BCBA® task B.13 · Area B

B.13 Stimulus discrimination

3 min read · BCBA® Test Content Outline, 6th edition

What it means

Discrimination is demonstrated when responding differs reliably across stimuli because of the relevant learning history. An SD is correlated with greater availability of reinforcement for a particular response than comparison conditions; an S-delta is correlated with the absence of reinforcement for that response. In a simple discrimination, one stimulus decides whether a response pays off. In a conditional discrimination, which choice is correct depends on another stimulus, the sample, as in matching-to-sample. A generalization gradient shows how responding changes as test stimuli move away from the trained SD: a steep gradient means tight control, a flat one means little control along that dimension.

Everyday example

Simple: a dog sits when it hears "sit" and not when it hears "sip." Conditional: in a matching task, the learner touches the dog picture when the sample is a dog and the cat picture when the sample is a cat; the same picture is right or wrong depending on the sample.

Common confusion

Responding controlled by position, one particular item or a prompt may demonstrate discrimination of those features, but does not establish the intended discrimination. Test whether responding tracks the relevant stimulus relation when incidental features change. Faster or less prompted responding alone does not establish discrimination. An S-delta is an antecedent correlated with no reinforcement for a specified response, not a punishing consequence.

Key points

  • Discrimination is shown when responding differs reliably across stimuli (SD versus S-delta).
  • Position, item or prompt control can demonstrate an unintended discrimination. Correct answers alone do not establish control by the intended stimulus relation.
  • In a simple discrimination, responding depends on one stimulus (SD versus S-delta); in a conditional discrimination, which comparison is correct depends on another stimulus, the sample (as in matching-to-sample).
  • A stimulus generalization gradient shows how responding changes as test stimuli differ from the trained SD; a steep gradient indicates tight stimulus control, a flat one little control along that dimension.

Examples from practice

Example 1

A warehouse picker scans boxes with a green sticker (scanning them earns credit) and skips boxes with a gray sticker (scanning them earns nothing). Green = SD; gray = S-delta.

Example 2

A new bartender pours the house red when a customer says "red" and the house white when the customer says "white," with the bottles moved around each night.

Non-example

A trainee who picks the "urgent" folder whenever it is on the top of the pile, whatever its label, is not discriminating urgency. Position is controlling.

Check yourself

One question from the course lesson on this task. Choose an answer to see why it is right or wrong.

Which data give the best evidence that a new receptionist discriminates "visitor badge" from "staff badge" when deciding whether to ask for ID?

Show the answer and the explanations

Answer: D. He asks for ID from 95% of visitors and 5% of staff

  • a. Speed alone is not evidence of discrimination. He could be faster at asking everyone.
  • b. Almost equal responding to both badges shows little discrimination.
  • c. High responding to one stimulus says nothing about the other. He might ask everyone.
  • d. Correct. Responding is high to the SD and low to the S-delta: a reliable difference.

The full lesson is in the course

The course lesson on this task goes further, with more sections, more examples and more practice questions, and keeps a record of what you have mastered.

Open the lesson Practice this task

Try a sample free, without an account

Key terms

Each term opens its definition in the ABA glossary.

All tasks in area B All 104 tasks ABA glossary: every term in one page

Independent study resource. INTERLAZA is not affiliated with or endorsed by the BACB. The explanations and examples are original, are not official exam content and do not reproduce any exam item.