1.1 Goals of Behavior Analysis & Foundational Assumptions

The Scientific Framework of Behavior Analysis

Behavior analysis is a natural science that investigates the interactions between the behavior of organisms and their environment. As formulated by B.F. Skinner and expanded across decades of basic and applied research, the discipline approaches human and non-human action not as arbitrary or internally willed occurrences, but as lawful physical events. To master the foundational content of the Behavior Analyst Certification Board (BACB) examination, an assistant behavior analyst (BCaBA) must possess an exacting understanding of both the overarching goals of scientific inquiry and the foundational philosophical assumptions that govern behavioral research and clinical practice.

Science is not merely a collection of verified facts; it is a systematic approach to seeking and organizing knowledge about the natural world. In behavior analysis, this search is structured around three distinct, cumulative levels of scientific understanding: description, prediction, and control.

---

Levels of Scientific Understanding: Description, Prediction, and Control

Scientific inquiry progresses through three hierarchical tiers of understanding. Each tier provides the necessary empirical and methodological foundation for the subsequent level. A behavior analyst must accurately identify which level of understanding has been attained during behavioral assessment and intervention.

1. Description

Description is the foundational level of scientific understanding. It involves systematic, objective observation and quantification of the phenomenon of interest. Through descriptive procedures, the scientist measures, categorizes, and classifies physical events, establishing operational definitions that allow independent observers to record the occurrence and nonoccurrence of behavior reliably.

In applied settings, description provides a clear, quantitative snapshot of baseline rates, topography, duration, latency, interresponse times, and immediate environmental correlates. However, description alone cannot explain why a behavior occurs, nor can it identify causal mechanisms. For example, documenting that a student engages in motor stereotypy an average of 42 times per hour during classroom instruction describes the topography and temporal parameters of the response class, but it does not establish why the behavior persists.

Key descriptive assessment tools in clinical practice include:

• Narrative Antecedent-Behavior-Consequence (ABC) data collection
• Structured ABC interval recording checklists
• Scatterplot matrices mapping behavior across times of day and routines

2. Prediction (Correlation and Covariation)

Prediction occurs when repeated systematic observations demonstrate that two or more events consistently covary. When the presence or change in one event reliably predicts a change in the relative probability of another event, a correlation or covariational relationship is established.

Prediction enables behavior analysts to anticipate the relative probability that a target response will occur under specific environmental conditions. For instance, if scatterplot data demonstrate that aggressive outbursts occur almost exclusively between 10:00 AM and 10:30 AM during small-group reading instruction, the clinician can predict an elevated probability of aggression during that instructional period. Similarly, when direct observation confirms that peer teasing is systematically followed by verbal outbursts in 85% of recorded instances, a strong predictive correlation exists between the antecedent trigger and the student's reaction.

Critical Exam Trap: Prediction does not demonstrate causation or functional control. Two events may consistently covary due to an unmeasured third variable or confounding environmental condition. Correlational observations derived from scatterplots, descriptive ABC logs, and parent interviews provide predictive hypotheses, but they never prove that an antecedent or consequence is the maintaining functional variable.

3. Control (Functional Relations)

Control represents the highest, most definitive level of scientific understanding. Scientific control is achieved when a functional relation (causation) is demonstrated between an independent variable (IV) and a dependent variable (DV).

A functional relation is demonstrated when an experimenter systematically manipulates the independent variable (such as an intervention procedure, schedule of reinforcement, or antecedent condition) while holding all potential confounding and extraneous variables constant, and reliably observes a specific change in the dependent variable (the target behavior) that is improbable to have resulted from extraneous factors.

In behavior analysis, experimental control requires believable replication across time, conditions, or participants using single-case experimental designs (such as ABAB reversal, multiple baseline, or multielement designs). In functional behavior assessment, control is established exclusively through an analog functional analysis (FA)—the systematic manipulation of hypothesized maintaining variables across test conditions (e.g., contingent attention, contingent escape, contingent tangible access, and alone) compared against a play control condition.

---

Comparative Analysis of the Three Goals of Science

The following matrix outlines the operational distinctions among description, prediction, and control as applied to behavior analysis:

---

Foundational Philosophical Assumptions of Behavior Science

All natural sciences rest on fundamental philosophical assumptions regarding the nature of reality, knowledge, and physical inquiry. Task A.2 of the BCaBA Test Content Outline (6th ed.) lists five assumptions as examples: selectionism, determinism, empiricism, parsimony, and pragmatism. Behavior-analytic textbooks add three related attitudes of science: experimentation, replication, and philosophic doubt. This section covers all eight.

                                 [ DETERMINISM ]
                   (The universe is lawful, orderly, and cause-and-effect driven)
                                        |
                   +--------------------+--------------------+
                   |                                         |
           [ SELECTIONISM ]                               [ EMPIRICISM ]
   (Phylogenic, Ontogenic, Cultural)             (Objective measurement & observation)
                   |                                         |
                   |                              [ EXPERIMENTATION ]
                   |                      (Systematic IV manipulation on DV)
                   |                                         |
                   |                               [ REPLICATION ]
                   |                        (Direct & systematic repetition)
                   |                                         |
                   +--------------------+--------------------+
                                        |
                   +--------------------+--------------------+
                   |                                         |
            [ PARSIMONY ]                           [ PHILOSOPHIC DOUBT ]
(Simplest explanation ruled out first)      (Continuous skepticism & empirical revision)

1. Selectionism

Selectionism is the philosophical premise that life, biology, and behavioral repertoires evolve and are shaped through a process of variation, interaction with the environment, and differential selection by consequences. B.F. Skinner recognized that the principle of selection by consequences operates across three distinct levels:

1. Phylogenic Selection (Phylogeny): The evolutionary history of a species. Natural selection operates across generations, selecting anatomical structures and unconditioned, innate behavioral repertoires (e.g., unconditioned reflexes, unconditioned reinforcers such as food, water, warmth, and biological defense reactions) that promote species survival.
2. Ontogenic Selection (Ontogeny): The development and shaping of an individual organism's behavioral repertoire over its lifetime. Operant conditioning is the primary vehicle of ontogenic selection: behaviors that contact reinforcement are selected and strengthened, while behaviors that encounter extinction or punishment are selected against and weaken.
3. Cultural Selection: The evolution and transmission of behavioral practices, traditions, and verbal rules across members of a social group or society. Cultural practices are maintained and passed between generations through imitation, observational learning, modeling, and shared verbal contingencies that benefit the survival of the culture.

2. Determinism

Determinism is the foundational presumption that the universe is a lawful and orderly place in which all physical phenomena occur as a result of other events. Behavior does not occur by accident, whim, spontaneous creation, or uncaused "free will." Every behavioral event is functionally related to preceding and concurrent physical events, encompassing an organism's biological constitution, genetic endowment, evolutionary history, prior learning history, and immediate environmental stimulation.

Determinism vs. Fatalism: A Vital Distinction

A critical conceptual distinction often tested on the BCaBA exam is that between determinism and fatalism:

• Fatalism is the belief that the future is predetermined regardless of what actions are taken—a resigned stance that human intervention is futile because "what will be will be."
• Determinism, by stark contrast, affirms that behavior is determined by specific, identifiable environmental and biological variables. Because behavior is functionally determined by the physical environment, actively modifying environmental antecedents and consequences directly changes behavior. Determinism is the very premise that empowers behavior analysts to design life-altering interventions.

3. Empiricism

Empiricism is the practice of objective observation and systematic measurement of the phenomena of interest. Scientific empiricism demands that behavioral investigations rely on thorough, objective quantification rather than personal beliefs, subjective impressions, speculation, philosophical dogma, or anecdotal testimony. If an event cannot be observed and measured directly or indirectly through physical instrumentation, it cannot serve as the empirical foundation of behavioral science.

In clinical practice, empiricism requires assistant behavior analysts to collect reliable interobserver agreement (IOA) data, utilize precisely defined operational definitions, and graph physical data points rather than relying on caregiver impressions such as "he had a much better day today."

4. Experimentation

Experimentation is the controlled comparison of a phenomenon of interest (the dependent variable) under the systematic manipulation of two or more independent variables, while extraneous and confounding variables are actively held constant. Experimentation transforms descriptive observation into functional analysis. It is the engine through which behavior analysis identifies causal functional relations rather than mere coincidental correlations.

5. Replication

Replication is the repeating of experiments (and experimental conditions within an experiment) to determine the reliability and generality of findings. In behavior analysis, replication is conducted at multiple levels:

• Direct Replication: Repeating the exact experimental protocol with the same participant or across identical conditions. Demonstrates the reliability of the behavior change.
• Systematic Replication: Intentionally varying one or more non-critical parameters (such as the participant population, setting, therapist, or task materials) while keeping the core independent variable constant. Demonstrates the external validity and generality of the behavioral technology.

Replication protects science against flukes, experimental error, researcher bias, and coincidental trends.

6. Parsimony

Parsimony (often referred to as Occam's razor) requires that all simple, logical explanations for a phenomenon must be ruled out experimentally or conceptually before more complex, abstract, or unverified explanations are adopted. A parsimonious account relies strictly on established principles and fewest possible assumptions.

Parsimony in Clinical and Diagnostic Evaluations

When confronted with an abrupt escalation of severe problem behavior in a non-vocal client, a parsimonious approach demands that the clinician first investigate basic biological, medical, and environmental variables before postulating complex neurochemical anomalies, personality disorders, or new psychiatric syndromes. For example, ruling out acute otitis media (ear infection), dental abscesses, gastroesophageal reflux, constipation, or abrupt sleep deprivation represents sound behavioral parsimony. Similarly, evaluating whether an intervention failure is due to a simple schedule strain or inconsistent reinforcement fidelity is far more parsimonious than assuming the client has "internal resistance to therapy."

7. Philosophic Doubt

Philosophic doubt is the continuous, skeptical attitude with which scientists view all scientific facts, interpretations, and established knowledge. Practitioners must constantly question the validity of their assessments, the effectiveness of their interventions, and the veracity of published literature. When new empirical data contradict established theories, the behavior analyst must be willing to abandon prior assumptions in favor of the data. Philosophic doubt keeps behavior analysis vibrant, self-correcting, and protected against dogmatic clinical complacency.

8. Pragmatism

Pragmatism is the position that the value (or "truth") of a scientific account is judged by its practical consequences: an explanation is good to the extent that it lets us predict and change behavior effectively. Skinner's radical behaviorism adopts this pragmatic truth criterion, sometimes called "successful working." An account is not accepted because it sounds plausible or matches a popular theory; it is accepted because acting on it produces the predicted change in behavior.

In practice, pragmatism shapes how an assistant behavior analyst chooses between explanations. Suppose one team member says a student "refuses math because he is lazy," and another says "difficult math worksheets have been followed by removal of the work, so escape now reinforces refusal." Only the second account points to variables the team can change (task difficulty, the escape contingency, a break request), and its usefulness can be tested with data. A pragmatic analyst therefore prefers the second account and keeps it only if the intervention it suggests actually works.

Exam trap: pragmatism is not "anything goes as long as it works." The criterion is effective prediction and control demonstrated by data, and it always operates inside the Ethics Code. A procedure that suppresses behavior through harm is not justified by pragmatism.

---

Philosophical Assumptions Matrix

The following table summarizes the eight philosophical assumptions and attitudes of behavior analysis, their theoretical essence, applied exemplars, and common anti-patterns or exam traps:

---

Clinical Translation: Applying Assumptions to Real-World Scenarios

To pass the BCaBA exam, candidates must move seamlessly from textbook definitions to complex, scenario-based decisions. Consider how these foundational assumptions guide clinical reasoning:

Distinguishing Scientific Goals Across Clinical Assessment

When an assistant behavior analyst initiates services for a client engaging in self-injurious behavior, the clinical workflow mirrors the three levels of science:

1. Descriptive Baseline: The clinician gathers frequency and duration data, records narrative ABC charts, and maps out a scatterplot. Outcome achieved: Description.
2. Pattern Identification: The clinician correlates the scatterplot data with the daily schedule, discovering that SIB occurs with high probability during independent seatwork and near-zero probability during 1-on-1 reading. Outcome achieved: Prediction.
3. Functional Analysis Verification: The clinician systematically alternates between 5-minute conditions of task presentation (escape) and free access to preferred toys (control) in an alternating treatments design, demonstrating that SIB occurs exclusively when academic tasks are presented and ceases immediately upon task removal. Outcome achieved: Control (Functional Relation).

Applying Parsimony to Differential Diagnosis

A 5-year-old child with autism abruptly begins biting their own wrist and screaming during meal times. The child's parents request a referral to an occupational therapist for "sensory integration therapy," suspecting an internal sensory processing crisis. An assistant behavior analyst adhering to parsimony advises the supervising BCBA and the family that physical and dental causes must be investigated first. A pediatric dental examination reveals a severe, impacted molar causing excruciating pain exacerbated by chewing. Once the tooth is treated, the wrist-biting ceases entirely. By holding to parsimony, the team identified the simplest, physical cause without subjecting the child to weeks of speculative sensory exercises.