Mand training is a method used in applied behavior analysis (ABA) therapy to teach a person, often a child with autism, how to make requests. The word “mand” comes from “demand” or “command,” and it simply means asking for something you want or need. B.F. Skinner coined the term in his 1957 book Verbal Behavior, where he classified different types of communication by their function rather than their form. A mand is unique because it’s driven by what a person wants in the moment, and the reward for making the request is getting that specific thing.
Why Requests Come First
Skinner described several types of communication, which he called verbal operants. A “tact” is labeling something you see, like saying “cat” when a cat walks by. An “intraverbal” is responding to someone else’s words, like answering “blue” when asked your favorite color. A mand is different from both because it’s not triggered by something you see or hear. It’s triggered by something you want.
This distinction matters for therapy because manding is the most immediately rewarding form of communication a child can learn. When a child says “cookie” and gets a cookie, the connection between speaking and getting what they need becomes concrete. That direct payoff makes manding the natural starting point for building communication skills. Once a child learns that their words (or signs, or picture cards) reliably get them what they want, they’re more motivated to keep communicating, which opens the door to more complex language.
The Role of Motivation
The engine behind mand training is something called a motivating operation. This is simply whatever makes a person want something right now. A child who hasn’t had a drink in a while is more motivated to ask for juice. A child who just finished a big glass of juice has no reason to request one. The motivation has to be genuine for the training to work, because the whole point is connecting a real desire to an act of communication.
Therapists and caregivers use two main strategies to work with motivation. The first is capturing it: noticing when a child naturally reaches for a toy, looks at a snack, or shows interest in an activity, then using that moment to teach a request. The second is contriving it: setting up the environment so motivation is more likely to appear. Placing a favorite toy on a high shelf where the child can see it but can’t reach it is a classic example. The child wants the toy, and now there’s a reason to communicate.
How a Session Works
A typical mand training session looks less like a classroom lesson and more like structured play. The therapist identifies items or activities the child is motivated by, often through a preference assessment done beforehand. During the session, the therapist creates or waits for a moment of motivation, then prompts the child to make the request.
The prompt might be a full verbal model (“say ‘ball'”), a partial prompt (“ba…”), a gesture, or a visual cue like pointing to a picture on a communication board. The child makes the request in whatever form they’re working toward, whether that’s spoken words, sign language, or selecting a picture. Then they immediately receive the item they asked for. That immediate delivery is critical. The reinforcer for a mand is always the specific thing that was requested, not a generic reward like praise or a sticker.
Over time, prompts are systematically faded. A child who initially needed to hear the full word before repeating it might progress to needing only the first sound, then just an expectant pause from the therapist, and eventually requesting completely on their own. This fading process is gradual and follows the child’s pace.
Natural Environment Teaching
While some mand training happens in structured, sit-down sessions, much of it takes place in everyday settings. This approach, called natural environment teaching, embeds learning into the child’s regular routines and preferred activities. Instead of sitting at a table with flashcards, the therapist follows the child’s lead. If the child gravitates toward blocks, that becomes the teaching opportunity.
The environment itself becomes a teaching tool. Giving a child a bowl of cereal but “forgetting” the spoon creates a natural reason to request one. Blowing bubbles, then pausing and holding the wand still, gives the child a reason to ask for more. These setups feel organic to the child, and the skills learned in real contexts tend to transfer more easily to other situations. The goal is for requesting to become something the child does throughout their day, not just during therapy hours.
Tracking Progress
Mand training relies on careful data collection to measure whether a child is actually gaining communication skills. Therapists typically track three categories of requests: prompted mands (the therapist had to model or cue the word), unprompted mands (the child requested independently when an item was present), and spontaneous mands (the child requested something on their own initiative, without any setup). The shift from prompted to spontaneous mands over time is one of the clearest signs of progress.
A common tracking method uses daily cold probes. Once a day, the therapist checks each target word by creating a moment of motivation for that item and seeing whether the child requests it independently. A child typically has two to five active target words at any given time. When a child hits mastery criteria on a target, usually requesting it correctly several times in a row, that word is considered learned and a new one takes its place.
Therapists also track frequency: how many total requests a child makes per minute during timed sessions. These sessions are usually short, anywhere from a few minutes to 20 or 30 minutes, and shorter, more frequent sessions tend to produce better data than long, infrequent ones. Both the total number of mastered words and the rate of requesting are graphed daily, creating a visual picture of how quickly a child’s communication repertoire is growing.
What Mand Training Looks Like Over Time
Early mand training often starts with just a handful of highly motivating items. A child might begin by learning to request three or four favorite snacks or toys. As those become reliable, the targets expand to include actions (“push” on a swing), social interactions (“tickle”), and information (“where” questions). The form of the request can also evolve. A child who starts with single words may progress to two-word combinations (“more bubbles”) and eventually full sentences.
For children who are pre-verbal or minimally verbal, mand training doesn’t require spoken language. Sign language, picture exchange systems, and speech-generating devices all work within the same framework. The functional logic is identical: the child communicates a want, and the want is fulfilled. The specific modality is chosen based on what’s most accessible and effective for that child, and many children use a combination of methods as their skills develop.
The broader purpose of mand training extends beyond getting snacks and toys. Children who can reliably communicate their needs show fewer problem behaviors, because much of what gets labeled as “challenging behavior” in young children with limited language is actually an attempt to communicate. When a child has an effective way to say “I want that,” “stop,” or “help,” the need for crying, grabbing, or tantrums decreases. Building a strong mand repertoire gives a child more control over their own environment, which is the foundation for independence.