Getting the Right Answer Isn’t Enough: Establishing Speaker–Listener Repertoires for Complex Learning

Free · 1 Learning CEU · certificate included

FreeDetailed certificate included
Register free & watch now

Claim your certificate as soon as you pass the quiz.

alphabeePRO1 Learning CEU60 minOn-demand

Presented by Dana Pettus

Getting the Right Answer Isn’t Enough: Establishing Speaker–Listener Repertoires for Complex Learning

About this CEU

Learners may follow multi-step instructions, complete worksheets, or answer comprehension questions correctly—yet performance often collapses when prompts fade, delays increase, or task demands shift during novel tasks. Accurate responding does not necessarily indicate functional problem solving. In many cases, the controlling variables reflect restricted stimulus control rather than established speaker–listener mediation. This presentation examines verbal mediation and joint control as critical teaching tools in moving learners from basic skills into complex learning. Joint control occurs when a rehearsed verbal operant and a tact jointly evoke a response under shared topographical control. Participants will analyze how self-echoic rehearsal, tacting, and intraverbal responding interact to support responding across delay, distance, and novelty. Emphasis will be placed on identifying missing prerequisite repertoires, establishing overt verbal mediation, and transferring control to covert mediation. Applications will include multi-step instructions, mediated selection, recalling past events, and reading comprehension—supporting more durable and generative repertoires.

Credits

1 CEU · 60 min

From the talk

What was covered

Why a right answer is not proof of problem solving, and how to teach the thinking steps behind it. A recap of Dana Pettus's talk on OpenCEU.

  • A correct answer alone does not prove understanding. Ask if it came from memory or from the learner talking through the task.
  • Run joint control (two cues controlling one response) programs with items and instructions the learner already knows. These are not language programs.
  • Prompt the learner to repeat the key words out loud two or three times. End with the learner saying it, then cue them to go.
  • After an error, bring the learner back and run the prompting procedure again instead of pointing to the right answer.
  • When a learner keeps missing, try a few fixes. Shorten the echoic, cut the distance, prime the tacts, or practice at the table first.
  • Make repeating instructions out loud a daily routine, then fade it so the rehearsal happens quietly.

Why a Correct Answer Can Hide Restricted Stimulus Control

Dana Pettus opened with a quick test. What is two plus two? Everyone knew four right away. Then came a harder question. What did you have for breakfast last Sunday? The chat went quiet for a while. Most people had to work for that one. Pettus walked through the steps out loud. Last Sunday meant the cottage. Breakfast was a personal job that day. Eggs and bagels were in the bag. So eggs and bagels was the answer. That answer came from covert problem solving (quiet thinking that finds answers). Two plus two did not.

Here is the problem with that approach. A lot of teaching treats every question like two plus two. We teach the answer and we reinforce the answer. We skip the quiet steps in the middle. The learner looks fine on the data sheet. Then prompts fade, delays grow, or the task shifts. Performance falls apart. That pattern means something about stimulus control (the cues that control a response) is missing.

Part one of this series set up the basics. A pure tact (naming what you see) is rare in daily life. You do not walk into a room and say chair. Most naming is an intraverbal tact (naming cued by a question too). In a color task, the answer blue has two sources. The question pulls for a color. The picture pulls for blue. That is convergent control (several cues, one response). Divergent control (one cue, many responses) works the other way. What color can evoke blue, red, or green.

Part one also used an autoclitic frame (a sentence starter that cues wording). Ask what does it say. Teach the learner to start with it says. That frame narrows the field. It pulls for meow or moo, not for car or kitchen. The learner is prompting themselves. Pettus then drew a line many of us miss. A simple intraverbal (answering words with other words) comes from direct reinforcement. A complex intraverbal (a harder question) works differently. The question sets off quiet verbal behavior (silent self-talk) first. That quiet behavior and the question together evoke the answer.

Here's a stimulus, here's a response.

From the talk — Dana Pettus

Problem Solving, Verbal Mediation, and Working Memory

Pettus leaned on Skinner's definition of problem solving. By that standard, most of what we ask learners to do counts. Even non-identical matching (matching items that are similar but not identical) is a small version of it. You see a folding chair and a recliner. You tact both as chair. They match, so they go together.

Other examples are easy to spot in the clinic. A learner is told to get a coat and boots, then line up. They have to hold coat, boots, and line up while they move. Complex questions work the same way. The breakfast question had no stored answer waiting. The answer had to be built.

Verbal mediation (talking yourself to an answer) is the tool for that work. Pettus gave plain examples. Repeating a grocery list on the way to the store. Using subcategories when you list. Ask someone to name countries and they stall. Then they prompt themselves with Europe, then South America. New answers come. Listeners in a webinar do it too. They repeat key words, reword them, and take notes.

This connects straight to working memory (holding information while you use it). Holding a thought in your head while you process it is the core of it. Pettus pointed to research that supports this link. Blocking verbal mediation hurt performance on false belief tasks (tests of guessing another person's mistaken belief). Visual imagining helped with a categorization task. Self prompts helped with a listing task. Teach this early, even in simple form. It can support executive functioning (planning and self-monitoring) skills down the line.

Skinner defined problem solving as any behavior which through the manipulation of variables makes the appearance of a solution more probable

From the talk — Dana Pettus

What Joint Control Actually Is

The next activity made the concept concrete. A number appeared on screen. Hold it. Then an array of similar numbers appeared. Find the match. The chat reported rehearsal and scanning. That is the whole process. You rehearse the digits. You scan the array. You tact each candidate and reject it. Then one matches and your hand stops.

Joint control (two cues control one response) is the technical name. A response is evoked by one SD (the cue that starts responding). Rehearsal carries it across the gap. Then a second stimulus comes under control of the same response. Two cues work at once and both strengthen it. Working memory is the broad term. Joint control is one slice of it.

Pettus used a peanut butter example. The empty jar is the cue. There is a long delay before the store. So rehearsal fills the gap: peanut butter, peanut butter. In the aisle, the scanning starts. Jam, no. Flour, no. Butter, no. Peanut butter, yes. The reach is controlled by both the rehearsed words and the tact on the shelf.

A shape matching activity showed the same four steps. Name or describe the model, such as a square riding a rocket ship. Rehearse it. Scan and tact each option. Stop and select on the match. Skilled adults do this without noticing. Many of our learners do not. Think of the learner sent to the bedroom for a sweater. They get three feet, see a toy, and the instruction is gone. They cannot hold that information in their head.

We do this automatically all the time and we don't think about it.

From the talk — Dana Pettus

Where the VB-MAPP and ABLLS Leave Off

Pettus mapped this against the tools we already use. The VB-MAPP (a common verbal skills assessment) is full of items that need mediation. Following three step directions. Continuing pattern sequences. Answering up to 300 intraverbals. Describing events. Yet the assessment stops at 48 months. Some complex conditional discriminations (choices based on two cues) may not show up until age five or later.

The ABLLS (another verbal skills assessment) reaches further. It includes questions about past and upcoming events. It includes novel questions, telling about an experience, and telling stories. Still, the jump is large. Answering concrete intraverbal questions is one thing. Telling a story about yesterday is another. Neither tool assesses the mediation skills that bridge the two.

That gap is where learners stall. They pass the simple items and then sit. We keep running more exemplars (practice examples) of the hard item. The missing piece was never in the plan, so progress stays flat.

And we have learners that go way past four and if we want to be supporting them we need to be looking at skills beyond just the VB map.

From the talk — Dana Pettus

Program One: Following Instructions Over Distance

Dr. Degli Espinoza suggests starting with instructions over distance. The learner hears a request and finds the item somewhere in the room. Some learners already handle this on their own. Others need the rehearsal made overt (said out loud) before they can do it.

Check three prerequisites first. The learner can identify every item used in the program. The learner self-echoes (repeats the item's name back) when picking items, so naming is already a habit. The learner has some echoic repertoire (repeating words they hear), since rehearsal starts out loud. That last part matters. We make the quiet step loud so we can see it happen.

The procedure is short. Give the instruction: go get your pencil. Prompt a repeat right away. The key word is enough, so pencil works. Point to yourself and say pencil. Point to the learner to cue the repeat. Go back and forth two or three times. End with the learner saying it. Then give a clear cue to respond, such as go do it, or a small nudge.

Then fade. The pointing prompt can drop out fast. The finished program is just the instruction and the trip across the room. The rehearsal keeps happening, but it happens quietly. When the learner starts rehearsing on their own, prompt sooner and less. Remember what this program is not. It is not a receptive identification (matching a spoken word to an item) program.

We're practicing this with items they already know.

From the talk — Dana Pettus

Error Correction and Troubleshooting a Stalled Program

Error correction is where this program lives or dies. Do not point to the right item. If you prompt and reinforce correct selection, the mediation never gets built. Stop the learner. Put the item back. Walk back to the start and run the prompting procedure again. The target is the learner's own self-echoic (repeating words back to yourself) as the prompt.

If errors continue, Pettus offered five fixes. Add repetitions of the key words, two or three in a row. Simplify or shorten the echoic, so backpack replaces watermelon backpack. A hard echoic eats the effort needed to rehearse. Cut the distance to three or five feet, then build it back up. Prime the tacts by having the learner name the items in the array first. Practice the same request at the table with no distance at all.

Walking a learner back can set off problem behavior. Shorten the distance so the trial runs more errorlessly (with fewer mistakes). Keep the correction warm and matter of fact. A kind try again does more than a sharp no. You can also start with a seated program instead, where there is no transition to manage.

One attendee asked about non-vocal learners. Sign works well here. Model the sign and let the learner's own sign serve as the rehearsal. AAC (a speech app or device) is trickier. The button stays visible on the screen. That turns the task into a visual match rather than rehearsal.

we want to bring them back and repeat the joint control prompting procedure

From the talk — Dana Pettus

Scaling Up: Multiple Items, Multi-Step Directions, Counting, and Blocks

The next program is selecting multiple items in order. Say give me the ball and cat. The learner repeats ball cat two or three times, then responds. Order is the point. Correct order is your evidence that rehearsal happened. Fade the prompt as soon as you can. Some learners need scripted combinations at first. Mastery means any novel combination of known items. Then add a third and fourth item.

Multi-step instructions come next. Each instruction must already be solid on its own. Teach the learner to shorten long directions to key words. You may need to model that a few times. Go get your lunch bag and head to the table becomes lunch bag table. Rehearse, then go do it. Then raise the demand in steps. Add three and four part directions. Add a detail that narrows the choice, so blue pencil instead of pencil. Add prepositions (words like under or on), such as the book under the table when a book sits on it. Add locations, so the bathroom rather than the closet. Finally combine them: bedroom, blue car, on bed.

Counting out a set amount uses the same mechanism. Say give me five pom poms. Prompt the target number: five, five, five. The learner counts and holds that five. The match between the count and the rehearsed number is the cue to stop. Pettus flagged a common staff error here. Many of us praise the moment the learner hits five. We never learn whether they would have kept going.

Block imitation (copying a block structure from a model) closes the set. Start block by block with the model in view. Prompt a tact of the first block, find the match, place it, repeat. Then delay it. Show the model, prompt tacts in build order, rehearse, remove the model, wait, and build while labeling. Recall needs three parts. Tacting in the moment, since you cannot recall what you never labeled. Rehearsal. Then mediation while generating the answer. Some learners hold the image visually instead. The strategy is the same, but words are easier to prompt than pictures.

This one, you are giving them that two to three seconds.

From the talk — Dana Pettus

Reading Comprehension and Carrying This Into the Day

Pettus ended with a short passage and one question: where did James go last night? The chat listed the skills involved. Decode (sound out) the words. Know what they mean. Know that where calls for a location, not for blocks. Recall the events in order. Scan the passage for the answer. That is a lot of behavior inside a simple question.

The failure pattern is familiar. The learner misses, the therapist says go reread it, the learner rereads and still misses. The passage was never the problem. They could not hold the key terms while they scanned. So teach that directly. Read the question. Pull out the key terms: James go last night. Rehearse them. Then scan the passage. When the rehearsed words match "came to my house last night," the answer comes out.

Two cautions came with it. Some learners benefit from repeating the passage language back, which also shows you they attended. And check comprehension level, not decoding level. Plenty of learners decode far past what they understand. The reading sounds great and the meaning is not there.

Then move it into the day. Make repeating instructions a routine, out loud at first and quiet later. Normalize tacting, since being verbally present feeds everything else. Practice the self-echoic across the day. It helps in math too. Learners must hold key terms from a word problem. They need to remember if they are adding or subtracting. When something stalls, ask what you do in your own head to solve it. Ask a colleague how they do it. Find where the control broke down: attending to the question, tacting, or a missing mediation step. Then teach that piece instead of handing over the answer.

When I hear an instruction, I say it back.

From the talk — Dana Pettus

Common questions

What is joint control in ABA (applied behavior analysis)?▾

Joint control happens when one cue evokes a response. Rehearsal carries that response forward. Then a second stimulus takes control of the same response too. Both cues strengthen it at the same time. The peanut butter example shows it. You rehearse the word peanut butter, then scan the shelf. The reach is controlled by both the rehearsed word and the tact of the jar.

How is joint control different from working memory?▾

Working memory is the broad ability to hold information while you process it. Joint control is one specific version of that. You repeat information to yourself, and that repeated information matches something in the environment, which evokes the response. Pettus described joint control as a possible subsection of working memory.

What prerequisites does a learner need before these programs?▾

The learner should already identify every item or follow every instruction used in the program. This is not a language teaching program. They should self-echo while selecting items, so naming is already habitual. They also need some echoic repertoire, since rehearsal starts out loud so you can see and prompt it.

Can you teach joint control to a non-vocal learner?▾

Yes, if the learner uses sign. Model the sign and let the learner's own sign work as the rehearsal step. AAC is harder, because the button stays visible on the device. That turns the task into a visual match between the screen and the item rather than true rehearsal.

What do I do if the learner keeps making errors?▾

Do not prompt the correct answer, because that teaches selection without mediation. Bring the learner back and run the prompting procedure again. If errors continue, try a few fixes. Add repetitions or shorten the echoic. Reduce the distance, prime the tacts in the array, or practice the request at the table first.

Why is multiple exemplar teaching (teaching with many practice examples) not enough for complex skills?▾

Complex responses depend on several sources of stimulus control at once, not a stored answer. More exemplars can build a memorized response while leaving the mediation skill untaught. The learner then falls apart when prompts fade, delays grow, or the task is novel.

About the speaker

Dana Pettus is a registered behavior analyst and a board certified behavior analyst. Pettus works as a clinical manager at AlphaBee and has been in the field since 2013. Pettus works mostly with children and teens with autism, plus some clients with dual diagnosis (autism and another condition). This talk was part two of a three part series on speaker-listener repertoires (talking and understanding skills) for complex learning.

This summary was generated from the recording’s transcript. Quotes are taken word for word from the talk.

What you'll learn

  1. 11. Define joint control as the onset of joint stimulus control over a common response topography and distinguish it from simple listener responding 2. Analyze learner performance to determine whether correct responding reflects mediated problem solving or restricted stimulus control. 3. Identify prerequisite repertoires (e.g., self-echoic, tact, listener, and mixed operants) necessary for establishing functional speaker–listener interactions. 4. Design instructional procedures that establish overt verbal mediation and systematically transfer responding to covert mediation. 5. Apply joint control procedures to teach a variety of skills including multi-step instructions, mediated selection, listing, and early academic problem-solving tasks.

Related talks

Talks that cover the same concepts

More from alphabeePRO

OpenCEU is supported by advertising from third-party sponsors. Sponsors are clearly labeled, have no influence over course content, instructors, or CEU decisions, and never appear on certificates. The BACB does not sponsor, approve, or endorse OpenCEU sponsors or their products.