14.3 Indirect & Descriptive Functional Behavior Assessment (FBA)

Indirect & Descriptive Functional Behavior Assessment (FBA)

Exam Tip: On the QASP-S exam, FBA questions focus on distinguishing between the three levels of the assessment continuum: Indirect, Descriptive, and Experimental. You must know the specific names and features of indirect screening tools (MAS, QABF, FAST) and understand why indirect methods have low reliability. For descriptive ABC recording, you must understand the mathematical calculation of conditional versus unconditional probabilities and recognize the major clinical pitfall: descriptive data show correlation, NOT causation, frequently producing false-positive results for social attention due to ubiquitous caregiver reprimands.

Under federal disability mandates (IDEA) and ethical clinical standards established by the QABA, designing a Behavior Intervention Plan (BIP) without first identifying the maintaining operant function of the problem behavior is considered professional malpractice. Prescribing generic deceleration techniques (e.g., response cost, time-out, or differential reinforcement) without knowing why a behavior occurs frequently exacerbates the behavior, escalates crisis events, and violates client dignity. The Functional Behavior Assessment (FBA) represents the systematic process of gathering empirical information to determine the environmental variables that occasion and maintain challenging behaviors.

---

The FBA Assessment Continuum

Functional Behavior Assessment is not a single instrument; rather, it is a hierarchical, multi-method continuum comprising three distinct tiers:

┌─────────────────────────────────────────────────────────────────────────────┐
│                     THE FBA METHODOLOGICAL CONTINUUM                        │
├─────────────────┬──────────────────────────────────┬────────────────────────┤
│ Tier / Level    │ Assessment Methodology           │ Experimental Rigor     │
├─────────────────┼──────────────────────────────────┼────────────────────────┤
│ 1. Indirect     │ Informant interviews, checklists,│ Low: Subjective recall,│
│    Methods      │ rating scales (MAS, QABF, FAST). │ no direct observation. │
├─────────────────┼──────────────────────────────────┼────────────────────────┤
│ 2. Descriptive  │ Direct observation in natural    │ Moderate: Objective    │
│    Methods      │ settings (ABC data, Scatterplots)│ correlation, no causal.│
├─────────────────┼──────────────────────────────────┼────────────────────────┤
│ 3. Experimental │ Systematic analog manipulation   │ High: Gold standard,   │
│    Methods (FA) │ of antecedents & consequences.   │ confirms causation.    │
└─────────────────┴──────────────────────────────────┴────────────────────────┘

While an Experimental Functional Analysis (FA) provides the only definitive demonstration of cause-and-effect relations, clinical practice routinely begins with Indirect and Descriptive methods to screen behaviors, establish operational definitions, identify idiosyncratic setting events, and generate testable hypotheses.

---

Indirect Functional Behavior Assessment

Indirect FBA methods gather retrospective information from informants who know the individual well (e.g., parents, teachers, paraprofessionals, group home staff, or the client themselves) without directly observing or measuring the behavior as it occurs.

Standard Indirect Assessment Tools:

1. Structured Behavioral Interviews:
   • Functional Assessment Interview (FAI; O'Neill et al.): A comprehensive clinical interview gathering information regarding behavioral topographies, physiological setting events (sleep deficits, allergies, pain), antecedent triggers, communicative repertoires, and potential reinforcers.
   • Student-Assisted Functional Assessment Interview: Engages verbal autistic learners in self-reporting triggers, academic preferences, and perceived peer dynamics.
2. Standardized Rating Scales & Questionnaires:
   • Motivation Assessment Scale (MAS; Durand & Crimmins, 1988): A 16-item questionnaire scored on a 7-point Likert scale evaluating four functional categories: Sensory (Automatic), Escape (Social Negative), Attention (Social Positive), and Tangible (Social Positive).
   • Questions About Behavioral Function (QABF; Paclawskyj et al., 2000): A 25-item rating scale evaluating five distinct behavioral functions: Attention, Escape, Non-social (Automatic), Physical Discomfort, and Tangible.
   • Functional Analysis Screening Tool (FAST; Iwata & DeLeon, 1996): A 16-item binary (Yes/No) questionnaire identifying social positive reinforcement, social negative reinforcement, automatic positive reinforcement, and automatic negative reinforcement.

Clinical Strengths of Indirect Methods:

• Rapid and convenient to administer (typically 15–30 minutes).
• Poses zero physical risk to the client or staff (no behavior is evoked).
• Identifies rare or low-frequency setting events that might not occur during a direct 60-minute observation (e.g., monthly menstrual cramps, parental custody transitions, changes in seizure medication).
• Helps build clinical rapport with family members and educational stakeholders.

Major Psychometric Limitations:

• Low Inter-Rater Reliability: Extensive research demonstrates poor agreement between different informants rating the same client (e.g., parent vs. teacher).
• Informant Biases: Subject to recency bias (over-reporting behaviors from yesterday), primacy bias, emotional burnout, and halo effects.
• Lack of Functional Validity: Informants consistently misattribute internal cognitive motives ("he does it out of spite" or "she has ADHD") rather than identifying environmental contingencies.
• Rule of Thumb: Indirect tools must never be used in isolation to develop a Behavior Intervention Plan.

---

Direct Descriptive Functional Behavior Assessment

Descriptive FBA involves direct, real-time observation and measurement of the target behavior and environmental events in the natural environment (e.g., classroom, lunchroom, therapy clinic, or home) without experimental manipulation.

1. Narrative ABC Data Collection

In narrative ABC recording, the observer writes an open-ended descriptive diary of everything that occurs immediately before the behavior (Antecedent), the precise topography of the response (Behavior), and what immediately followed the response (Consequence).

• Example: Antecedent: "Teacher hands client math worksheet." -> Behavior: "Client rips paper and screams." -> Consequence: "Teacher says 'We don't rip papers,' removes worksheet, and sends client to cool-down corner."
• Limitations: Highly subjective, inconsistent between observers, time-intensive, and difficult to quantify mathematically.

2. Structured / Continuous ABC Data Collection

In structured ABC recording, the observer uses a pre-printed data sheet containing standardized, objective checkboxes for common antecedents (e.g., demand presented, attention diverted, transition, peer interaction), target behaviors, and consequences (e.g., verbal reprimand, physical guidance, task removed, item returned). Observers record data continuously across observation intervals, allowing for objective statistical coding.

3. Scatterplot Data Collection

As introduced in Section 14.1, scatterplots divide the client's day into standardized time blocks across successive calendar days to identify temporal, environmental, and schedule-correlated patterns.

---

Empirical Calculation of Conditional Probabilities

A sophisticated clinical competency tested on the QASP-S exam is the calculation of conditional and unconditional probabilities from structured ABC data (Vollmer et al., 2001).

Core Mathematical Formulas:

1. Unconditional Probability of Problem Behavior: The overall likelihood that problem behavior occurs during any observed interval: $P(B) = \frac{\text{Total Number of Intervals with Target Behavior}}{\text{Total Number of Observation Intervals}}$

2. Conditional Probability of a Consequence Given Behavior: The likelihood that a specific environmental consequence ($C$) follows the emission of the target behavior ($B$): $P(C|B) = \frac{\text{Number of Behavioral Episodes Followed by Consequence } C}{\text{Total Number of Behavioral Episodes}}$

3. Conditional Probability of Behavior Given an Antecedent: The likelihood that the target behavior ($B$) occurs when a specific antecedent event ($A$) is presented: $P(B|A) = \frac{\text{Number of Antecedent Presentations Eliciting Behavior } B}{\text{Total Number of Antecedent Presentations}}$

Step-by-Step Clinical Calculation Vignette:

A QASP-S conducts a 2-hour structured ABC observation consisting of 60 consecutive 2-minute intervals. The target behavior is aggressive hitting.

• Total intervals observed = 60.
• Total intervals containing aggressive hitting ($B$) = 15.
• Consequence Analysis across the 15 hitting episodes:
  • In 12 episodes, staff immediately delivered verbal reprimands / attention ($C_{\text{attention}}$).
  • In 3 episodes, staff removed the academic task ($C_{\text{escape}}$).
• Antecedent Analysis:
  • Staff presented academic demands ($A_{\text{demand}}$) exactly 20 times during the session.
  • Aggressive hitting occurred immediately following 14 of those 20 demand presentations.

Mathematical Computations:

• Unconditional Probability of Hitting: $P(B) = \frac{15}{60} = 0.25 \text{ (or } 25\%\text{)}$
• Conditional Probability of Attention Given Hitting: $P(C_{\text{attention}}|B) = \frac{12}{15} = 0.80 \text{ (or } 80\%\text{)}$
• Conditional Probability of Escape Given Hitting: $P(C_{\text{escape}}|B) = \frac{3}{15} = 0.20 \text{ (or } 20\%\text{)}$
• Conditional Probability of Hitting Given Demand Antecedent: $P(B|A_{\text{demand}}) = \frac{14}{20} = 0.70 \text{ (or } 70\%\text{)}$

Interpretation: Although staff delivered attention following 80% of hitting episodes, hitting occurred in 70% of demand presentations. This sets up the classic descriptive assessment dilemma.

---

The Critical Flaw: The Attention False-Positive Trap

While descriptive FBA provides valuable observational context, it carries a profound, well-documented methodological vulnerability: Correlation does NOT equal causation.

The Attention False-Positive Phenomenon:

In natural educational and residential environments, caregivers, teachers, and peers naturally react to loud, disruptive, or aggressive behavior by looking, gasping, talking, comforting, or verbally reprimanding the individual. Because reprimands occur with high frequency in human social environments, descriptive ABC data almost always demonstrate an exceptionally high conditional probability for social attention ($P(C_{\text{attention}}|B) \approx 0.75 - 0.95$).

However, in translational and clinical research (e.g., St. Peter et al., 2005; Thompson & Iwata, 2007), when these same individuals are exposed to a systematic experimental Functional Analysis, less than 15% to 20% of those behaviors are actually maintained by social attention! In reality, the true maintaining reinforcer is frequently negative reinforcement (escape from demands) or automatic reinforcement, and the caregiver's verbal reprimands are simply an irrelevant, co-occurring epiphenomenon.

If a clinician mistakenly relies on descriptive data showing 80% attention consequences and prescribes an attention extinction protocol (e.g., planned ignoring during table tasks), the client will continue to experience the task demand, the behavior will escalate dramatically into severe crisis, and the intervention will fail catastrophically.

---

Formulating Testable, Function-Based Hypotheses

The ultimate clinical deliverable of an Indirect and Descriptive FBA is a synthesized, testable hypothesis statement. A comprehensive hypothesis statement must include four mandatory operant elements:

┌─────────────────────────────────────────────────────────────────────────────┐
│                     FOUR-PART FUNCTIONAL HYPOTHESIS MODEL                   │
├─────────────────┬───────────────────────────────────────────────────────────┤
│ 1. Setting      │ Motivating Operations and physiological/environmental     │
│    Events (MOs) │ context (e.g., poor sleep, noisy classroom, hunger).      │
├─────────────────┼───────────────────────────────────────────────────────────┤
│ 2. Antecedent   │ Specific discriminative stimulus ($S^D$) triggering the   │
│    Triggers     │ behavior (e.g., presentation of multi-step math tasks).   │
├─────────────────┼───────────────────────────────────────────────────────────┤
│ 3. Target       │ Objective, measurable response topography emitted         │
│    Behavior     │ (e.g., screaming, ripping paper, biting forearm).         │
├─────────────────┼───────────────────────────────────────────────────────────┤
│ 4. Maintaining  │ Operant reinforcer maintaining the behavior ($S^R+$ or    │
│    Function     │ $S^R-$: e.g., negative reinforcement via task escape).    │
└─────────────────┴───────────────────────────────────────────────────────────┘

Exemplar Clinical Hypothesis Statement:

"When presented with non-preferred multi-step writing tasks ($S^D$) following nights with fewer than 6 hours of sleep (Setting Event / MO), Marcus engages in severe table-banging and verbal profanity (Target Behaviors) in order to obtain temporary cessation of academic demands (Maintaining Function: Social Negative Reinforcement / Escape)."

---

Comparative Matrix: Indirect vs. Descriptive FBA Methods