Every science rests on a few working assumptions about how the world operates and how we come to know it. Behavior analysis has five that show up again and again, and they quietly shape how you assess, explain, and change behavior.
Why it matters
These assumptions are not just history-of-science trivia. They explain why you collect direct observation data instead of trusting impressions, why you look for environmental variables instead of blaming a client's "personality," and why you keep an intervention only if the data show it works.
They feed Domain A, Behaviorism and Philosophical Foundations. Practice items here are usually short vignettes where a behavior analyst says or does something and you name the assumption behind it. Several assumptions sound alike, so the skill is discrimination: picking the one that fits best.
Key ideas
- Determinism: behavior is lawful. Behavior analysts assume the universe is orderly and that events, including behavior, happen as a result of other events. For behavior, those causes are the person's environment and history plus their genetic endowment. If behavior were random or uncaused, prediction and change would be impossible.
- Empiricism: knowledge comes from objective observation. Claims are accepted on the basis of careful observation and measurement, not opinion, intuition, tradition, or authority. This is why direct measurement sits at the center of the field.
- Parsimony: try the simplest account first. Before reaching for complex or abstract explanations, rule out the simplest, logical ones, ideally by testing them. Simple, observable environmental variables are considered before hypothetical inner causes.
- Pragmatism: judge ideas by how well they work. An explanation is valuable to the extent that acting on it helps you predict and influence behavior. "Does this account let us do something useful?" matters more than "Does it sound deep?"
- Selectionism: consequences select. Just as natural selection shapes species traits, consequences select behavior. Selection operates at three levels:
- Phylogenetic: traits and reflexes selected across many generations of a species.
- Ontogenetic: an individual's behavior selected by its consequences during their lifetime (operant learning).
- Cultural: practices shared by a group, passed on and selected by their effects on that group.
Determinism, empiricism, and parsimony also appear on the classic list of scientific attitudes, alongside experimentation, replication, and philosophic doubt (continually questioning what you believe to be true, including your own findings).
Definitions
- Determinism: the assumption that the world, including behavior, is lawful and orderly, so behavior is a function of identifiable variables.
- Empiricism: basing knowledge on objective observation and measurement of the events being studied.
- Parsimony: ruling out the simplest, logical explanations before considering more complex or abstract ones.
- Pragmatism: judging the worth of an explanation or practice by how well it works in predicting and changing behavior.
- Selectionism: the view that behavior, like species traits and cultural practices, is shaped by selection by consequences.
Examples from a session
The assumptions at work
- Determinism: A teacher says Mateo, age 9, "just melts down for no reason." His BCBA replies that the meltdowns are likely related to something in his environment or history, and starts looking for patterns.
- Empiricism: A school principal wants every classroom to use a new token board because a respected colleague says it "obviously works." The BCBA suggests measuring on-task behavior before and after the change in a few classrooms before deciding.
- Parsimony: Staff at an adult day program believe a client yells because he is "trying to control the room." The BCBA first checks a simpler account: yelling is reliably followed by staff attention.
- Pragmatism: Two explanations for a teen's school refusal are on the table. The team keeps the one that has let them predict refusal and reduce it, and drops the one that only sounds insightful.
- Selectionism (ontogenetic): Each time Aiko, a toddler, reaches toward a cup and says "wa," a parent hands her water. Saying "wa" becomes more frequent. Consequences selected the response.
- Non-example (not empiricism): A clinic adopts a sensory diet because the director has used it for years and "has seen it work." Experience and authority are the basis here, not objective measurement.
Common mistakes / exam traps
- Reading determinism as "behavior can't change." It is the opposite. Because behavior is lawful, changing the variables it depends on changes the behavior. Determinism is what makes intervention possible.
- Treating parsimony as "always accept the simplest answer." Parsimony means you rule out the simple, logical explanations first. If testing rules them out, you move on to more complex ones.
- Calling any experience "empirical." Empiricism refers to objective, systematic observation. A strong impression or a confident report is a starting point, not evidence.
- Mixing up the levels of selection. Reinforcement of one person's behavior is ontogenetic. Reflexes inherited by a species are phylogenetic. A practice spreading through a family, classroom, or organization is cultural.
Exam trap
Parsimony vs. pragmatism. Parsimony is about the order in which you consider explanations: the simple, testable one first. Pragmatism is about the standard you judge them by: does acting on it let you predict and change behavior? If the vignette turns on which account is checked first, it is parsimony; if it turns on which account is kept because it works, it is pragmatism.
Exam trap
Determinism vs. parsimony. Both reject "he just does it," but for different reasons. Insisting the behavior has identifiable causes rather than happening "for no reason" or by free choice → determinism. Setting aside an inner cause ("he's angry," "he wants control") to test a simpler environmental account first → parsimony. Quick map for the rest: collecting data → empiricism; consequences shaping behavior over time → selectionism.
Quick check
At a group home, one shift starts a quick end-of-shift huddle to share what went well. Incidents drop on that shift, other shifts adopt the huddle, and new staff learn it in orientation long after the original staff have left. Which level of selection best describes the huddle spreading and lasting?