Preference Assessments

Preference Is Not Yet Reinforcement

A preference assessment identifies items, activities, people, sensory events, or routines a client approaches, selects, consumes, or engages with. It estimates what may function as reinforcement — but a stimulus is only a reinforcer if delivering it increases or maintains the behavior it follows. Confirming that requires a separate reinforcer assessment (e.g., a concurrent-operants or progressive-ratio test) showing the behavioral effect.

This preference-vs-reinforcer distinction is one of the most frequently tested ideas in this section. A 'highly preferred' item that does not raise responding is not a reinforcer for that behavior.

Preference data drive assessment-to-goal decisions. A learner with few known preferences is at risk of failed acquisition because consequences are weak; expanding the preference pool may itself be an early target. A client who strongly prefers peer interaction makes social goals and natural reinforcement easier to design.

The cleanest way to remember the logic: a preference assessment answers 'what does this person tend to choose?'; a reinforcer assessment answers 'does delivering that thing actually make the behavior happen more?' The first is descriptive (approach, selection, engagement); the second is functional (a behavioral effect on rate or persistence). High preference makes a stimulus a strong candidate reinforcer and is where you start — but the exam will punish any answer that treats 'most preferred' as 'proven reinforcer.'

The Formats and When to Use Them

Each format trades off precision (does it rank items or just find the top one?) against demands on the learner (can they scan an array, wait, tolerate removal?).

MSWO vs. MSW is a favorite exam contrast. MSWO removes the selected item, forcing the learner to choose among the remainder, so it builds a graded hierarchy efficiently. MSW replaces the item each trial, so a learner can pick the same favorite repeatedly — it reliably reveals the top item but a poor hierarchy. Free-operant removes nothing, so it is the least likely to evoke problem behavior triggered by withdrawing items, making it humane for learners who escalate when items are taken away.

Selection Logic, Motivating Operations, and Traps

Choosing a format

Choose by the learner's ability to scan, choose, wait, and tolerate removal, plus time, setting, staff training, and the cultural acceptability of the stimuli.

• Cannot make choices yet → single-stimulus.
• Need a precise ranking and have time → paired-stimulus.
• Need a ranking efficiently → MSWO.
• Removal evokes problem behavior → free-operant.
• Only need the single best item quickly → MSW.

Motivating operations change value

Preference is dynamic. Satiation (recent free access, just ate) lowers a stimulus's value; deprivation raises it. Food, attention, movement, devices, breaks, and sensory materials shift across the day, so a morning hierarchy may be wrong by afternoon. The exam punishes answers that treat a single assessment as permanent.

Common traps

• Assuming the most-selected item is automatically a reinforcer for any behavior — it must still pass a reinforcer test.
• Ignoring MOs when interpreting low selection of a normally preferred item.
• Picking a format the learner can't perform (e.g., paired-stimulus for a learner who cannot make a choice response).
• Letting position or color bias masquerade as preference — a learner who always picks the left or the red item is showing a side bias, not a true choice, so vary positions across trials and watch for it.
• Using only edibles; rotate in activities, attention, and sensory items so the hierarchy reflects real options and resists rapid satiation.

A preference result should lead to a practical next step: test reinforcement effects, arrange teaching conditions, expand preference variety, or choose goals that contact naturally occurring reinforcers.

Confirming reinforcement after preference

Once a hierarchy exists, confirm function with a reinforcer assessment. Two common methods:

• Concurrent-operants (choice) arrangement — make two responses available, each producing a different stimulus; the response that increases reveals the more effective reinforcer.
• Progressive-ratio (PR) assessment — steadily raise the response requirement and find the breakpoint (the last ratio completed). A higher breakpoint means the stimulus sustains more work, i.e., it is a stronger reinforcer.

This two-step sequence — preference assessment to nominate candidates, reinforcer assessment to confirm and rank by strength — is the defensible workflow. It explains why a 'highly preferred' item can still fail in a program: if it never passed a reinforcer test, its effect on behavior was never demonstrated, only its selection.