function of other verbal behavior, such as “I think” softening an assertion ²⁵.
• Listener responding (receptive language): the behavior of the listener whose actions are altered by the speaker’s verbal behavior, such as following an instruction ⁵.
• Establishing operation (EO): the motivational variable that momentarily makes an item more valuable and evokes the mand for it ³.
• Functional communication training (FCT): teaching a functionally equivalent request to replace problem behavior maintained by the same reinforcer ³⁴.
• Functional vs. formal analysis: analyzing verbal behavior by its environmental controlling relations rather than by its structure or grammar ².

Resources & Further Reading

• Verbal Behavior — B. F. Skinner (1957), Google Books
• Verbal Behavior — Wikipedia
• Empirical Application of Skinner’s Verbal Behavior to Interventions for Children with Autism: A Review — PMC
• Empirical Applications of Skinner’s Analysis of Verbal Behavior with Humans — PMC
• Understanding Skinner’s Verbal Operants — Hidden Talents ABA
• ABA Therapy: Verbal Operants (Teaching Language) — video

Reflective / Supervision Questions

• For this learner, have I actually assessed which operants exist under which controlling variables, or am I treating “language” as one undifferentiated thing? ³⁵
• When I say the child “has” a word, do they have it as a mand, a tact, both, or neither — and am I teaching the function that is actually missing? ³⁵
• Is the mand I am teaching under genuine motivational control, or have I built a prompted response that will vanish when the prompt does? ³
• For a behavior problem that looks like aggression or escape, have I asked what the person would be manding for if they could, and could functional communication training meet that need? ³⁴
• Am I extending the framework into well-evidenced territory (manding) or thinly studied territory (complex intraverbals, autoclitics), and is my confidence calibrated to the evidence? ³⁴
• Where am I leaning on Skinner’s theory as if it were settled, and where does the generativity of real language outrun what operant-by-operant teaching can produce? ²
• Have the communication targets and reinforcers been chosen with the family and the person’s culture in mind, or imposed from a generic curriculum? ³