Mastering Mand Training: Avoiding Top 5 Mistakes for Effective Communication
The mand is unique from all other verbal operants as it is the only operant driven by the motivating operation and reinforced by the specified reinforcer. Early in a child's development, mands emerge spontaneously, epitomized by utterances like "mama," "dada," or "baba" for bottle. However, for children with autism, expressing wants and needs through manding can present challenges, necessitating targeted interventions. Before embarking on mand training interventions, it's crucial to address the 5 common mistakes that may hinder progress.
Mand Training Error # 1: Forgetting the MO
Mands are based on the motivation for the item, activity, or information. The client wants a cookie, they say “cookie”, they want you to blow bubbles, they say “blow”, they want to know where the toy is, they say “where?”. It is easy in our drive to get in the required number of trials to push for more responses, but if you don’t start with the client’s motivation, you are just training scripted responses that have no meaning. To fix this issue, ensure that you:
· Conduct preference assessments regularly to verify the MO
· Varying reinforcers to avoid satiation
· Have potential reinforcers visible in the environment, but under control of the therapist to prompt the mand when needed
Mand Training Error # 2: Creating Prompt Dependency
“What do you what?” This is a common question in ABA therapy when prompting mands, but also with typical developing children in real life. There is nothing inherently wrong with asking “What do you want?” but if all your mands a prompted with this question, how will your clients learn to independently mand for things? Before jumping into the cycle of “what do you want?” over and over again, consider if it might be better for your client to also include using a model prompt that you fade over time.
Let’s look at an example of both prompting options. Let’s pretend your client is showing desire for bubbles so you prompt “What do you want?”. Well now the antecedent is the question “What do you want?” instead of just the motivation for the bubbles. This might be ok, they are still interested in bubbles, but if we do this over and over again the control of the response “bubbles” could transfer over to the question instead of the MO. If instead the client shows that they want the bubbles and you model for them “I want bubbles” with appropriate pausing to give them a chance to complete the sentence independently, then we are teaching them to make requests for desired items on their own. We also have a way to fade the prompt by saying less and less words and increase pausing, which is difficult to do when using the verbal prompt.
Mand Training Error # 3: Training Vague Requests
Before getting into the field of ABA, I was a speech language pathologist assistant for over 6 years. During that time, I was guilty of some common mand training errors, but the biggest one was vague requests. One of the first mands (though I was unaware of verbal behavior and we didn’t call them “mands”) was to teach my clients to ask or sign for more, not “more chips” or “more bubbles”, just “more”. So after weeks of teaching my client to ask for “more” without a specific request, we had a child that would come to me and other adults wanting something and just saying “more, more, more”.
Guess what? They now had a mand and we STILL didn’t know what they wanted! It was also often used incorrectly because they would ask for “more” to initiate accessing something, like asking for “more” when they wanted a new item or activity instead of just at times when they actually wanted more of something. When teaching requests, it’s important to start with the specific then add modifiers like first teaching “bubbles” then teaching “more bubbles” later on. “More” isn’t the only potential issue, but also teaching “my way”, “please”, etc.
Mand Training Error # 4: Mindreading
It is obvious to follow through on requests immediately to reinforce the mand, but what do you do when a client mands for something that you know they don’t want? You probably know your client very well and know when they make an incorrect mand, but other people do not and motivations can also change, so we should not try to “mindread” our clients and act based off of what they say or sign. Clients can make incorrect requests due to scrolling, faulty mand training, or other errors, but the goal of mand training is to teach the client that when they request for something it has an effect on the environment. They same goes for incorrect requests.
I had a client that had previous mand training which inadvertently trained scripted responses instead of actual mands based on the client's motivation. When we asked “What do you want” (which is still appropriate to do at times) he would make a request for something that he had been reinforced for saying in the past, but didn’t necessarily want at that time. Now, we could have just stopped him and prompted another mand, but that wouldn’t teach him the consequence of his verbal behavior. Instead, when he made an incorrect request, we would still offer the item to him, then when he declined it, we would reset and model the correct request, ultimately fading the model over time.
Mand Training Error # 5: Failing to Use Error Correction
As I mentioned above, scrolling can be problematic when it comes to mand training. You could have a client that just gives an incorrect response, or they could give multiple responses at once, like requesting “juice, milk, water” when they want water. If you then reinforce them with the water because that’s what they are reaching for, you’ve just reinforced the response “juice, milk, water”. Error correction is vital when training mands. While there may be variations for certain clients, in general, error correction for mands should involve the following steps:
Client makes an error in form or scrolled response
Reset the environment by removing or moving back the reinforcer for 3-5 seconds
Represent the reinforcer and immediately prompt the correct response
Then still provide the reinforcer, but for less time then they would have received if they didn’t make an error
If we skip the steps of resetting the environment, you risk reinforcing a chained response of error >> prompt >> reinforcement which always starts with the errored response.
By addressing these five pitfalls, behavior analysts can optimize their interventions and effectively train clients to express themselves to get their wants and needs met. If you want a more in-depth look at increasing communication using verbal behavior and Functional Communication Training (FCT), check out our Masterclass: Increasing Communication through Functional Communication Training.
