Behavior Atlas

In 1967, Ogden Lindsley and his colleagues, including Carl Koenig, started the Behavior Bank, a shared store of Standard Celeration Chart projects sent in by teachers, parents, and researchers (Potts et al., 1993). Its 1971 handbook listed 11,947 projects, and by November 1973 it held 32,192 (Lindsley, 1990). Koenig (1972) used it to show that straight celeration lines describe charted behavior well. In Lindsley's words, the Bank failed because "the people who sent in the projects did not ask the bank questions" (Lindsley, 1990, p. 10). It stopped operating in the mid-1970s.

The data never stopped being collected. It sits on paper charts, in journals, and on hard drives, but it is rarely searchable or comparable. The Behavior Atlas gathers Standard Celeration Chart records so anyone who teaches, treats, or studies learning can ask: for this behavior, this intervention, and this dose, what rates of change have been seen, and how long did it take?

Every deposited chart adds to that reference. Lindsley started with thousands. We are starting again.

Between 2024 and 2026, the HHS Office of Inspector General audited Medicaid payments for Applied Behavior Analysis in Indiana, Wisconsin, Maine, and Colorado. In each state, every one of the 100 sampled enrollee-months contained at least one improper or potentially improper payment, and estimated improper payments ran into the tens of millions of dollars per state (Office of Inspector General, 2024, 2025, 2026a, 2026b). Common problems included session notes that did not support what was billed, billed time that included meals, naps, or recreation, and staff without required credentials. Much of the field treated this as a billing problem. It is also a measurement problem.

Applied Behavior Analysis is largely billed by the hour. Hours spent, however, do not tell you whether learning occurred, how fast it occurred, or whether the amount of time prescribed was producing the change the learner needed. A provider can bill forty hours a week for a year and have no data showing whether those hours were working for that individual learner.

Precision Teaching's foundational principle is that if the learner is not progressing, the program is wrong. Not the learner. The program. That principle requires measuring whether the learner is actually progressing, which requires measuring rate of change, which requires celeration. A celeration record does not just satisfy an auditor. It answers the clinical question: is this dose working for this person, and if not, what needs to change?

The Behavior Atlas exists to make that kind of decision possible from data rather than convention. What dose has produced what rate of change, for what behavior, under what conditions? The field has collected the data to answer that question for decades. It has just never put it in one place.