Correspondence Training

Description of the Strategy

The term correspondence refers to congruence between verbal and nonverbal behavior. For example, if a child says that he or she is going to play with Legos later in the morning and actually plays with Legos when the time comes, then a specific instance of correspondence between saying and doing has occurred. Correspondence may also occur if an accurate verbal report follows the behavior. For example, if the child played with Legos in the morning and then in the afternoon said that he or she played with Legos that morning, then a specific instance of correspondence between doing and saying has occurred. Thus, correspondence involves agreement between verbal and nonverbal behaviors that may follow two temporal sequences: saying-then-doing and doing-then-saying.

If individuals typically do what they say for many different behaviors across different environmental settings, then verbal-nonverbal correspondence is well established and generalized. Generalized correspondence is an important skill for a number of reasons. First, if people promise to do something, you can usually have confidence that they will fulfill their promise. Second, if you ask people about past behaviors, their answers are usually accurate. In many ways, important human traits such as trustworthiness, honesty, dependability, reliability, and so on are judged by high levels of verbal-nonverbal correspondence. An equally important function of correspondence is its potential use as a behavior change strategy for nonverbal target behaviors that are less accessible for direct modification. Another important function of well-established correspondence is the role that verbalizations may have in mediating generalizations. That is, verbalizations (i.e., covertly talking to yourself) can occur at any time in nontreatment settings and serve as cues for the generalized performance of nonverbal target behaviors.

Most young children do not engage in high levels of verbal-nonverbal correspondence. Correspondence training was developed to teach children increased convergence between saying and doing. The most effective procedures for correspondence training consist of positive reinforcement, modeling-imitation, and chaining. Reinforcement seems to be a critical procedural component, and in order to be maximally effective, it must be contingent upon correspondence between saying and doing. That is, if children say they will play with blocks, then they receive a reinforcer after they play with blocks and do not receive a reinforcer if they do not play with blocks. Since reinforcement must be arranged contingent upon both the verbal and nonverbal behavior (often in different settings as well as temporally separated), a very practical concern in correspondence training is direct access to both the verbal and nonverbal behavior so that observational data can be used to determine when reinforcement should occur. Reinforcement of the verbal behavior without verifying that the nonverbal behavior occurred or vice versa does not increase verbal-nonverbal correspondence.

Modeling both the verbal and nonverbal behavior for children and having them imitate correctly can facilitate specific instances of correspondence that then can be reinforced. If the verbal behavior is simple, modeling is usually unnecessary. For example, if the verbal behavior is answering the question, “Are you going to play with the soccer ball at recess?” and the child says “Yes” and the nonverbal behavior is a simple selection of a toy or activity, then modeling probably would not be necessary. If either behavior is [Page 791]more complex, though (e.g., social behaviors such as sharing or making a new friend), then modeling-imitation could be critical.

...