Preference Assessments in ABA: A Clinical Guide for Behavior Technicians

Overview & Clinical Significance

Preference assessments are a foundational component of ABA service delivery, yet they are among the most frequently underutilized procedures in clinical practice. The ability to identify what an individual currently prefers — what items, activities, and social interactions are most reinforcing in the current context — is the prerequisite for selecting the reinforcers that will drive skill acquisition and maintain motivation across sessions. Without accurate and current preference data, reinforcement delivery becomes guesswork, and the effectiveness of every ABA program is compromised.

The significance of this topic for behavior technicians lies in their role as the primary implementers of preference assessments and reinforcement programs. RBTs conduct more direct interactions with clients than any other member of the treatment team, and their sensitivity to what is reinforcing — and what is losing its reinforcing value — directly affects session productivity. A behavior technician who knows how to conduct multiple types of preference assessments, who understands why assessments are repeated rather than conducted once, and who can match the assessment type to the individual and the situation is a more effective clinical partner than one who uses the same approach with every client.

Preference assessments are also relevant to client dignity and self-determination. Identifying what individuals prefer — rather than deciding for them what should be reinforcing — is a practice that respects client autonomy, increases the probability that intervention is experienced as pleasant and motivating rather than aversive, and directly supports the commitment to person-centered care that underlies ethical ABA practice.

Background & Context

The systematic assessment of reinforcer preferences in ABA emerged from research demonstrating that reinforcement effectiveness varies among individuals and fluctuates over time within the same individual. The recognition that caregivers and practitioners often inaccurately predict what will function as a reinforcer — particularly for individuals with limited communication skills — motivated the development of standardized preference assessment methods.

Four primary preference assessment formats have been established in the ABA literature: free operant observation, single stimulus (successive choice), paired stimulus (forced choice), and multiple stimulus (either with or without replacement — MSWO or MSW). Each format provides different information about relative preference and has different efficiency, effort, and application characteristics. Research has established that items identified as higher preference through more systematic assessment methods — particularly paired stimulus and MSWO — are more likely to function as reinforcers in intervention contexts than items identified through less systematic methods such as caregiver report alone.

The distinction between a preferred item and a reinforcer is clinically important and often misunderstood. A preferred item is one the individual selects more often or approaches more readily than alternatives. A reinforcer is a stimulus whose delivery following a behavior increases the future probability of that behavior. Preference assessment identifies likely reinforcers — it does not confirm reinforcer function. That confirmation requires observation of the item's effect on behavior in the context of the skill program being run.

Satiation — the reduction in reinforcing value following repeated or extended exposure to a stimulus — is a critical concept for understanding why preference assessments must be repeated rather than conducted once. An item that was highly preferred during an assessment conducted three months ago may be neutral or even aversive now. Preference is a dynamic characteristic of the individual's current motivational state, not a fixed property of the stimulus.

Clinical Implications

Free operant preference assessments are conducted by making multiple items available simultaneously in the individual's environment and observing which items they approach, interact with, and spend the most time with during an unstructured period. Free operant assessments are low effort to set up and provide a naturalistic sample of preference. They are most appropriate as an initial screen or as a regular brief check at the start of sessions. Their limitation is that they do not provide precise ordering of preference strength among items.

Single stimulus assessments present items one at a time and record whether the individual approaches or interacts with each item. This format is useful for assessing preference for items that are difficult to present simultaneously, for individuals who are overwhelmed by multiple choices, and for generating a list of candidate items for more structured assessment. It does not provide information about relative preference among items with any precision.

Paired stimulus (forced choice) assessments present two items simultaneously and record which item the individual selects. Items are paired systematically so that each item is presented with every other item in the assessment set. The result is a rank ordering of items by percentage of trials on which each was selected. Paired stimulus assessments generate the most precise preference hierarchy among items but require the most effort to administer and score.

Multiple stimulus without replacement (MSWO) presents all items simultaneously and records the individual's sequential selection as items are removed from the array once chosen. MSWO is more efficient than paired stimulus for generating a preference hierarchy across multiple items and is the most commonly used format in clinical practice when a preference hierarchy across five or more items is needed. The first item selected is the most preferred; subsequent selections provide an ordered hierarchy.

Rotation of reinforcers — systematically varying the reinforcers used across trials or sessions — is a practical strategy for maintaining reinforcer effectiveness by preventing satiation. Behavior technicians should know how to implement reinforcer rotation based on preference hierarchy data and should be alert to signs that a reinforcer is losing its effectiveness, including reduced responding, longer latency to respond, or the client pushing items away after contact.

Ethical Considerations

The BACB Ethics Code (2022) Section 2.14 requires that behavior analysts use positive reinforcement as the foundation of behavioral programming. Preference assessment is the practical tool through which this commitment is implemented: without accurate preference data, the selection of positive reinforcers is unreliable. BCBAs who supervise RBTs in conducting preference assessments are responsible for ensuring that assessments are conducted correctly, that results are used to inform reinforcement selection, and that preference assessments are repeated with sufficient frequency to reflect the individual's current motivational state.

Client dignity in preference assessment requires that the process respects the individual's choices. When an individual consistently selects an item or activity, that preference should be honored in the reinforcement program even if the item or activity is one that caregivers find unusual or do not understand. The ethics of self-determination support using the individual's actual preferences rather than substituting items that others believe should be preferred.

Forcing an interaction with an item during a preference assessment — physically guiding the individual to touch or engage with a stimulus they are not approaching — contaminates the assessment and is ethically problematic. Preference assessment must allow voluntary responding; the value of the assessment depends on the individual's choices being genuine expressions of current preference rather than compliance with adult prompting.

Privacy considerations apply to the information gathered in preference assessments and to the data collected about reinforcer effectiveness. Preference data are part of the clinical record and should be handled with the same confidentiality protections as other clinical information. Sharing preference information with unauthorized parties — for example, discussing a client's unusual preferences with family members who are not involved in the treatment team — is a confidentiality concern.

Assessment & Decision-Making

Selecting the appropriate preference assessment format requires consideration of the individual's communication and response repertoire, the number of items being assessed, the time and resources available, and the purpose of the assessment. For a brief check at the start of a session to identify what the client is most interested in today, a free operant assessment is sufficient. For building a reinforcer menu to support a new skill acquisition program, a paired stimulus or MSWO assessment that generates a preference hierarchy provides more useful data.

The frequency of preference assessment should be specified in the behavior plan or in clinical procedures and should be sufficient to capture changes in preference over time. Quarterly formal assessments at minimum, with brief informal checks before each session, is a common and reasonable standard. Clients with rapidly changing preferences or whose motivational state is affected by physiological variables — sleep, illness, hunger — may require more frequent assessment.

Decision-making about reinforcer effectiveness should be based on data, not intuition. If a client's response rate is declining despite consistent reinforcement delivery, the first hypothesis to investigate is whether the reinforcer has lost its effectiveness — either due to satiation, changing preference, or inadequate delivery. Conducting a preference reassessment and rotating to higher-ranked alternatives is the behavioral response to declining reinforcer effectiveness.

Data collection during preference assessments should be systematic. For paired stimulus and MSWO assessments, data sheets that record trial-by-trial selections allow for accurate calculation of preference percentages and rankings. Informal assessment data — notes about what the client reached for first, what they ignored, what they pushed away — provides supplementary information that complements formal assessment data and captures information about aversion that formal assessments may not fully capture.

What This Means for Your Practice

For RBTs, the practical takeaway is that preference assessment is not a one-time task but a recurring clinical responsibility. Starting each session with a brief informal preference check — asking yourself what the client is showing interest in today, making a few items available and observing — is a habit that improves reinforcement selection immediately and costs very little time.

Knowing the types of preference assessments — their names, their procedures, and when each is most appropriate — means being able to implement them accurately and to communicate with the supervising BCBA about assessment findings and reinforcer effectiveness clearly. BCBAs depend on behavior technicians to be accurate observers of preference in the moment, because the technician's real-time observations often provide more current preference data than a formal assessment conducted during a supervision session.

For supervising BCBAs, building preference assessment into the regular session structure — with specific instructions in the session protocol about when and how to conduct preference checks, how to record findings, and how to rotate reinforcers — ensures that reinforcement selection remains data-driven across the full range of clinical contact hours rather than only during supervised sessions.

The connection between preference assessment, reinforcer effectiveness, and client motivation is direct and immediate. Clients who encounter sessions as reinforcing — because the activities and consequences offered are things they actually prefer — engage more readily, learn more quickly, and experience the therapy environment positively. This is not an abstract commitment to 'positive practice'; it is a data-based clinical reality with direct implications for how effective ABA services are.