Leer en español: Por qué los niños pueden hacer algo con un adulto, pero no con otro
Understanding generalization, relationships, and the small instructional differences that affect children’s performance.
Have you ever noticed that a child can do something perfectly well with one adult but seems unable to do it with someone else? Maybe a child follows directions beautifully for one teacher but struggles with another. Or perhaps a therapist reports that a child has mastered a skill, but when someone else tries the same activity, the child does not demonstrate it.
This happens more often than people realize. It can be frustrating, particularly when everyone is trying to help the child learn. But instead of assuming that the child is being inconsistent or choosing not to cooperate, I think we need to look more carefully at what is happening.
Sometimes the difference has very little to do with whether the child knows the skill. It may have much more to do with how the skill is being taught, the materials being used, the environment, or the child’s relationship with the adult.
Sometimes One Adult Just Has a Way of Getting Through
I have worked with therapists who seem to have a particular ability to help certain children demonstrate what they know. It is not necessarily that they have some kind of magic. For whatever reason, the child responds better to that person.
Maybe the therapist has developed a particularly strong relationship with the child. Maybe they use a tone of voice or pace of instruction that the child finds easier to follow. Maybe they have discovered small ways to make the activity more interesting. Maybe they notice the child’s signals quickly and adjust before frustration builds.
Whatever the reason, the child performs better with that adult. When another therapist comes along and tries the same program, the results may be very different. That is when we need to investigate rather than immediately deciding that one person is doing something wrong.
The adult who gets the best response may have learned something important about the child. The next step is to understand what that adult is doing and determine which parts of the approach can be shared, practiced, or built into the teaching plan.
Look at How the Instruction Is Presented
One of the first things I want to know is whether the adults are actually presenting the instruction in the same way. Sometimes we assume they are doing the same thing because they are using the same data sheet or treatment plan. But when we observe them, we discover meaningful differences.
For example, imagine that we are teaching a child to identify pictures. One therapist places the cards flat on the table, while another holds them vertically in front of the child. Both therapists may be presenting the same pictures and giving the same instruction, but the child may respond much better when the cards are held up.
That difference matters. Maybe the child can see the pictures more easily, or maybe the vertical presentation helps them attend. Whatever the explanation, the instructional arrangement is affecting performance.
We should also look at the words being used, the timing of the instruction, the distance between the adult and child, the amount of waiting time, and what happens after the child responds. A small difference in any of these areas may change how clear or manageable the task feels.
This does not mean that every adult must copy another person exactly. It does mean that we should identify the features that are helping the child respond and decide whether they should be included in the broader teaching plan.
The Materials and Environment Matter
I have also seen children respond much better when we turn a learning activity into a game. Suppose we are teaching a child to identify colors. We could sit at a table and repeatedly ask the child to touch red, blue, or yellow. That may be appropriate for some children, but it may not be particularly interesting to this child.
Now imagine putting those same colors on a whiteboard and turning the activity into something more active. The therapist stands back, gets the child’s attention, and says, “Okay, are you ready? Touch red!” Suddenly, the child is engaged and participating.
We are still teaching the same skill, but we have changed the experience. Sometimes that change makes a tremendous difference in how successfully the child responds.
The environment can also affect performance. A child may respond at a quiet table but struggle in a busy classroom. They may identify pictures with one set of materials but not another. They may follow directions when seated near the adult but lose track of the instruction when they are across the room.
These differences are useful information. They help us understand what supports the child’s performance and what conditions still need to be taught. If we treat every change in performance as a problem with the child, we miss information about the teaching arrangement.
This Is Why Generalization Matters
In applied behavior analysis, we use the term generalization to describe situations where a learned skill occurs under conditions different from those in which it was originally taught. If a child learns to identify a car using one particular picture, we want them to recognize cars in other pictures and in the real world. If they learn to follow an instruction from one therapist, we want them to understand that instruction when other appropriate people give it.
Generalization does not always happen automatically. A child may learn a skill under very specific conditions and have difficulty demonstrating it when those conditions change. That is why we need to think about generalization as part of teaching rather than assuming it will happen on its own [1].
We can practice skills with different people, materials, and environments. We can vary the way instructions are presented and make sure children have opportunities to demonstrate what they have learned outside the original teaching arrangement.
We also need to make changes thoughtfully. If we change the adult, materials, room, and instructions all at once, it may be difficult to tell what affected the child’s performance. Gradual changes can help us see which differences matter and give the child a fair opportunity to adjust.
Generalization is not just a final test after teaching is finished. It is something we can plan for from the beginning.
Do Not Confuse Performance With Mastery
Suppose a child responds correctly on 100 percent of trials with one therapist. That is encouraging, but it does not necessarily mean the skill is fully established across people and settings. We need to consider the conditions under which those responses occurred.
Maybe the child has learned to respond to subtle cues from that particular therapist. Maybe the materials are always arranged in the same order, or perhaps the therapist is providing prompts without realizing it. Maybe the child has learned the routine in that room but does not yet understand how the skill applies elsewhere.
Those possibilities need to be examined before we conclude that the child has mastered the skill. A strong performance with one adult is valuable evidence. It tells us that the child can respond under at least some conditions. It does not tell us that every relevant condition has been taught.
Sometimes the child genuinely knows the skill but needs additional opportunities to demonstrate it under different conditions. Other times, we discover that the original teaching arrangement was unintentionally supporting responses that were not fully independent. Either way, the information helps us make better instructional decisions.
Data are most useful when they include the conditions surrounding the response. Who gave the instruction? What materials were used? Where did the teaching occur? What prompts were provided? How much time did the child have to respond? Recording those details can help a team see patterns that a single percentage cannot show.
Relationships Are Part of the Teaching Context
A child’s relationship with an adult can influence how comfortable, engaged, and ready to respond they are. A child may be more willing to try with someone who has spent time playing with them, noticed their interests, or responded consistently to their communication.
This does not mean that rapport replaces instruction. A warm relationship does not automatically teach a skill. It does mean that the social context of teaching matters, and we should not remove it from our analysis.
If a child performs better with one adult, we can ask what that adult has learned about the child. Does the adult offer more time to respond? Do they notice signs of uncertainty and adjust the task? Do they use preferred interests to make the activity meaningful? Do they give clearer feedback? Do they approach the child in a way that feels predictable?
The goal is not to label one adult as the favorite or another as the problem. The goal is to identify useful practices and help the child build successful relationships with more than one appropriate person.
Share What Works
This is one reason I encourage therapists and other team members to communicate with one another. If you have discovered that a child responds better when you hold the cards vertically, tell the other adults working with the child. If you have found that turning an activity into a game helps the child participate, share that information.
Those small instructional discoveries are valuable. They can help other adults become more effective and prevent unnecessary frustration for the child. Sometimes a simple conversation between therapists can explain why their data look so different.
I also think it is important to approach these conversations with humility. Instead of assuming that another therapist is doing something wrong, we can ask, “How are you running that program?” or “Can you show me what you have been doing?” We may discover something useful that we had not considered.
The adult who has been most successful may also need support. It is not enough to say, “The child does it for me.” The team needs to understand what is happening so the child can experience success with other people too. Demonstrating the approach, practicing together, and comparing observations can turn individual knowledge into shared teaching practice.
Teach the Child, Not Just the Program
Ultimately, the goal is not for a child to perform a skill perfectly with one particular adult. We want the child to use that skill in the situations where it matters.
That means teaching in ways that support generalization and paying attention when performance changes. It means looking at the instruction, materials, environment, prompts, timing, and relationships involved. It means treating differences between adults as information rather than as proof that the child is being difficult.
If a child can do something with one adult but not another, do not immediately assume the child does not know the skill. Look carefully at what each adult is doing and what the child may be experiencing. There may be an important teaching opportunity hidden in those differences.
And sometimes the best thing we can do is learn from the adult who has figured out how to help that child succeed.
References
[1] Stokes, T. F., & Baer, D. M. (1977). An implicit technology of generalization. Journal of Applied Behavior Analysis, 10(2), 349–367. [https://doi.org/10.1901/jaba.1977.10-349] [2] Schreibman, L., et al. (2015). Naturalistic developmental behavioral interventions: Empirically validated treatments for autism spectrum disorder. Journal of Autism and Developmental Disorders, 45, 2411–2428. [https://doi.org/10.1007/s10803-015-2407-8]