RBT Discrete Trial Teaching Questions

A neutral stimulus becomes a conditioned reinforcer by being paired repeatedly with an established reinforcer. Saying 'Great job!' while handing Max the cracker pairs praise with a primary reinforcer on every correct response, so praise can gradually gain reinforcing value of its own. Praising about a minute later breaks the pairing; the two must occur together, with praise just before or at the same time as the cracker. Stopping crackers abruptly and using only praise removes the effective reinforcer before praise has any value, which can cause correct responding to drop. Praising only when he refuses the cracker never pairs praise with the edible and may accidentally reinforce refusals. Later, the BCBA may direct fading the crackers as praise gains strength.

In match-to-sample trials, the learner is given a sample stimulus and must pair it with the corresponding comparison stimulus from an array. The dog card given to Jae is the sample, the item to be matched. The dog card on the table is the correct comparison stimulus, the choice that matches the sample, so it is not the sample itself. All three cards on the table make up the comparison array, which contains the correct comparison plus the hat and spoon as distractors. The instruction 'Find the same' tells Jae what to do and is part of the discriminative stimulus, but it is not the stimulus being matched. Knowing these terms helps the RBT set up and record each trial correctly.

A vocal discriminative stimulus in discrete-trial teaching should be a clear, concise statement or instruction, so 'Touch your nose.' tells Hana exactly which response is expected. Phrasing the instruction as a question, such as asking whether she wants to touch her nose, implies the response is optional and invites a 'no' or a head shake instead of the target behavior. Asking 'Would you touch your nose now?' has the same problem and adds extra words that may compete for her attention. Adding 'okay?' to the end of an otherwise clear instruction turns it into a request for agreement, again making compliance seem optional. Delivering the same brief, consistent instruction across trials also helps bring the response under stimulus control.

Instructional control develops when following the RBT's instructions reliably produces reinforcement. Giving easy, mastered instructions, such as clapping or touching her head, and reinforcing Ivy's compliance builds a history of success and pairs the RBT and the table with reinforcement before harder work is introduced. Starting with the hardest new targets while she is most alert makes errors and refusals likely, which can strengthen escape. Presenting new targets and using full physical prompts when she refuses can make the setting more aversive before rapport and compliance are established. Holding all reinforcement until she finishes a block of new targets sets the requirement too high for a learner who often refuses, so she may rarely contact reinforcement at all.

A gestural prompt is a movement by the instructor, such as pointing toward or motioning at the correct stimulus, that guides the learner's response without physical contact. Pointing toward the red block without touching it delivers exactly that. Guiding Ben's hand to the red block is a physical prompt, a more intrusive level that the program does not call for. Saying 'red' again as he reaches is a verbal prompt, which adds a spoken cue rather than a gesture. Sliding the red block closer to Ben than the others is a positional prompt, which changes where the stimulus sits. Using the specified prompt type matters because the BCBA's fading plan and the RBT's prompt-level data both depend on it.

Stimulus generalization occurs when a skill learned under one set of conditions also occurs with different people, materials, or settings. Having different RBTs run the same trials in the kitchen and at home both promotes this spread and checks whether Hugo's responding has generalized beyond one RBT and one table. Response generalization refers to the learner emitting new, untrained responses that produce the same outcome, such as pointing to the cup instead of touching it, which this change does not target. Skill maintenance is about the skill lasting over time after teaching ends, not across people and places. Interobserver agreement checks require two observers recording the same session at once to evaluate data accuracy, not separate sessions in new settings.

Stimuli in an array should be placed the same distance from the learner so that no item is easier to reach or more noticeable, giving each an equal chance of being selected. Placing all three pictures equidistant removes the unplanned cue. Leaving the target closer adds a positional prompt the program does not include, so a correct response may reflect proximity rather than discrimination, and the data would overstate Leo's skill. Moving the target to the middle spot on every trial creates a position pattern Leo can follow without learning the picture; target placement should be rotated. Removing the other two pictures turns the trial into a single-item presentation, which is a different teaching step that the BCBA has not directed.

A prompt is a supplementary antecedent stimulus added to the instruction to increase the likelihood of a correct response. The instruction 'Clap hands' is the discriminative stimulus, and the RBT's simultaneous clap is an extra, model-type cue delivered before Mia responds, which makes it a prompt; that is also why the response is recorded as prompted. Calling the clap a reinforcer is incorrect because reinforcers follow a response, and this clap came before Mia's. Calling it the discriminative stimulus confuses the extra help with the instruction that should eventually control the response alone once prompts are faded. An establishing operation changes how valuable a reinforcer is, such as hunger increasing the value of food, which does not describe this demonstration.

The learner should finish consuming or engaging with the reinforcer before the next trial begins, so waiting until Nia finishes chewing is the best change. Presenting an instruction while she is eating competes with the reinforcer for her attention, makes errors more likely, and blurs the separation between one trial and the next. Giving the next instruction even sooner worsens that overlap. Giving the pretzel only every other trial changes the reinforcement schedule, which is a decision for the BCBA, and may weaken correct responding while the skill is still being acquired. Repeating a missed instruction louder while she is eating adds a repeated instruction, which can teach her to wait for several, and still competes with the reinforcer.

A tact is a label controlled by a nonverbal stimulus, so a tact trial presents an item and reinforces the learner for naming it. Holding up the toy car and praising Eli for saying 'car' is a tact trial because the car itself evokes the label. Saying 'car' and praising him for repeating it is an echoic trial, since his response is controlled by the RBT's spoken word. Saying 'touch car' and praising him for touching it is a listener, or receptive, trial; he responds to an instruction without speaking. Asking what he rides to school and praising 'bus' is an intraverbal trial, because a spoken question, not the item, evokes the answer. Choosing the correct trial type keeps the tact data accurate.