Which training approach provides practice with a variety of stimulus conditions, response variations, and response topographies to promote generalization?

• A. Naturally Existing Contingency
• B. Multiple Exemplar Training
• C. Momentary Time Sampling
• D. Interobserver Agreement

Answer: (B)

Multiple exemplar training builds generalization by giving the learner extensive practice across many different instances of the target skill. By varying the stimulus conditions (different items, settings, people), the response variations (different forms of the response), and the response topographies (different physical forms of the behavior that achieve the same function), the learner learns to apply the skill in new, untrained situations. This broad exposure helps the behavior transfer beyond the exact training moments, because the learner understands the underlying relation or function rather than relying on a single cue or response form. For example, if teaching a child to request a preferred item, you’d practice with many different items, in various rooms and with different people, and allow different ways to request (verbal, sign, gesture, or use of pictures). The other approaches don’t systematically promote this broad, varied practice: one focuses on using real-world contingencies without controlled variety, another is a data-collection method that records behavior at intervals, and another concerns reliability of observers rather than training for generalization.