Conditioned motivating operation

A CMO has two linked effects

First, it changes the current effectiveness of some consequence as reinforcement or punishment. Second, it changes the current frequency of behavior that has produced that consequence in the past.

For example, encountering a locked cabinet while a desired material is inside may increase the value of a key and evoke asking for, finding, or using a key. The lock-key relation depends on learning.

Learning history is the key boundary

An unconditioned motivating operation has relevant effects without the learning history being analyzed. A CMO acquires its effect through a person's experience with related events and consequences.

The same visible event can have different effects for different people. A technical error screen may make help valuable for someone who learned that staff resolve it, while having no such effect for a new user.

A CMO differs from a discriminative stimulus

A discriminative stimulus signals that reinforcement is available for a response because of a prior differential reinforcement history. A motivating operation changes the current value of a consequence and behavior related to it.

One event can participate in both relations. Analysis should ask two separate questions: Is the consequence currently more or less valuable, and is its availability signaled? The BACB BCBA Test Content Outline explicitly includes distinguishing motivating operations from stimulus control.

Three proposed CMO variants

A tutorial on motivating operations describes three commonly proposed forms:

• Surrogate CMO, or CMO-S: a formerly neutral event acquires value-altering effects through association with another motivating operation.
• Reflexive CMO, or CMO-R: an event establishes its own worsening or removal as significant and evokes behavior that has prevented or ended it.
• Transitive CMO, or CMO-T: a condition changes the value of another stimulus as a consequence, as a blocked activity can make a needed tool valuable.

These labels summarize proposed functional relations. The letters alone do not replace data.

A fictional transitive example

Aya chooses to assemble a model with parts stored in a latched box. The closed latch makes the opener valuable only while access to those parts matters. Across eight blocked-access opportunities, Aya uses speech or AAC to request the opener in 6 of 8. When the box is already open, the same request occurs in 0 of 5 opportunities.

This pattern is consistent with a transitive CMO interpretation. It does not isolate the mechanism. The visible latch may also signal which response will contact reinforcement, and the conditions differ in more than one way.

Measure the operation and outcome separately

Define the suspected operation, when it begins and ends, the consequence whose value may change, the relevant response class, and the comparison condition. Record raw opportunities and behavior under both conditions.

Preference, latency, choice, work allocation, or reinforcer-assessment measures may help evaluate value. Response frequency alone can reflect availability, prompting, effort, rules, or other antecedent control.

CMO-R examples need ethical care

A warning correlated with worsening conditions can establish its own removal as reinforcing. Deliberately arranging threats, escalating demands, or distress to evoke escape carries substantial ethical and clinical risk.

Prioritize prevention, clear communication, choice, predictable routines, accessible breaks, and less intrusive supports. A conceptual label supplies no authority for coercion, restraint, or continuing after withdrawal.

Classification has limits

A conceptual review of motivating operations and discriminative stimuli questions the weight historically placed on CMO subtypes and emphasizes interaction among variables. Analysts should report observable operations and outcomes even when classification remains uncertain.

Useful notes say what changed, which consequence became more or less effective, what behavior changed, and what comparison supports the claim. “The CMO was present” is a conclusion, not an observation.

A practical analysis checklist

• What exact environmental change is the candidate operation?
• Which consequence may change in value?
• Which behavior has produced that consequence?
• What learning history is relevant?
• Which condition provides a credible comparison?
• Could availability, prompting, rules, pain, fatigue, or access explain the pattern?
• How are assent, communication, safety, and choice protected?
• What result would weaken the CMO hypothesis?

Keep hypotheses provisional and revise them when new evidence conflicts.

Use repeated within-person comparisons

Sample the suspected operation on several naturally occurring occasions and compare it with occasions when the operation is absent or differs in strength. Keep consequence availability and response effort visible. A single before-and-after episode can generate a hypothesis, while repeated comparisons show whether the value and behavior changes recur. Report exceptions because they often reveal a missing variable or an overly broad CMO label.