Precision teaching

1. Introduction

Version followed: Ogden Lindsley's Precision Teaching (PT) as set out in the Athabasca University Lindsley resource (five steps: pinpoint, set an aim, count and teach, develop a learning picture, decide), cross-checked against the five-step process of Evans, Bulla & Kieta (2021: pinpoint, practice, chart, decide, try again, as summarised in Vostanis et al., 2023). Charting and decision conventions follow Kubina & Yurich (2012) as reported in Kubina's practitioner articles. SAFMEDS follows the basic procedure of Eshleman (2000) and Graf & Auman (2005) as reported by Quigley et al. (2021). Inner-behaviour work follows Calkin (1981) and Kostewicz, Kubina & Cooper (2000).

What it is. PT is not a psychotherapy. It is a measurement and decision system that can sit on top of any teaching or treatment method. The learner or clinician pinpoints a countable behaviour, times it, records it as a frequency (count per minute or per day) on the Standard Celeration Chart (SCC), and uses the slope of change (celeration) to decide each week whether to keep or change the procedure [source: Athabasca concept page; Wikipedia; Vostanis et al. 2023]. Lindsley developed it from Skinner's free-operant rate research. The chart dates from the 1950s–60s and was standardised in 1967, and the name was coined in 1966 [source: SCS history page]. Its main settings are classrooms and tutoring. Psychotherapy-relevant uses come from a small self-management literature on counting and changing "inner behaviours" such as thoughts, feelings and urges [source: Kubina CentralReach article; Kostewicz et al. 2000].

How change happens (core idea). Behaviour is best described by how often it occurs and how fast that frequency multiplies or divides over calendar time. Frequent, timed practice with immediate feedback builds fluency, meaning accurate and fast performance. Fluent performance is claimed to be retained, to endure, to hold up under distraction and to transfer to new tasks. When progress stalls, the procedure is at fault, not the learner, and the chart shows when to change it [source: Athabasca concept and set-aim pages; Kubina 2005].

For a psychotherapist [synthesis]: use PT as (a) a daily self-monitoring and decision layer around whatever treatment you deliver, (b) a fluency-building method for skills that must be produced quickly under stress, such as coping statements, assertive replies or psychoeducation terms (SAFMEDS), and (c) the specific inner-behaviour procedures in 4.14–4.16. The originators did not develop it as a psychotherapy. Applications beyond those published procedures are marked [synthesis].

2. Framework

PT has no fixed number of sessions. It runs as a continuous loop with brief daily measurement and weekly decisions for as long as a target is being worked on [source: Athabasca process pages; Evans et al. 2021 via Vostanis et al. 2023].

Timescale. The daily SCC spans 140 calendar days (about one school semester). Celeration is read per week, so decisions usually run on a weekly cycle [source: Wikipedia; Binder 2004 charting slides]. Morningside Academy requires learners to keep doubling their performance frequency each week (×2/week) and changes the procedure if the chart shows otherwise [source: Athabasca decide page, citing Johnson & Layng 1992]. Kubina gives ×1.1/week as too slow, needing about 27 weeks to reach a typical aim from a low start [source: Kubina, standard celeration metrics].

Criteria for moving between phases [source: Athabasca decide page; Kubina learning-pictures article; Evans aims article]:

• Continue when corrects are above errors and accelerating, or when the learning picture is in the "improving" family.
• New target ("hard-to-do has become easy-to-do") when the aim is met with zero errors on two consecutive counting periods (McGreevy, 1983). Some authors also require fluency outcomes (maintenance, endurance, stability, application) before moving on.
• Change procedure ("hard-to-do has become hard-to-learn") when growth is too slow or absent, when the projected celeration line misses the aim star, or immediately when the picture is worsening.
• Mark every change with a phase change line on the chart, so the next celeration line belongs to the new condition [source: Aninao et al. 2015].

3. Guiding principles

[source: Athabasca concept page (Lindsley's four guidelines), SCS history, Evans et al. 2021 critical features via Vostanis et al. 2023; practical meanings are synthesis]

• Focus on directly observable behaviour. Count actions or their products, not labels such as "anxious" or "lazy". The site's "Dead Man's Rule" is "If a dead man can do it, then it ain't behavior" (Athabasca). Practical: "not panicking" and "staying calm" fail the test; "says a coping statement" or "notices an urge and clicks the counter" pass.
• Make private behaviour countable. Silent or inner events are made public, for example by reading aloud or self-counting thoughts. Practical: the client is the only observer of inners, so train them to count.
• Frequency (count per time) is the basic measure, not percent correct. Two people can both be 100% accurate and differ hugely in fluency. Practical: always time; always record the counting period.
• Use the standard chart. Plot on a multiply/divide (logarithmic) scale against calendar days, so equal proportional changes look alike and learning appears as a straight line. Practical: plot every day, including weekends and days with no data, and keep the same chart so trends are comparable across clients.
• The learner knows best. If the learner is not progressing, the programme is wrong and must change (from Skinner's remark that the rat knows best, per SCS history). Practical: never blame the client for a flat chart. Change one thing and look again.
• Count both corrects and errors (the correct/incorrect pair). Define what to do as well as what not to do, and give every decelerating target an accelerating replacement. Practical: pair "self-critical thoughts" with "self-supportive statements".
• Daily, continuous measurement. Count and chart every day [source: SCS history]. Practical: short daily timings beat long weekly ones.
• Learner self-charting. Learners who count and chart their own behaviour do at least as well, and gain self-management skills and trust in the data [source: Athabasca; SCS history]. Practical: hand the chart to the client as early as possible.
• Aim for fluency, not mere accuracy. Aims are frequency ranges seen in competent performers [source: Athabasca set-aim page; Evans aims article]. Practical: don't stop at "can do it"; stop at "can do it fast and smoothly".
• Behaviours are independent. Gains in one learning channel or pinpoint do not guarantee gains in another, and changed inners do not guarantee changed public behaviour (Lindsley 1992, as cited by Kubina). Practical: measure the outcome you care about directly.
• Timely decisions. Act on the chart promptly, ideally weekly or sooner for worsening trends [source: Evans et al. 2021 critical feature via Vostanis et al. 2023].

4. Intervention inventory

PT's "interventions" are measurement, charting and decision procedures plus a few practice formats. The teaching or treatment content itself (prompting, reinforcement, exposure, cognitive work and so on) comes from whatever method you pair with PT [source: Athabasca count page; Wikipedia]. Entries 4.1–4.13 are core PT, 4.14–4.16 are the published inner-behaviour procedures, and 4.17 is a procedure named in sources but not drawn on.

4.1 Pinpointing

• Purpose / when: At the start of every target and at every change. It turns a vague complaint or goal into something countable. [source: Athabasca pinpoint page; Evans et al. 2021 via Vostanis et al. 2023]
• Steps:
  1. Ask what the person would be doing if the problem were better or worse. Write it as an action verb plus object.
  2. Apply the dead-man test: if a dead person could "do" it ("not shout", "be calm"), rewrite it as an action ("says 'I need a minute' and leaves the room").
  3. Check that the behaviour is countable: it is either observable or a product the person can tally.
  4. Set the counting period and keep it identical every day. Make it long enough for roughly 8–10 occurrences. Use a special daily practice period for infrequent behaviours, and a whole day (up to 24 hours) for behaviours that occur through the day.
  5. Define the correct/incorrect pair: what to do and what not to do. Give every behaviour you want to reduce a replacement to increase. If only one can be counted, count the correct one (McGreevy 1983).
  6. Name the learning channel set (4.2).
  7. Check it is "hard-to-do": there must be room to improve on both corrects and errors. Near-zero errors with low corrects suggests the person is avoiding errors by going slowly.
• Example language (illustrative): "If I filmed you on a good day, what would I see you doing that I don't see now?" / "'Not worrying' is something a statue can do. What would you be doing instead?" / "Let's pick something you could click a counter for."
• Watch for: pinpoints that are traits or outcomes rather than movements; counting periods that change from day to day; decelerating targets with no replacement behaviour ("dead-man" goals). It is working when you and the client would agree on the count for any given minute.

4.2 Learning channel specification

• Purpose / when: At pinpointing, and whenever skills fail to transfer. The channel specifies how the stimulus comes in and how the response goes out. Fluency in one channel set does not guarantee fluency in another. [source: Athabasca pinpoint page; Lin & Kubina 2004 (title only) [uncertain]]
• Steps:
  1. Choose an input: see, hear, touch, smell, taste, feel, or an internal event ("think").
  2. Choose an output: say, write, point, draw, do or perform.
  3. Write the pinpoint as "input/output", for example "see/say" or "think/say".
  4. Plan the progression from heavily supported to independent input. The source's example is guided → heard → self-initiated ("think") shooting.
  5. Measure the channel the real-world task demands before declaring mastery.
• Example language (illustrative): "Reading the coping card and saying it is see/say. In the real moment there's no card, so we'll build up to think/say."
• Watch for: assuming a skill fluent on one channel (written) transfers to another (spoken under stress). Shifts in channel are a legitimate step in the programme and should be marked on the chart.

4.3 Setting frequency aims

• Purpose / when: Before charting. The aim defines "done" in fluency terms. [source: Athabasca set-aim page; Evans aims article; Kubina 2005]
• Steps:
  1. Look for a published performance standard for the skill and channel, such as the appendix of Kubina & Yurich (2012) or aims from a PT curriculum publisher.
  2. If none exists, time 2–3 competent performers (adults or skilled peers) for 1–2 minutes on the identical task and channel. Use their range; do not use class averages.
  3. Write the aim as a range, high number first (for example "100–80 per minute"), with an error aim of 0 (or 0–1).
  4. Cap the aim by the learner's tool skills: written aims cannot exceed the learner's free-writing rate, and spoken aims cannot exceed their speaking rate. Make sure the aim counts the same unit as the timing (words, not letters).
  5. Put the aim on the chart: an aim star at the deadline date, or an aim band across the chart if there is no deadline.
  6. Treat the aim as provisional. If the aim is reached without the expected fluency outcomes (4.13), reconsider it. Do not lower it just because it "feels high".
• Example language (illustrative): "I'll time myself on the same cards so we know what fluent looks like. You're aiming for my range, not perfection."
• Watch for: aims set on accuracy alone, and aims that exceed the tool skill. Kubina (2005) gives basic maths facts at roughly 80–120 digits per minute as an example of a published standard. For clinical skills there are generally no validated aims [uncertain]; derive them from competent performers [synthesis].

4.4 Timings (timed counting)

• Purpose / when: Daily. Timings yield the frequency datum. [source: Athabasca pinpoint and count pages; SCS history; Kubina 2005; Binder 2004]
• Steps:
  1. Prepare the materials and a countdown timer. Timings of 1 minute are standard; shorter (10–30 seconds) or longer (up to 24-hour whole-day counts) are fine if kept constant within a phase.
  2. Give a short standard instruction, start the timer, and say nothing during the timing.
  3. Count corrects and errors (skips can be counted with errors, as in Binder's charts). The teacher counts observable responses; the learner self-counts inners; products are scored afterwards.
  4. Divide the count by the minutes to get count per minute, and record the counting time as well.
  5. Give brief feedback, then plot (4.5).
• Example language (illustrative): "Ready? Go as fast as you comfortably can. Skip anything you're stuck on. Please begin." ... "Stop. Let's count."
• Watch for: changing timing length mid-phase, since a 2-minute and a 4-minute timing are effectively different tasks; frequency falling over the course of the timing (an endurance problem, see 4.10); timing-related anxiety (see section 5).

4.5 Daily Standard Celeration Charting

• Purpose / when: Every day there is a count. The chart makes learning visible and comparable. [source: Wikipedia; SCS history; Binder 2004; Aninao et al. 2015; Kostewicz et al. 2000]
• Steps:
  1. Use the daily SCC: calendar days (140) across the bottom, and a multiply/divide scale from about 1 per day to 1000 per minute up the side. Paper charts or charting software both work.
  2. Plot corrects (accelerating targets) as dots and errors (decelerating targets) as x's on the actual calendar day. Days with no timing get no mark, but the day still passes on the chart.
  3. Mark the counting-time bar or record floor (1 divided by the counting time). A zero count is plotted just below the floor.
  4. Draw a vertical phase change line whenever anything changes (materials, timing, procedure, aim), and label it.
  5. Draw the aim star or aim band (4.3).
• Example language (illustrative): "Today's dot goes on Wednesday. Each line up multiplies by ten, so moving from 5 to 10 looks the same as moving from 50 to 100."
• Watch for: reverting to percentage graphs (they hide fluency differences; always show raw counts [source: Binder 2004]); skipped calendar days making trends look steeper than they are; missing phase lines.

4.6 Learner self-counting and self-charting

• Purpose / when: As early as feasible, and always for inner behaviours. [source: Athabasca concept page; SCS history (Curtis & Lovitt 1969; Slezak)]
• Steps:
  1. Model one timing and plot it together.
  2. Have the learner count and plot while you check.
  3. Hand over fully, and review the chart together at each session.
  4. Invite the learner to read their own learning picture and propose changes, with your agreement before they are adopted (Kubina learning-pictures article).
• Example language (illustrative): "It's your chart. Tell me what your week looks like and what you'd change."
• Watch for: plotting errors (check a sample) and self-criticism triggered by the chart (frame flat lines as "the plan needs changing"). It is working when the client brings the chart unprompted and suggests changes.

4.7 Celeration lines, projection and bounce

• Purpose / when: Once a phase has several data points. Celeration quantifies the speed of learning; bounce quantifies day-to-day variability. [source: Kubina standard celeration metrics; Binder 2004; Evans aims article]
• Steps:
  1. Within one phase, draw a best-fit straight line through the corrects and another through the errors. The SCS uses a quarter-intersect or least-squares type line; the exact method was not specified in the sources [uncertain].
  2. Read the line's value per week: ×1.4 means growth of 40% per week, ×2 means doubling, ÷2 means halving.
  3. Extend the line forward. If it reaches the aim star by its date, stay on course; if it does not, plan a change.
  4. Draw parallel lines above and below the data to form an envelope. The ratio between them is the bounce (for example ×2 or ×5), and high bounce means unstable performance.
• Example language (illustrative): "You're multiplying by about 1.3 a week. At this rate we reach the aim in about eight weeks; to hit the date we need a steeper line."
• Watch for: lines drawn across phase changes (don't); over-reading two or three points; high bounce, which suggests inconsistent conditions, so check timing, setting and materials [synthesis].

4.8 Reading learning pictures

• Purpose / when: At each review. Classify the joint pattern of corrects and errors to choose a decision quickly. [source: Kubina learning-pictures article, citing All 1977 and Kubina & Yurich 2012]
• Steps:
  1. Look at both lines together.
  2. Classify the picture:
     • Improving: cross-over, jaws, take-off, climb (corrects growing, errors falling or flat), uphill (both rising) and dive (corrects flat, errors falling).
     • Maintaining: aim (both at aim), get truckin' and rock bottom (little change).
     • Worsening: snowplow (unlearning), landing (corrects falling, errors flat), surface (corrects flat, errors rising) and downhill (both falling).
  3. Act on the family:
     • Improving: continue. For uphill, monitor, because the learner may be practising an incorrect form. For dive, check that corrects begin to rise.
     • Aim: move to maintenance.
     • Get truckin' and rock bottom: do something different.
     • Worsening: remediate immediately.
• Example language (illustrative): "This is a 'jaws' shape — the right answers are rising and the errors are closing in from below. We keep going."
• Watch for: reading one line in isolation.

4.9 Decision rules (continue / new target / change)

• Purpose / when: At weekly review, or sooner if the picture is worsening. [source: Athabasca decide page; Evans aims article; Kubina standard celeration metrics]
• Steps:
  1. Continue if corrects are above errors and accelerating (White & Haring 1980), or the picture is in the improving family, and the projected line meets the aim on time.
  2. New target when the aim is met with zero errors on two consecutive counting periods (McGreevy 1983). Before moving on, ideally confirm fluency outcomes (4.13).
  3. Change when celeration is below the programme's standard (Morningside uses ×2/week; ×1.1 is "too slow" per Kubina), when the projection misses the aim star, or when the picture is flat or worsening.
  4. Change one variable at a time, draw a phase line, and re-read after the new phase has enough data [synthesis on "one at a time"; phase line per sources].
• Example language (illustrative): "Two flat weeks means the plan isn't working, not you. Let's pick one thing to change."
• Watch for: waiting several flat weeks (decisions should be timely) and changing several things at once, which hides what worked.

4.10 Instructional change options (slice back, step back, change timing, change aim, change method)

• Purpose / when: After a "change" decision. Match the change to the learning picture. [source: Athabasca decide page]
• Steps:
  • Slice back: practise a smaller part of the full movement (for example, the first three of four steps). Use it when errors exceed corrects and both are accelerating, or corrects are slightly higher but flat or falling while errors climb.
  • Step back: practise an easier prerequisite. The most basic step back is a "tool" movement repeated rapidly. Use it when errors exceed corrects and are rising or flat while corrects are flat or falling.
  • Change the counting period: if performance fades within the timing (endurance), build fluency at shorter durations and then lengthen them. In one source example, moving from 1-minute to 15-second timings cut errors and noncompliance. Eventually evaluate at the duration real life requires (Binder et al. 1990).
  • Change the aim: set a temporary lower aim to restore success, then raise it again.
  • Change the teaching method: alter what happens before the response (materials, prompts, modelling), after a correct response (reinforcement, feedback), after an error (error correction), or the practice arrangement (4.11).
• Example language (illustrative): "The full script is too much right now. This week we'll just practise the opening line until it's automatic."
• Watch for: stepping back so far that the task becomes "easy-to-do" with no room to grow. Check that the new phase actually changes the celeration.

4.11 Frequency building (rate-building practice)

• Purpose / when: To move an accurate but slow skill up to its fluency aim. [source: Kubina 2005; Binder 2004; Quigley et al. 2021; Kostewicz et al. 2000]
• Steps:
  1. Confirm the skill is accurate at a slow pace. If it is not, teach it first; frequency building is practice, not initial instruction.
  2. Choose the practice arrangement. Kubina (2005) lists the dimensions: daily or intermittent; single or multiple trials per session; with or without feedback; massed or spread through the session or day; with or without cumulative review; extrinsic or intrinsic reinforcement.
  3. Use short repeated timings ("sprints"), for example several 10–60-second practice timings with brief rests, followed by one charted test timing.
  4. Give immediate feedback after each sprint and practise the missed items.
  5. Spread practice: Binder (2004) reports faster celeration with more, shorter drills per day at equal weekly time.
• Example language (illustrative): "Three quick 20-second rounds to warm up, then we'll time one for the chart. Beat your own score, not anyone else's."
• Watch for: accuracy collapsing as speed rises (slow down or slice back); rehearsing errors; fatigue. It is working when the celeration line rises while errors fall or stay at zero.

4.12 SAFMEDS (Say All Fast a Minute Each Day Shuffled)

• Purpose / when: Fluency on discrete verbal facts and terms. Lindsley coined the term in the late 1970s (Quigley et al. 2021). Clinical uses (psychoeducation terms, cue → coping response pairs) are [synthesis]. [source: Quigley et al. 2021, describing Eshleman 2000 and Graf & Auman 2005; Wikipedia]
• Steps (basic procedure):
  1. Make a deck of cards about 7.6 × 12.7 cm with a prompt on the front and a short answer on the back. Replace bent or marked cards, since learners can recognise them.
  2. Shuffle the deck.
  3. Start a 1-minute timer.
  4. See the front, say the answer aloud (see/say), flip the card to check, and sort it into a correct or an incorrect pile. Go as fast as possible.
  5. When time is up, count both piles and chart them on the SCC.
  6. Do this every day. One timing per day is the minimal version; multiple daily timings are optional.
  7. Adjust deck size and timing length to performance (Eshleman; Graf & Auman).
• Variants (Quigley et al. 2021):
  • Review: up to 10 minutes of studying the whole deck before the timing.
  • Sprints: three 1-minute practice timings before the test timing.
  • Incremental whole-deck (Kim et al. 2001): 15-card subdecks combined into a 60-card deck.
  • Incremental sprints: subdecks plus practice timings.
  • Participants also informally pulled missed cards for extra practice.
  • All supplemented variants outperformed the basic procedure, and the incremental variants looked strongest. The authors call this ranking tentative.
• Example language (illustrative): "Shuffle, say each answer out loud, flip, sort. Don't stop to think for long — if it's not there, guess or skip and keep going."
• Watch for: silent "thinking" answers, since the procedure requires an observable spoken response; decks too large for novel material, where the basic procedure alone gave low corrects and high errors; an unshuffled deck, which builds serial-order memory instead of the skill.

4.13 Fluency outcome checks (REAPS / RESAA / MESAG)

• Purpose / when: At or after aim. These checks verify that fluency produced the outcomes it is meant to: retention, endurance, stability, application and adduction or generativity. The acronyms vary: REAPS (Binder 1996; Haughton 1980), RESAA (Johnson & Layng 1996), MESAG (Evans aims article) and RESAG (Vostanis et al. 2023). [source: Kubina 2005; Evans aims article; Vostanis et al. 2023]
• Steps (the sources name the outcomes but give no probe protocol, so these operational steps are [synthesis]):
  1. Retention: retime after a break with no practice (for example 1–2 weeks) and compare with the last frequency.
  2. Endurance: time for longer than the practice duration and look for a fall-off within the timing (Kubina defines endurance as the trend of performance as the interval lengthens).
  3. Stability: time under distraction or in a less comfortable setting.
  4. Application: check that fluent component skills combine into the composite skill without being taught directly. Kubina, Young & Kilwein (2004) used two fluent spelling components that yielded accurate spelling of untaught words.
  5. Adduction / generativity: watch for new, untaught combinations.
• Example language (illustrative): "You haven't practised these for two weeks — let's see if the speed stayed."
• Watch for: a big drop on any check, which means you should raise or modify the aim (Evans aims article) or add maintenance timings.

4.14 Counting inner behaviours (whole-day self-count)

• Purpose / when: To make private events such as thoughts, feelings and urges measurable for self-management. The prerequisites are "a pinpoint, a timing device, and counter" (Kubina CentralReach article). [source: Kubina CentralReach article on Calkin 1981, McCrudden 1990, Cooper 1991; Kostewicz et al. 2000; Duncan 1971 cited in SCS history]
• Steps:
  1. The client defines each inner in their own words and writes the definition down (Calkin and McCrudden used the learner's own criteria).
  2. Count thoughts and feelings as separate pinpoints. Kostewicz et al. counted them together and later listed this as a limitation.
  3. Use a wrist or golf-style tally counter: one click per occurrence, over a fixed period (Kostewicz counted midnight to midnight), then reset.
  4. Chart the daily totals as count per day on the SCC. Plot zero days below the floor.
  5. Take baseline before any intervention. Baselines in the published cases ran for weeks.
• Example language (illustrative): "Each time you notice the 'I'm useless' thought, one click — no arguing with it, just count. Tonight you write the number down and plot it."
• Watch for: counting becoming a cue for rumination [synthesis]; forgetting to count, where the published fixes are a wristwatch timer, reminders near personal items and written prompts (Kostewicz et al.). The accuracy of private counts cannot be checked by an observer (Kostewicz et al.).

4.15 One-minute counting procedure (accelerating a replacement inner)

• Purpose / when: To accelerate positive or replacement inners in daily timed bursts while monitoring the target inner. [source: Kubina CentralReach article (Calkin 1981; Cooper 1991; Kubina, Haertel & Cooper 1994); Kostewicz et al. 2000]
• Steps:
  1. Pinpoint the replacement inner. Examples from the cases: positive statements about oneself (Calkin), loving thoughts about one's spouse and family (Cooper), non-aggressive statements and imagery (Kostewicz).
  2. Set a 1-minute countdown timer.
  3. During the minute, produce and tally as many distinct replacement thoughts as possible. Calkin wrote them; Kostewicz tallied them on paper. Count a duplicate only once per timing; it may recur in later timings.
  4. Chart the per-minute count as dots, and continue whole-day counts of the target inner as x's on the same chart.
  5. Repeat daily, reviewing celeration on both.
• Example language (illustrative): "For one minute, write down as many true, kind things about yourself as you can. Different ones each time count; repeats don't."
• Watch for: the one-minute timing alone may not reduce the target. In Kostewicz et al. the target inner did not change during the 1-minute phase and fell only after the distributed procedure (4.16). Clients may find generating positive inners invalidating; adapt the content to self-compassionate or neutral statements rather than forced positivity [synthesis].

4.16 Distributed brief timings

• Purpose / when: When a single daily timing fails to shift the target inner. [source: Kostewicz et al. 2000]
• Steps:
  1. Schedule six 10-second timings spread across the waking day (the published times ran roughly every 2–3 hours from mid-morning to late evening).
  2. Use the same replacement-inner tally as 4.15 at each prompt.
  3. Chart the six counts stacked on the day, and keep the whole-day count of the target.
  4. Phase it in and out to check its effect (the study used a reversal design).
• Example language (illustrative): "Your phone will buzz six times a day. Ten seconds each time: as many calm, helpful thoughts as you can."
• Watch for: missed prompts, so use alarms. The authors suggest the effect may come from the frequent cue to relax and reorient, and it is a single-case self-experiment [uncertain].

4.17 Talk-aloud problem solving (named, not drawn on)

• Purpose / when: Listed among strategies combined with PT (Dembek & Kubina 2018, cited in Vostanis et al. 2023). Its procedure was not reviewed for this guide [uncertain], so no steps are given. Consult the original source before using it.

5. Handling problems

PT sources handle problems almost entirely through the chart: "the learner knows best", so a flat or worsening chart means the programme changes. Responses below cite sources where they exist; everything else is [synthesis].

• Stuck client (flat chart: "get truckin'" or "rock bottom"). Do something different (Kubina learning-pictures article). Choose a slice back, step back, timing change, temporary aim or method change according to the picture (Athabasca decide page). Check the pinpoint is "hard-to-do", not "hard-to-learn" or already "easy-to-do". [source]
• Low corrects with near-zero errors (error avoidance). The learner may be going slowly to avoid mistakes. Encourage speed and treat errors as learning opportunities (Athabasca pinpoint page). [source] With anxious or perfectionistic clients, say this explicitly and praise attempts as well as answers [synthesis].
• Rising errors ("uphill" or "surface"). The learner may be practising an incorrect form. Stop rate building, reteach accurately, and slice back (Kubina; Athabasca). [source]
• Performance falls within a timing (endurance). Shorten the timing and build back up (Athabasca decide page; Evans aims article). [source]
• Skills fluent in session but absent in real life. Check the learning channel (see/say vs think/say) and the real-world duration. Behaviours are independent, so measure the real-life behaviour directly (Lindsley 1992 via Kubina; Binder et al. 1990 via Athabasca). [source]
• Inners change but behaviour doesn't. Expected, given that behaviours are independent (Lindsley 1992 via Kubina). Add a public-behaviour pinpoint and chart it alongside. [source + synthesis]
• One-minute procedure doesn't shift the target inner. Move to distributed brief timings (Kostewicz et al. 2000). [source]
• Non-adherence to daily counting or timing. The published remedies are a wristwatch countdown timer, reminders near personal items and written prompts (Kostewicz et al.). [source] Further options [synthesis]: shrink the task to one 10–30-second timing per day; anchor it to an existing routine; let the client plot on a phone; review the chart at the very start of each session so the data clearly matter; and chart adherence itself as a pinpoint.
• Timing anxiety or shame about the numbers [synthesis]. Frame timings as measuring the plan, not the person. Let the client compete only with their own prior frequency. Use shorter timings, and allow the client to keep the chart private and choose what to share. If timing anxiety persists, make it a pinpoint or drop timings for that target and use whole-day counts instead.
• Self-monitoring increases distress or rumination [synthesis]. Counting a distressing inner can cue more attention to it. Pair it with a replacement-inner acceleration target (4.15); limit counting to tallying with no elaboration; or count only the replacement behaviour, which is permitted when only one of the pair can be worked with (McGreevy 1983 via Athabasca).
• Resistance to pinpointing ("my problem isn't countable") [synthesis]. Validate first. Then ask for the smallest observable sign that would show the problem was better, and offer a provisional pinpoint to test for one week, revisable at the next decision point.
• Ruptures over a "failing" chart [synthesis]. Repair by explicitly owning the plan ("the learner knows best" puts responsibility on the programme). Ask the client what felt wrong and make their suggestion the next phase change, marked on the chart.
• Aim feels impossibly high. Don't lower it reflexively. Set a temporary aim and raise it later (Athabasca decide page; Evans aims article). [source] For clinical skills with no published aims, sample competent performers, including yourself (4.3) [synthesis].
• Data unreliable for private events. Accuracy of self-counted inners cannot be checked (Kostewicz et al.). [source] Look for convergent public indicators, such as behaviour products or reports from others with consent [synthesis].
• Counted inners involve self-harm or suicidal thoughts or urges [synthesis]. PT measurement is not a risk-assessment tool. Assess risk with your service's standard structured clinical risk assessment and safety planning, including lethal-means counselling (safe storage or removal) where indicated. Treat any acceleration in such counts, or any new intent or plan, as a prompt for immediate clinical review, not a phase change. Do not rely on chart trends to judge safety.
• Modality-specific pitfalls [synthesis]: drawing celeration lines across phase lines; changing several variables at once; using percent-correct graphs; skipping calendar days; dropping a target before checking retention and endurance (4.13); using decks or practice sheets that are unshuffled or worn, which cue serial or physical-feature responding (Quigley et al. 2021) [source].

Sources

Sources

• Wikipedia: Precision teaching
• Standard Celeration Society: History of Precision Teaching
• Athabasca University: Precision Teaching — Concept Definition and Guiding Principles
• Athabasca University: Select a Task (Pinpoint)
• Athabasca University: Set an Aim
• Athabasca University: Count and Teach
• Athabasca University: Develop a Learning Picture
• Athabasca University: Decide What to Do
• Quigley et al. (2021) An Evaluation of Multiple SAFMEDS Procedures (PMC)
• Kubina: Can thoughts and feelings be counted, measured and changed? (CentralReach)
• Kostewicz, Kubina & Cooper (2000) PDF
• Kubina (2005) The relations among fluency, rate building, and practice (PDF)
• Kubina: Jaws, snowplows and dives — using learning pictures for decision making
• Kubina: The power of standard celeration metrics
• Evans: Without aims you are aimless
• Binder (2004) Standard Celeration Charting (ISPI slides, PDF)
• Aninao, Acevedo, Newsome & Newsome (2015) Putting the precision in precision teaching (PDF)
• Vostanis et al. (2023) Campbell protocol: Evaluating the application and effectiveness of precision teaching (PDF)
• Lindsley (1992) Precision teaching: Discoveries and effects — PMC record (metadata only; full text not retrieved)

Further reading (not drawn on unless listed above)

• Lindsley, O. R. (1990). Precision teaching: By teachers for children. Teaching Exceptional Children, 22(3), 10–15.
• Lindsley, O. R. (1991). Precision teaching's unique legacy from B. F. Skinner. Journal of Behavioral Education, 1, 253–266.
• Lindsley, O. R. (1992). Precision teaching: Discoveries and effects. JABA, 25, 51–57.
• White, O. R., & Haring, N. G. (1980). Exceptional Teaching (2nd ed.). Merrill.
• White, O. R. (1986). Precision teaching — precision learning. Exceptional Children, 52, 522–534.
• McGreevy, P. (1983). Teaching and Learning in Plain English.
• Binder, C. (1996). Behavioral fluency: Evolution of a new paradigm. The Behavior Analyst, 19, 163–197.
• Johnson, K. R., & Layng, T. V. J. (1992). Breaking the structuralist barrier. American Psychologist, 47, 1475–1490.
• Kubina, R. M., & Yurich, K. K. L. (2012). The Precision Teaching Book. Greatness Achieved.
• Pennypacker, H. S., Gutierrez, A., & Lindsley, O. R. (2003). Handbook of the Standard Celeration Chart.
• Calkin, A. B. (1981). One minute timing improves inners. Journal of Precision Teaching, 2(3), 9–21.
• Calkin, A. B. (2005). Precision teaching: The standard celeration charts. The Behavior Analyst Today, 6(4).
• Cooper, J. O. (1991). Can this marriage be saved? Self-management of destructive inners. Journal of Precision Teaching, 8(2), 44–46.
• Kubina, R. M., Haertel, M. W., & Cooper, J. O. (1994). Reducing negative inner behavior of senior citizens: The one-minute counting procedure. Journal of Precision Teaching, 11(2), 28–35.
• Kostewicz, D. E., Kubina, R. M., & Cooper, J. O. (2000). Managing aggressive thoughts and feelings with daily counts of non-aggressive thoughts and feelings. Journal of Behavior Therapy and Experimental Psychiatry, 31, 177–187.
• Evans, A. L., Bulla, A. J., & Kieta, A. R. (2021). The precision teaching system: A synthesized definition, concept analysis, and process. Behavior Analysis in Practice, 14, 559–576 (cited via Vostanis et al. 2023; not drawn on [uncertain]).
• Eshleman, J. W. (2000); Graf, S. A., & Auman, J. (2005) — SAFMEDS basic procedure, cited via Quigley et al. 2021 (not drawn on).
• Quigley, S. P., et al. (2021). An evaluation of multiple SAFMEDS procedures. Behavior Analysis in Practice, 14, 679–703.