Fixed-ratio schedule

The ratio counts qualifying responses

An FR schedule is response based. If an outcome follows every third completed packing step, the arrangement is FR 3. Time may pass while the ratio is completed, but elapsed time does not determine eligibility.

Define what counts as one response. “Work” is too broad. A completed envelope that meets the agreed quality check is countable. A prompt, partial attempt, corrected error, or repeated movement needs an explicit rule.

Fixed ratio differs from fixed interval

An FR 5 schedule requires five eligible responses. An FI 5-minute schedule makes reinforcement available for the first eligible response after five minutes. A person can complete an FR ratio quickly or slowly; the count remains five.

Baum’s analysis of interval and ratio schedules reviews their basic definitions and common performance differences. The paper does not prescribe clinical use or a preferred ratio.

Reset rules change the arrangement

The ratio usually resets after reinforcement. If a session ends with two of four responses complete, the plan must say whether the next session starts at zero or carries the count forward. If errors reset the ratio, that is another contingency and potential source of burden.

Staff should record the count rather than rely on memory. A visible tracker can make the arrangement predictable when the person wants it, provided the display is accessible and does not expose private information.

Response patterns vary

Fixed-ratio schedules can produce steady responding followed by a pause after reinforcement. Larger ratios may be associated with longer pauses or breakdown under some conditions. These are empirical patterns, not part of the definition.

Task effort, consequence quality, delay, prompts, fatigue, health, instructions, and alternative activities may shape performance. Hackenberg’s token-reinforcement review illustrates schedule effects in complex token and exchange arrangements, mostly from laboratory research.

A fictional FR record

Keon chooses a brief audio clip after packaging each set of four envelopes. One response is an accurately sealed and labeled envelope. The count resets after the clip. Prompts are recorded, but prompted completions still count because the goal is a predictable work-break routine rather than independent mastery.

Keon completes 20 eligible responses and receives 5 scheduled outcomes. The arithmetic matches FR 4 because 20 divided by 4 equals 5. This record establishes delivery integrity for the session. Those counts do not show that FR 4 improved productivity, accuracy, or satisfaction.

Integrity and outcome are separate

Useful measures include:

• scheduled outcomes delivered divided by completed ratios
• eligible responses divided by all response attempts
• average and range of time to complete a ratio
• prompts, errors, pauses, and requests to change the arrangement
• task quality, preference, fatigue, and other unwanted effects

If five outcomes were due and four arrived, schedule integrity is 4/5. A high response rate cannot repair missed consequences or an inaccessible tracker.

Clinical use requires workload safeguards

Choose ratio size from assessment and the person’s informed preferences. Begin with achievable requirements, monitor burden, and adjust when the task changes. A ratio that fits one response may be unreasonable for another with greater effort or duration.

Do not place food, water, bathroom access, communication and AAC, mobility, prescribed care, rest, or emergency help behind a response count. Ordinary social connection and freely available leisure should not become payment for every action.

The BACB BCBA Test Content Outline, 6th edition includes reinforcement schedules and schedule thinning in examination content. Examination scope does not set a ratio, mastery rule, or treatment plan.

Plan thinning by function and fit

If programmed reinforcement is temporary, define how the ratio may change and what evidence supports that decision. Abruptly increasing requirements can create errors or abandonment. Natural outcomes, choice, and useful supports may replace parts of the program over time.

Keep independent performance, prompted teaching, maintenance, and generalization data separate. Report every ratio change with its effective date so trends are not combined across different schedules.

Version each ratio and workload

An FR 2 phase and an FR 4 phase should have separate labels, dates, and graphs. Record any simultaneous change in response quality, task effort, consequence, or prompt level. When the response becomes longer or harder, the same numerical ratio may represent a larger workload. Reassess fit from current effort and preference instead of treating the ratio number as fixed intensity.