Using RFT to Promote Generative Language: Complexity and Flexibility

My consultation work always begins with some problem—in some way, someone is “stuck” or they wouldn’t be looking for help from me. In early intervention work that is often a learner who isn’t making progress, and one of the first things I consider is whether the response requirements are too complex and need to be broken down, or whether we haven't built enough flexibility into the repertoire. The HDML framework gives us a way to think about this more precisely. In my last post, I described the levels of relational responding. Today I'll focus on two of the four dimensions of relational framing: complexity and flexibility. (I'll cover derivation and coherence in my next post.)

Complexity

When planning intervention, we want to systematically increase the complexity of response requirements to strengthen responding while maintaining a supportive context for learning. Relational complexity refers broadly to the level of detail or intricacy in a particular relational framing response, and several variables contribute to it.

Consider the difference between asking a child to compare an apple and an orange as same or different in terms of their color versus their shape versus what parts are edible. That involves more complex response requirements than relating "apple" as the same as "pomme," or simply relating apples and oranges as "different." The number of stimuli being related matters—the more stimuli involved, the more complex the response. The number of types of relations matters too. If A is different from B, the derived response is symmetrical—B is different from A. But if A is more than B, then B is less than A: two types of relations are involved, and the response is more complex. And the number of components in the relevant contextual cues adds complexity as well—multi-component cues such as for a particular relation and for particular properties or functions would be more complex than responding to a single requirement.

In practice, this means starting simple and building out. When working within the early relational pattern of coordination, relational responses are relatively simple—we start with only a few stimuli that are related, and coordination is itself a simpler pattern (there is only one type of relation: sameness). As difference is introduced, opportunities arise to significantly increase complexity, such as by adding additional cues (e.g., different or same color versus shape, but also texture, material, and so on). Introducing patterns of comparison, opposition, spatial, and temporal relations continues the progression in complexity of relations and expansion of cues.

Flexibility

Relational flexibility refers to the ease with which a response pattern can be changed by introducing a new contextual cue. Given our overarching goal of supporting psychological flexibility, attending to relational flexibility specifically can be a particularly important aspect of intervention—and one that is easy to underestimate.

Consider what inflexible relational responding looks like in practice. If you've ever tried to teach a learner to select a "different" item in a match-to-sample format after they've had extensive practice selecting "same" items, you've seen relational inflexibility in action. The well-practiced response pattern is difficult to shift, even when the contextual cue has changed. This is a useful reminder that every time we build fluency with one pattern of responding, we're also potentially creating a repertoire that will need to be flexible enough to change when new cues are introduced.

The introduction of new patterns of relational responding also allows for increasing the flexibility of responding to cues for those relations. Whether this means fluently switching from one relational cue to another (e.g., same vs. different) within a match-to-sample format, or being able to give the "wrong" response when asked to, relational flexibility can be an important dimension to establish and strengthen.

This connects directly to the broader framework of psychological flexibility that I have argued is the aim of all of our work. Relational flexibility at the level of early language development is a critical behavioral foundation for psychological flexibility. At its core, this is what flexible behavior looks like at any level of complexity: the ability to identify what is most important to attend to in a given context, and shift or persist in behavior accordingly.

I hope you've found this helpful—as always, I'd love to hear your thoughts and questions. Next time, I'll turn to derivation and coherence, and how they connect to some familiar clinical challenges around rigidity and rule governance.