12.3 Teaching Elementary Verbal Operants: Mands, Echoics, Tacts, and Intraverbals
Key Takeaways
Mand training starts by capturing or contriving a motivating operation, prompts the response (often with an echoic), delivers the specific item or action requested, and fades prompts until the MO alone evokes the mand.
Tacts are taught by transferring control from an echoic prompt to the nonverbal stimulus ("What is it?" plus "Ball" fading to the ball alone) and are reinforced with generalized social reinforcement.
Intraverbals are taught by transferring control from echoic, picture, or textual prompts to the verbal stimulus (fill-ins, then WH-questions), and questions should be varied to avoid rote chains.
Transfer-of-stimulus-control procedures pair the target antecedent with a prompt that already evokes the response, then fade the prompt so the target antecedent alone controls responding.
From Classification to Teaching
Section 3.3 defined the verbal operants by their controlling variables. Task G.10 asks you to teach them. The central idea is transfer of stimulus control: start with a prompt that already evokes the response, present the target antecedent together with that prompt, and fade the prompt until the target antecedent alone controls the response. Which antecedent you transfer to depends on the operant.
| Operant | Target antecedent | Typical prompt to transfer from | Reinforcer |
|---|---|---|---|
| Mand | Motivating operation (with the item present or absent) | Echoic ("Say ball"), sign model, or physical prompt for a picture exchange | The specific item or action requested |
| Echoic | A vocal model | Shaping successive approximations | Generalized social reinforcement (and often preferred items early on) |
| Tact | A nonverbal stimulus (object, action, property) | Echoic | Generalized social reinforcement |
| Intraverbal | A verbal stimulus with no point-to-point correspondence | Echoic, picture (tact), or text prompt | Generalized social reinforcement |
| Listener response | A spoken instruction (a conditional discrimination) | Gestural, model, or physical prompt | Social or other reinforcement |
Mand Training
Mands are usually taught first because they directly benefit the learner and can replace problem behavior that serves the same function (Sundberg & Partington, 1998).
- Identify strong motivating operations: preferred items and activities, help with difficult tasks, breaks, attention.
- Capture or contrive the MO. Capture an MO when the learner already wants something (reaching for juice). Contrive one by arranging the environment: give a puzzle with a missing piece, hand over a toy that needs batteries, or place preferred items in clear containers that are hard to open (a transitive CMO, section 3.2).
- Prompt the mand with the response form the learner can best produce: an echoic prompt ("Juice"), a sign model, or physical guidance to exchange a picture.
- Reinforce with the specific item or action immediately, not general praise. Saying "Good talking!" without giving juice turns the mand into something else.
- Fade prompts (time delay, less intrusive prompts) until the learner mands when the MO is present without a prompt, and when the item is not in sight.
- Expand: teach mands for actions ("push"), information ("Where is it?"), help, and removal of aversive stimuli ("No more").
Picture Exchange Communication System (PECS): developed by Bondy and Frost (1994), PECS teaches learners to hand a picture to a communication partner in exchange for the item. Its phases progress from the physical exchange, to distance and persistence, to picture discrimination, to sentence structure ("I want ___"), to answering "What do you want?", and finally to commenting. PECS starts with manding because the exchange is directly reinforced by the item.
Common errors: requiring the learner to say "please" as part of every mand (adding effort and a rote chain), asking "What do you want?" on every trial (making the mand dependent on an intraverbal prompt), and delivering a different item than the one requested.
Echoic Training
Echoics are the foundation for using vocal prompts to teach other operants. If the learner has some vocal imitation, reinforce close imitations of single sounds and words, then shape toward clearer approximations (section 10.3). Pairing the instructor's voice with reinforcement and reinforcing any vocal play can increase early vocal behavior. For learners with very limited vocal repertoires, mands may be taught with signs or picture exchange while echoics are built separately.
Tact Training
Tacts are taught by echoic-to-tact transfer:
- Present the object and the question ("What is it?") along with an immediate echoic prompt ("Ball").
- Reinforce the echoed word with praise.
- Delay the echoic prompt (for example, 2 to 3 seconds) so the learner can respond to the object before the prompt.
- Fade the prompt until the learner names the object when shown it.
- Fade the question too, so that the object alone (or with a neutral "What do you see?") evokes the tact.
Use multiple exemplars (many different balls) so the tact generalizes to the class, not to one picture. Extend tacts to actions, features, prepositions, and private events ("My tummy hurts").
Intraverbal Training
Intraverbals answer verbal stimuli without point-to-point correspondence. Start with fill-ins the learner can complete ("Ready, set, ___"; "Twinkle, twinkle, little ___"), then move to simple WH-questions ("What do you drink?").
- Prompts: an echoic prompt ("Milk"), a picture prompt (show milk and let the learner tact it, then fade the picture), or a text prompt for readers.
- Avoid rote responding by varying the wording of questions and the order of targets, and by teaching several answers to the same question ("Name an animal").
Listener Responding
Following instructions and selecting items when named (receptive identification) are listener behaviors, taught as conditional discriminations (section 12.2). Listener and speaker repertoires often need to be taught separately; learning to tact "cup" does not guarantee the learner will select the cup when asked, although for many learners teaching one supports the other.
Measuring Progress
Use cold probes (the first trial of the day, without prompts) to judge independent responding, record prompted and unprompted responses separately, and probe untaught exemplars and settings for generalization. For mands, count spontaneous (unprompted) mands per hour in natural routines; this is often the most socially meaningful measure.
Common Exam Traps
- Reinforcing a mand with praise only. The mand's reinforcer is the specific item or action.
- Teaching a tact under MO control. If the learner says "cookie" only when hungry and receives the cookie, it is a mand.
- Treating rote chains as intraverbals. "Hi, how are you?" recited regardless of what was said is not intraverbal control by the verbal stimulus.
- Never fading the echoic. If the learner still needs "Say ball," the tact has not been established.
A learner says "juice" when a therapist holds up a cup of juice. The therapist says "Great talking!" but keeps the juice on the table. Over several days the learner says "juice" less often. What is the most likely error?
The response was reinforced with general praise instead of the requested juice.
Juice should not be targeted because edibles cannot reinforce verbal behavior.
The therapist should have required "I want juice, please" before giving praise.
The therapist should have used a picture of juice instead of the real cup.
An instructor holds up a toy truck, asks "What is it?", and immediately says "Truck." The child repeats "Truck" and receives praise. Over sessions the instructor waits longer before saying "Truck" until the child names it alone. What procedure is this?
Intraverbal training, because the question "What is it?" controls the answer
Echoic-to-tact transfer, fading the vocal prompt so the object controls the response
Mand training, because the child receives the truck after naming it
Listener training, because the child must attend to the instructor's words
A learner answers "What do you sleep in?" only when shown a picture of a bed. Which next step best builds an intraverbal response?
Fade the picture prompt across trials so the spoken question alone evokes "bed."
Stop asking the question and present the bed picture alone so the learner tacts it.
Keep presenting the picture on every trial so the learner never makes an error.
Switch to receptive trials in which the learner selects the bed from three pictures.
Sections you finish are checked off in the contents.