An intraverbal is a type of verbal behavior in which a person responds to someone else’s words with different words, without any object or event in the environment prompting the specific response. In Applied Behavior Analysis (ABA), the intraverbal is one of several categories of language first described by B.F. Skinner, and it covers a surprisingly large portion of everyday communication: answering questions, continuing a conversation, singing the next line of a song, or filling in a phrase someone else started. Teaching intraverbal skills is one of the more complex tasks in ABA programming, particularly for children with autism, because these responses depend on verbal cues alone rather than something the learner can see or touch.
What Makes an Intraverbal Different From Other Kinds of Language
Skinner’s framework divides verbal behavior into categories based on what controls the response. A mand is a request driven by motivation: a child says “cookie” because they want a cookie. A tact is a label driven by something in the environment: the child sees a dog and says “dog.” An echoic is a direct repeat of what someone else said: you say “ball” and the child says “ball.” The intraverbal is the odd one out because the response does not match the prompt and is not triggered by any visible object. When you ask “What do you eat for breakfast?” and a child answers “cereal,” the word “cereal” has no point-to-point relationship with any word in the question. The child’s answer is controlled entirely by the verbal stimulus and a learning history connecting that question to that answer.
This distinction matters practically because a child might be able to label a dog when they see one (tact) but be unable to answer “What animal barks?” (intraverbal) even though both responses require the word “dog.” In ABA, these are treated as functionally different skills. A child who can tact dozens of items may still have very few intraverbals, and building one type of verbal behavior does not automatically build the others. The technical literature makes the point that intraverbal control is woven through nearly all verbal behavior, since we almost always speak in response to what someone else has said, but the term “intraverbal” as a distinct category is most useful when the verbal antecedent alone is enough to produce the response.
Intraverbals in Typical Development
In children without developmental delays, intraverbal responses emerge gradually and tend to increase with age, though not on a neat timetable. Research comparing typically developing children and children with autism found a correlation between age and the number of correct intraverbal responses in the typically developing group, but there was variability among children of the same age.1PubMed Central. Intraverbal behavior and verbal conditional discriminations in typically developing children and children with autism Some three-year-olds can answer a broad range of “wh-” questions while others the same age are still limited to fill-in-the-blank responses like finishing familiar songs or phrases.
Early intraverbals tend to be simple and rote: a parent says “ready, set…” and the child says “go.” These evolve into answering “what” and “where” questions, then into more complex exchanges involving “why” and “how.” The progression is relevant for ABA practitioners because it informs the sequencing of intraverbal targets. Therapists typically start with simple fill-ins and work toward open-ended conversational exchanges, roughly mirroring the developmental trajectory that unfolds naturally in children without disabilities.
Why Intraverbals Are Especially Hard for Children With Autism
Children with autism often develop other verbal operants, such as mands and tacts, more readily than intraverbals. Several factors contribute. Intraverbals require a learner to respond to an abstract verbal cue without the support of visible objects. When a therapist holds up a picture of a cat and asks “What is this?”, the picture does the heavy lifting. When the therapist puts the picture away and asks “What animal says meow?”, the child has to retrieve the answer from memory based on a verbal prompt alone. For many children with autism, that shift from concrete to abstract stimulus control is the core challenge.
Another layer of difficulty comes from what researchers call multiple control. In natural conversation, a single response is often controlled by more than one verbal stimulus at the same time. If someone asks “What is something red that you eat?”, a correct answer like “strawberry” has to satisfy both the color condition (red) and the category condition (food). Training programs for children with autism have identified this kind of multiply controlled responding as a major instructional challenge, one that often requires specialized teaching procedures beyond basic prompting.2PubMed Central. Intraverbal Training for Individuals with Autism: The Current Status of Multiple Control
Prerequisite Skills That Set the Stage
One of the more practical findings in this area is that intraverbals do not emerge in isolation. They depend on a set of component skills that, when all are in place, can produce novel intraverbal responses the child was never directly taught. Research with children with autism has shown that when a specific set of prerequisite skills was trained to mastery, convergent intraverbals (responses where multiple verbal stimuli converge on a single answer) emerged without being directly taught.3PubMed. Facilitating the emergence of convergent intraverbals in children with autism
A related study looked at a slightly different form called intraverbal tacts, responses that involve describing features of something not currently present. Researchers found that intraverbal tacts emerged only when all component skills were mastered, and preliminary data suggested that at least four of the six targeted skills appeared to be necessary rather than just helpful.4PubMed. Facilitating the emergence of intraverbal tacts in children with autism spectrum disorder: A preliminary analysis The practical takeaway for therapists and parents is that skipping foundational skills to jump straight to teaching intraverbals often backfires. When a child is struggling with a particular intraverbal target, the issue may not be with that target at all but with a missing prerequisite further down the chain.
What counts as a prerequisite varies somewhat depending on the type of intraverbal being targeted, but common foundations include strong tacting skills (being able to label items and their features), listener responding (following instructions involving those items), and some echoic ability (being able to repeat words on request). The idea is that these existing repertoires give the learner the raw material from which intraverbal responses can be assembled.
Teaching Strategies That Work
ABA practitioners use a variety of procedures to teach intraverbals, and the research has explored which approaches are most effective and under what conditions.
Transfer of Stimulus Control
One widely used method involves starting from a response the child can already produce under one type of control and gradually shifting control to a verbal antecedent. For example, if a child can tact a picture of a shoe, the therapist might show the picture and ask “What do you wear on your feet?” then fade the picture over successive trials until the child answers the question without it. This is called transfer of stimulus control, and it is a cornerstone of intraverbal instruction in many ABA programs.
Research on this procedure found that for all participants in one study, whichever prompting method had been used most recently to teach intraverbals required fewer trials to teach new ones, suggesting that learners become more efficient at the transfer process itself with repeated exposure.5PubMed Central. The role of increased exposure to transfer-of-stimulus-control procedures on the acquisition of intraverbal behavior In practical terms, the first few intraverbals a child learns may take many sessions, but the pace tends to pick up as the child gets more practice with the general format.
Prompt Delay With Error Correction
Another approach is prompt delay: the therapist presents the verbal question, waits a set number of seconds for the child to respond, and if no correct response comes, provides a prompt (often an echoic model of the correct answer). When errors occur, a correction procedure follows. This method has been shown to be effective for teaching multiply controlled intraverbals, though some learners need additional strategies layered on top. In a study with seven participants with autism, prompt delay with error correction improved accuracy for at least one set of stimuli for all participants, but four required supplementary procedures such as differential observing responses, which essentially require the learner to attend to each part of a complex question before answering.6Journal of Applied Behavior Analysis. Teaching multiply controlled intraverbals to children and adolescents with autism spectrum disorders
Teaching Children to Say “I Don’t Know”
An often-overlooked piece of intraverbal instruction is teaching children what to do when they do not know the answer. Guessing or echoing the question are common error patterns. Researchers taught children with autism to respond to unknown questions by saying “I don’t know, please tell me,” which functioned as a mand (request) for information. All four participants in the study learned the response, and two of them went on to acquire correct answers to previously unknown questions simply through hearing the answers provided after their request.7PubMed Central. Acquisition of intraverbal behavior: teaching children with autism to mand for answers to questions The “I don’t know” response also generalized to new untargeted questions for three of the four participants, making it a useful general strategy rather than a rote response tied to specific questions.
Feature, Function, and Class Questions
A significant category of intraverbal targets in ABA programs involves questions about an item’s features (what it looks like), functions (what it does or what it is used for), and class membership (what category it belongs to). These are often abbreviated as FFC in clinical shorthand. “What color is a banana?” targets a feature. “What do you use to cut paper?” targets a function. “Name an animal” targets a class.
FFC intraverbals are considered a bridge between simple fill-in responses and the kind of flexible language needed for conversation. They require the child to think about properties of items that are not present, which exercises a more abstract kind of verbal control. Research has shown that using FFC-based intraverbal prompts can increase the number of different responses a child gives to open-ended questions. In a study with children with autism, prompts involving function, feature, and class effectively increased the number of divergent responses across all three targeted questions.8PubMed Central. Increasing Response Diversity to Intraverbals in Children with Autism Spectrum Disorder
A related line of work has shown that training tacts and listener responses by feature, along with name-feature and feature-name intraverbals, can lead to the emergence of responses that were never directly trained.9PubMed. Promoting the development of verbal responses using instructive feedback This is one of the more encouraging findings in the field: when instruction is well-structured, children sometimes show novel, untaught language. It does not happen reliably or for every child, but it suggests that the right combination of training can create conditions for language to grow beyond what was explicitly programmed.
Answering Wh- Questions
“Wh-” questions (what, where, who, why, when, how) represent some of the most complex intraverbal targets. Each type of question requires a different class of response: “what” questions typically call for an object or action, “where” questions call for a location, “who” questions call for a person, and so on. A child might learn to answer “what” questions well but struggle with “where” questions, because the two are controlled by different stimulus features even though both are phrased as questions.
A study using multiple exemplar training across classes of wh- questions found that all five participating children with autism became able to answer novel questions with complete sentences within each trained class, and these skills transferred across different people, settings, and over time.10PubMed. Teaching children with autism to answer novel wh-questions by utilizing a multiple exemplar strategy The key word here is “novel”: the children were not simply parroting memorized answers but were producing appropriate responses to questions they had never encountered before. The approach relied on training enough examples within each question class that the child could abstract the general pattern.
Research has also explored different prompting formats for wh- questions, including visual prompts (pictures), echoic prompts (verbal models), and technology-delivered prompts. A comparison study with an adolescent with Down syndrome found that picture prompts, echoic prompts, and iPad-delivered prompts were all effective for teaching wh- question responses.11PubMed Central. A Comparison of Prompting Strategies to Teach Intraverbals to an Adolescent with Down Syndrome The finding that multiple formats work is useful for practitioners because it means the prompting strategy can be matched to the learner’s strengths rather than requiring a one-size-fits-all approach.
When Errors Pile Up
Error patterns during intraverbal training can be stubborn. Common problems include the child echoing the question instead of answering it, giving the same answer to every question (a “scrolling” error where a recently reinforced response gets recycled), or failing to discriminate between questions that sound similar. Error correction procedures are critical, but not all correction methods work equally well for every learner.
One study compared three error correction approaches for teaching intraverbal behavior: vocal modeling (the therapist says the correct answer and the child repeats it), single response repetition (the child repeats the correct answer once), and multiple response repetition (the child repeats the correct answer several times). Results were mixed: the brief assessment matched the full validation assessment for only one of three participants, highlighting how individualized error correction needs to be.12Scholarship Repository @ Florida Tech. Implementing an Abbreviated Assessment to Compare Error-Correction Procedures for Teaching Intraverbal Behavior to Children The broader lesson is that ABA practitioners often need to test multiple approaches with a given learner rather than defaulting to a single correction strategy.
Joint Control and Emergent Responding
A theoretical concept that has gained traction in intraverbal research is joint control, which describes a situation where two or more sources of stimulus control converge on the same response. When someone asks you to “name a fruit that is yellow,” you might covertly cycle through fruit names while simultaneously checking each against the criterion “yellow” until “banana” emerges as the response that satisfies both conditions. That convergence of two streams of control is joint control.
A systematic review of the literature found that training based on joint control principles is a promising approach for teaching complex, multiply controlled verbal and nonverbal repertoires to children with autism and other developmental disabilities.13PubMed Central. The Effect of Joint Control Training on the Performance of Multiply Controlled Behavior: A Systematic Literature Review Relevant to Children With Autism Spectrum Disorder and Other Developmental Disabilities The evidence is still developing, but the idea has practical appeal because it gives therapists a way to think about what is happening internally when a child fails to answer a complex question and how to build the covert behaviors that bridge the gap.
The broader question of when and how untaught intraverbal responses emerge is an active area of research. A scoping review identified 99 experiments across 79 articles that have studied emergent intraverbals, reflecting significant interest in understanding the conditions under which language generalizes beyond what is directly trained. The research spans a range of populations and teaching procedures, but a unifying theme is that structured training of related skills can produce novel verbal behavior that was never explicitly reinforced.
Technology-Assisted Instruction
Tablets and speech-generating devices have become common tools in ABA classrooms, and they have been applied to intraverbal instruction as well. For children who can speak, technology often serves as a prompting tool, delivering visual or auditory cues during instruction. For children who communicate through augmentative and alternative communication (AAC) devices, the device itself becomes the medium through which intraverbal responses are produced.
A study examining iPad-based speech generating devices found that both participating school-aged children with autism acquired the ability to respond to three different intraverbal statements using the device.14Journal of Developmental and Physical Disabilities. The Acquisition of Intraverbal Responding using a Speech Generating Device in School Aged Children with Autism This is relevant because it demonstrates that intraverbal repertoires are not limited to vocal speech. A child who navigates a speech-generating app to answer a question is engaging in the same functional class of behavior as a child who answers vocally, and the same teaching principles apply.
Intraverbals Beyond Childhood Autism Programs
While most of the intraverbal research focuses on children with autism, the concept has applications across other populations. One area that might surprise people is dementia care. A study with adults in long-term care examined whether having participants engage in tacting, echoic, and intraverbal responses when presented with visual stimuli could improve recall. Compared to a condition with no verbal responding, incorporating these verbal operants improved recall for the population studied.15PubMed. Applying Skinner’s analysis of verbal behavior to persons with dementia The finding suggests that the intraverbal framework is not just a teaching tool for children but a way of analyzing and supporting verbal behavior across the lifespan.
Animal cognition researchers have also borrowed concepts from Skinner’s analysis of verbal behavior to study derived stimulus relations in non-human species. While animals have not yet demonstrated the flexibility or generativity seen in human verbal behavior, the research strategies used in this work draw on the same principles that underlie intraverbal training.16PubMed Central. Abstraction, Multiple Exemplar Training and the Search for Derived Stimulus Relations in Animals The gap between what animals can learn in these paradigms and what children demonstrate with intraverbal training underscores just how much of human conversation rests on the ability to respond to words with contextually appropriate, novel words, something that looks simple until you try to teach it systematically.
Common Misconceptions About Intraverbals in Practice
One persistent misconception is that intraverbals are just “answering questions.” While question-answering is the most obvious example, intraverbals also include completing phrases, continuing a topic in conversation, associating related words, and responding to social scripts. A child who hears “Twinkle, twinkle, little…” and says “star” is producing an intraverbal. So is a child who, upon hearing a peer say “I went to the zoo,” responds with “I like the zoo too.” Reducing the category to question-answering leads practitioners to neglect the conversational and social dimensions of intraverbal training.
Another misunderstanding involves the relationship between memorization and intraverbal ability. Rote memorization of answers to specific questions is not the same as having a flexible intraverbal repertoire. A child who has memorized that the answer to “What color is the sky?” is “blue” but cannot answer “Tell me something that is blue” has a narrow, stimulus-specific response rather than a generalized intraverbal skill. Effective intraverbal instruction aims for generativity: the ability to answer novel questions and engage in spontaneous verbal exchanges that were never rehearsed. The research on multiple exemplar training and prerequisite skills speaks directly to this goal, showing that when instruction is structured to promote flexible responding, children can go beyond rote recall and demonstrate genuinely creative language use.
A third common confusion is assuming that if a child has strong receptive language, they can be expected to produce intraverbals easily. Receptive skills (understanding what you hear) and intraverbal production (generating a verbal response to verbal input) are dissociable. Some children comprehend questions perfectly well but cannot produce the words to answer them. In ABA, this means assessment should evaluate intraverbal responding directly rather than assuming it will follow from receptive competence. Relying on receptive performance as a proxy for expressive ability often leads to overestimation of what a child can do in conversation and underestimation of what they still need to be taught.