15.1 Preference & Reinforcer Assessment Methodologies

Preference & Reinforcer Assessment Methodologies

Exam Tip: On the QASP-S exam, you must clearly distinguish between a Stimulus Preference Assessment (SPA) and a Reinforcer Assessment. An SPA identifies items, activities, or social interactions a learner prefers (yielding a preference hierarchy), but it only identifies potential reinforcers. A Reinforcer Assessment is an empirical operant test that delivers the preferred stimulus contingent upon a specific target behavior to determine if that behavior actually increases or maintains. You will be tested on the procedural steps, prerequisites, and clinical indications for Single-Stimulus, Paired-Stimulus, MSW, MSWO, and Free-Operant formats, as well as Progressive-Ratio schedules and the concept of the "breaking point."

In Applied Behavior Analysis (ABA), identifying effective reinforcers is the cornerstone of intervention design. Without functional reinforcers, teaching new skills, expanding communication, and reducing challenging behaviors become virtually impossible. However, clinicians cannot assume that items reported by caregivers or preferred by peers will serve as reinforcers for a specific autistic individual. Systematic, empirical assessment methodologies are required to identify what an individual likes, rank their relative value, and verify that delivering those stimuli increases behavioral repertoires.

---

Stimulus Preference Assessment vs. Reinforcer Assessment

The fundamental distinction between preference assessments and reinforcer assessments is rooted in operant conditioning principles:

$\text{Preference} \neq \text{Reinforcement}$

$\text{Preference Assessment} \longrightarrow \text{Identifies Highly Preferred Candidate Stimuli} (S^P)$ $\text{Reinforcer Assessment} \longrightarrow \text{Empirically Demonstrates Functional Control} (\uparrow R \text{ following } S^R)$

1. Stimulus Preference Assessment (SPA)

An SPA encompasses a variety of procedures designed to identify:

• Stimuli for which a learner demonstrates an affinity (approach, selection, or engagement).
• A preference hierarchy (e.g., high-preference [HP], moderate-preference [MP], and low-preference [LP] stimuli).
• Conditions under which preference values change across time, setting, and motivating operations.

Clinical Limitation: Even if a learner selects a spinning light-up toy on 100% of trials during an SPA, that toy is merely a candidate reinforcer. If delivering the light-up toy contingent upon vocal requesting does not produce an upward trend in requests, the toy is preferred but does not function as an effective reinforcer under those instructional demands.

2. Reinforcer Assessment

A reinforcer assessment is an active experimental evaluation wherein one or more candidate stimuli are delivered contingent upon the occurrence of a target response, and the resulting response rate is measured against a baseline or control condition (such as extinction or non-contingent reinforcement [NCR]).

• If response frequency, rate, or endurance increases above control levels when the stimulus is delivered contingently, the stimulus is confirmed as an operant reinforcer.
• If responding does not increase or maintain, the stimulus fails to function as a reinforcer, despite high ranking in an SPA.

The Impact of Motivating Operations (MOs)

Preferences are neither fixed nor permanent traits; they are dynamic states governed by Motivating Operations:

• Establishing Operations (EOs): Environmental events or biological states (such as deprivation) that increase the momentary reinforcing effectiveness of a stimulus (value-altering effect) and evoke behaviors that have produced that stimulus in the past (behavior-altering / evocative effect). For example, 3 hours of physical activity without liquids establishes cold water as a potent reinforcer.
• Abolishing Operations (AOs): Environmental events or biological states (such as satiation) that decrease the momentary reinforcing effectiveness of a stimulus (value-altering effect) and abate behaviors that have produced it (behavior-altering / abative effect). For example, consuming two full juice boxes immediately prior to an ABA session abolishes juice as an effective reinforcer for table tasks.
• Clinical Practice Standard: The QASP-S must train behavior technicians to conduct brief preference checks (e.g., a 3-item MSWO probe or free-operant sampling) at the start of every session and across activity transitions, rather than relying on a static preference assessment conducted weeks earlier.

---

Methods of Assessing Stimulus Preferences

Preference assessments fall along a continuum from indirect, informant-based approaches to direct, empirical observational protocols.

                               ┌───────────────────────────────┐
                               │ PREFERENCE ASSESSMENT METHODS │
                               └───────────────┬───────────────┘
                                               │
                       ┌───────────────────────┴───────────────────────┐
                       ▼                                               ▼
          ┌─────────────────────────┐                     ┌─────────────────────────┐
          │   INDIRECT METHODS      │                     │     DIRECT METHODS      │
          │  Caregiver Interviews   │                     │  Direct presentation &  │
          │  Surveys (RAISD)        │                     │  engagement measurement │
          └─────────────────────────┘                     └────────────┬────────────┘
                                                                       │
         ┌───────────────────────────────┬─────────────────────────────┼───────────────────────────────┐
         ▼                               ▼                             ▼                               ▼
┌─────────────────┐             ┌─────────────────┐           ┌─────────────────┐             ┌─────────────────┐
│ SINGLE-STIMULUS │             │ PAIRED-STIMULUS │           │    MULTIPLE     │             │  FREE-OPERANT   │
│ Pace et al.     │             │ Fisher et al.   │           │    STIMULUS     │             │ Roane et al.    │
│ (Approach)      │             │ (Forced-Choice) │           │ (MSW / MSWO)    │             │ (No Removal)    │
└─────────────────┘             └─────────────────┘           └─────────────────┘             └─────────────────┘

Indirect Assessment (Informant-Based)

Indirect methods gather retrospective subjective reports from parents, educators, caregivers, or the client:

• Informant Surveys: Checklists such as the Reinforcer Assessment for Individuals with Severe Disabilities (RAISD) or open-ended questionnaires.
• Clinical Strengths: Highly time-efficient, non-intrusive, and generates a broad initial pool of sensory, tangible, edible, and social stimuli to test.
• Clinical Weaknesses: Low predictive validity. Caregivers frequently identify items they wish the learner liked (e.g., educational books, vegetables) or overlook idiosyncratic sensory stimuli (e.g., vibrating toothbrushes, crinkly packaging tape). Direct empirical assessment must follow.

---

Systematic Direct Preference Assessment Formats

1. Single-Stimulus (SS) Preference Assessment / Approach Method (Pace et al., 1985)

• Protocol: The clinician presents one stimulus at a time to the learner in a randomized sequence across multiple trials. The clinician places the item directly in front of the learner (or within arm's reach) and allows a fixed latency (typically 5 seconds) for an approach response.
• Measurement:
  • An approach response is operationally defined as reaching, touching, picking up, or consuming the stimulus.
  • The percentage of trials approached is calculated: $\text{Approach Percentage} = \left( \frac{\text{Number of Trials Approached}}{\text{Total Trials Item Was Presented}} \right) \times 100$
  • Duration of engagement (how long the learner manipulates the item during a 30- to 60-second access interval) can also be recorded.
• Clinical Strengths: Ideal for learners who lack visual scanning repertoires, cannot make comparative choices between two or more items, or engage in impulsive dual-handed grabbing when presented with arrays.
• Clinical Limitations: Highly susceptible to false positives. Learners who reach for whatever is placed in front of them will show an approach rate near 100% for almost every item, failing to produce a differentiated preference hierarchy.

2. Paired-Stimulus (PS) / Forced-Choice Assessment (Fisher et al., 1992)

• Protocol: The clinician presents two stimuli simultaneously, separated by approximately 20–30 cm. The learner is instructed to "Pick one" or given an unprompted 5-second choice window.
  • If the learner reaches for one item, access is granted for 15–30 seconds while the unchosen item is immediately removed.
  • If the learner attempts to grab both items simultaneously, the clinician blocks the response and re-presents the trial with an explicit reminder to pick only one.
  • Every item in the assessment pool is paired systematically with every other item exactly twice, counterbalancing the spatial position (left vs. right) to prevent position bias.
• Trial Calculation Formula: For a pool of $N$ items, the total number of paired presentations is: $\text{Total Trials} = N \times (N - 1)$ Example: For 6 items, the clinician must run $6 \times (6 - 1) = 30$ trials.
• Measurement: Percentage of trials selected when available: $\text{Selection \%} = \left( \frac{\text{Number of Times Item Selected}}{\text{Number of Times Item Was Paired/Available}} \right) \times 100$
• Clinical Strengths: Gold standard for producing a sharply differentiated, highly accurate preference hierarchy (High, Moderate, and Low Preference).
• Clinical Limitations: Very time-consuming (often requiring 30–45 minutes). Requires visual scanning between two stimuli. Critically, because items must be taken away after 20–30 seconds of access, it frequently evokes extinction bursts, tantrums, or severe aggression in learners with tangibly maintained problem behaviors.

3. Multiple Stimulus with Replacement (MSW)

• Protocol: An array of 5 to 8 items is placed simultaneously in front of the learner. The learner selects one item and is allowed 15–30 seconds of access.
  • Following access, the selected item is returned back into the array.
  • The spatial positions of the items are rearranged/shuffled to control for side biases.
  • The clinician presents the identical array again for the next trial, repeating for a predetermined number of trials (e.g., 5–10 trials).
• Clinical Limitation: If a learner has a single, overwhelmingly dominant preference, they will select that identical item on every single trial. Consequently, the clinician gathers zero data regarding the relative preference ranking of the remaining items in the array.

4. Multiple Stimulus without Replacement (MSWO; DeLeon & Iwata, 1996)

• Protocol: An array of 5 to 8 items is placed in a horizontal line in front of the learner, spaced evenly.
  • The learner is directed to "Choose one."
  • Upon selecting an item, the learner receives 15–30 seconds of access.
  • While the learner consumes or manipulates the item, the clinician removes that item from the field entirely.
  • For the next trial, the remaining items are rotated and centered (e.g., moving the item from position 1 to position 4) so the array shrinks by one item on each successive trial ($N, N-1, N-2, \dots$).
  • The sequence continues until all items have been selected or until the learner makes no selection within 20 seconds.
  • The entire procedure is typically repeated 3 to 5 times (blocks) to calculate an aggregate selection rank across trials.
• Clinical Strengths: Extremely rapid (typically 5 to 10 minutes total). Yields a comprehensive, reliable preference hierarchy while preventing single-item monopolization.
• Prerequisites: Requires the learner to visually scan an array of 5–8 items, maintain choice consistency, and tolerate the permanent removal of selected items without engaging in severe problem behaviors.

5. Free-Operant (FO) Preference Assessment (Roane et al., 1998)

• Protocol: The learner is provided unrestricted access to a therapy room or play area containing multiple potential reinforcers. No items are ever removed, and no task demands or instructional prompts are presented.
  • Naturalistic Free-Operant: Observation conducted in the learner's everyday environment (e.g., preschool playroom or home) without altering the layout or introducing novel items.
  • Contrived Free-Operant: The clinician deliberately "salts" the environment with 6 to 10 specific target items (sensory toys, electronics, tactile manipulatives) placed strategically around the room. The learner is brought into the room, allowed to sample each item briefly for 10 seconds, and then observed for 5 to 15 minutes of uninterrupted exploration.
• Measurement: Cumulative duration of continuous physical engagement with each stimulus. Items are ranked by percentage of total session duration spent actively engaging.
• Clinical Strengths: Zero problem behavior evoked by item removal! Because staff never take an item away from the learner, this is the safest and most ethical direct assessment format for clients who engage in severe, dangerous aggression, property destruction, or self-injurious behavior (SIB) when items are removed.
• Clinical Limitations: Can yield an incomplete hierarchy if the learner hyper-fixates on a single preferred item for the entire 15-minute observation, revealing little information about other potential reinforcers.

---

Reinforcer Assessment Paradigms

Once an SPA establishes a hierarchy of candidate reinforcers, the clinician must empirically test their functional reinforcing capacity using one of three formal reinforcer assessment paradigms.

┌─────────────────────────────────────────────────────────────────────────────┐
│                        REINFORCER ASSESSMENT PARADIGMS                      │
├──────────────────────────┬──────────────────────────────────────────────────┤
│ Concurrent Schedule      │ 2+ responses available simultaneously; evaluates │
│ Reinforcer Assessment    │ relative reinforcing efficacy of stimuli.        │
├──────────────────────────┼──────────────────────────────────────────────────┤
│ Multiple Schedule        │ Single response across alternating signaled S^D  │
│ Reinforcer Assessment    │ conditions (e.g., HP vs LP vs Extinction).       │
├──────────────────────────┼──────────────────────────────────────────────────┤
│ Progressive-Ratio (PR)   │ Response requirements increase systematically to │
│ Schedule Assessment      │ determine the learner's "breaking point."        │
└──────────────────────────┴──────────────────────────────────────────────────┘

1. Concurrent Schedule Reinforcer Assessment

In a concurrent schedule, two or more operant behaviors are simultaneously available, each associated with an independent schedule of reinforcement delivering a different stimulus:

• Arrangement: Response A (e.g., sorting red blocks) delivers Stimulus X on an FR1 schedule. Response B (e.g., sorting blue blocks) delivers Stimulus Y on an FR1 schedule. Both tasks require equal physical effort.
• Measurement: Allocation of responding. If the client completes 45 trials of Response A and only 2 trials of Response B, Stimulus X demonstrates superior relative reinforcing efficacy.
• Clinical Utility: Essential for determining whether a moderate-preference tangible item can compete against automatic reinforcement or whether a social praise condition can maintain responding when paired with token systems.

2. Multiple Schedule Reinforcer Assessment

In a multiple schedule, a single target behavior is measured across alternating, distinct stimulus conditions, each signaled by an explicit discriminative stimulus ($S^D$):

• Arrangement: During Condition 1 (signaled by a green tablecloth), completing a math worksheet earns high-preference iPad access on an FR2 schedule. During Condition 2 (signaled by a yellow tablecloth), completing the same math worksheet earns low-preference verbal praise. During Condition 3 (red tablecloth), extinction is in effect.
• Measurement: Responding is compared across the alternating components. Higher response rates in the iPad condition confirm its operant reinforcing function.

3. Progressive-Ratio (PR) Schedule Reinforcer Assessment & The Breaking Point

While concurrent schedules evaluate relative preference, a Progressive-Ratio schedule measures the absolute potency and elasticity of a reinforcer under increasing effort or response demands.

• Operant Mechanism: The response requirement systematically increases following each reinforcer delivery according to a predetermined arithmetic step (e.g., FR1, FR3, FR5, FR7...) or geometric step (e.g., FR1, FR2, FR4, FR8, FR16, FR32...).
• The Breaking Point: The breaking point is operationally defined as the highest schedule requirement completed before responding ceases entirely for a specified duration (e.g., 5 consecutive minutes without a response):

$\text{Breaking Point} = \text{Last Successfully Completed Ratio Requirement prior to Responding Cessation}$

Clinical Example of Breaking Point Determination:

• A QASP-S evaluates two potential reinforcers (Bubbles vs. Tablet) for a learner learning difficult multi-step expressive language tasks:
  • Condition A (Bubbles): Ratio progression = FR1, FR3, FR6, FR10, FR15. The learner completes FR1, FR3, and FR6. At FR10, the learner responds twice and then stops responding for 5 minutes. Breaking Point = FR6.
  • Condition B (Tablet): Ratio progression = FR1, FR3, FR6, FR10, FR15, FR21, FR28, FR36. The learner completes all ratios up to FR28. At FR36, responding stops. Breaking Point = FR28.
• Clinical Decision: Tablet has a significantly higher breaking point than Bubbles. Tablet will maintain responding under high-effort academic demands and thin reinforcement schedules, whereas Bubbles will only support brief, low-effort instructional tasks.

---

Comparative Matrix of Stimulus Preference Assessment Formats

---

Preference Assessment Selection Algorithm