Cold probe data vs trial-by-trial data

By the IEP Atlas editorial teamLast updated November 1, 2026

Short answer

A cold probe records whether the student gets the first trial of a target right, before any teaching or prompting that session. Trial-by-trial data records every trial during teaching, with prompts. The deciding test: do you need to know what the student can do on their own right now (cold probe), or how they perform while being taught (trial-by-trial)?

Both methods count correct responses, so they are easy to confuse. The difference is timing. A cold probe is taken at the start of a session, on the first opportunity, with no warm-up, model or prompt. It answers one question: does the student know this today? Trial-by-trial data is taken across a block of teaching trials and shows how the student does with help, how prompts are fading and how accuracy changes within the session.

Each method has a cost. Trial-by-trial data is detailed but takes attention away from teaching, and it can overstate skill, because accuracy in trial 9 benefits from the model in trial 1. Cold probes are fast and show independent skill, but they give one data point per target per session. Many BCBAs and SLPs use both: cold probes to judge mastery, trial-by-trial data when a target is stuck. For the broader picture, see the IEP data collection pillar.

The deciding question

Ask: what decision will this data drive?

If the decision is "has the student mastered this, and can we move on?", use a cold probe. Mastery means independent performance without a warm-up.
If the decision is "why is this target not moving, and which prompt level is the student at?", use trial-by-trial data for that target until you know.

Side-by-side table

	Cold probe	Trial-by-trial data
Definition	First trial of each target, before teaching that session; scored correct or incorrect	Every trial in a teaching block, scored with the prompt level used
When taken	At the start of the session or the start of each target	Throughout teaching
Prompts allowed	None on the probe trial	Yes; the prompt level is recorded
Data points per target per session	1	Often 5 to 20
What it shows	Independent skill retained from the last session	Performance during teaching, prompt dependence, within-session learning
Time cost	A few seconds per target	Constant recording during the session
Typical mastery rule	Correct on the cold probe on 3 consecutive sessions, sometimes with 2 different adults	A percent correct over a set number of sessions, such as 80% of trials independent on 3 sessions
Basis in law or practice	Practice choice; the IEP must say how progress is measured, not which method^[1]	Same: practice choice within the IEP's stated method
Who decides	BCBA, SLP or teacher writing the program; IEP team for goal measurement	Same
Where it is documented	Probe sheet; skill acquisition graphs	Trial data sheet; session graphs

How to collect a cold probe

List the targets for the session on a probe sheet, one row per target.
Start the session by presenting the first target with the standard instruction and no prompt, model or warm-up.
Wait the response time your program sets, often 3 to 5 seconds.
Score it: correct and independent, or not. Any prompt, self-correction after a hint or delayed response counts as incorrect for the probe.
Then teach that target as planned, collecting trial data if the program calls for it.
Repeat for each target in the session.

Probing every target at the very start of the session is ideal. When that is not practical, probe each target the first time it comes up. Keep the condition the same each session.

How to collect trial-by-trial data

Present each trial, record the response and the prompt level used, and total the independent trials at the end of the block. The trial-by-trial data guide covers coding and prompt levels in detail, and the opportunity trial data sheet has the columns. The opportunities measurement guide explains how to report the result in a goal.

Three scenarios

All three are fictional.

Scenario 1: An SLP working on 12 vocabulary targets. A first grader is learning 12 category words. The SLP has 25 minutes, twice a week. Collecting 10 trials on each word would leave little time to teach. Answer: cold probe each word at the start of the session, then teach. One probe per word gives a clear mastery signal. After 3 consecutive correct cold probes, the word moves to a maintenance list checked every other week.

Scenario 2: A BCBA reviewing a stuck target. A student's cold probe on "identify coins" has been incorrect for 9 sessions. The program reports 85% correct during teaching. Answer: switch to trial-by-trial data with prompt levels for this target. The 85% hides the fact that most correct trials followed a gestural prompt. Trial data shows the student is prompt-dependent at the gesture level, so the team changes the prompt-fading plan. The prompt hierarchy examples page covers fading steps.

Scenario 3: A teacher reporting an IEP goal. A goal reads: "the student will read 20 of 25 functional sight words correctly on 3 consecutive probes." The teacher teaches the words daily in a small group. Answer: a cold probe of all 25 words once a week, before that day's teaching. That measures independent reading, which is what the goal promises. Daily teaching data can stay informal.

When to use both

Most skill acquisition programs use both in a predictable pattern:

Target status	Cold probe	Trial-by-trial data
New target, first sessions	Yes, to set a baseline	Yes, to watch prompt levels
Making steady progress	Yes, every session	Optional, or a sample of trials
Stuck for 5 or more sessions	Yes	Yes, full trial data until the problem is clear
Met cold probe mastery	Move to maintenance probes	No
Maintenance	Every 1 to 4 weeks	No

This pattern saves time on targets that are moving and spends it where the data is needed. The maintenance data and generalization data guides cover the later stages.

Common mistakes

Warming up before the probe. One practice trial before the "cold" probe turns it into a warm probe. Probe first.
Counting a prompted response on a probe. A gesture, a pause with a look or a repeated instruction makes it incorrect for the probe.
Reporting teaching-trial percentages as mastery. Accuracy during teaching usually overstates independent skill. Report cold probe data for mastery.
Probing at different times of the session. A probe at minute 25 after related practice is not comparable to one at minute 1.
Too few probe data points before a decision. One cold probe per session means you need several sessions before judging a trend. Repeated measures over time are what make the data useful.^[2]

Tools

Find all free IEP data sheets, including the opportunity trial data sheet for trial data and the accuracy data sheet for probe lists.
Read how to choose an IEP data collection method and digital IEP data collection.
Glossary: probe, trial-by-trial data and data sheet.

Frequently asked questions

What is a cold probe in ABA?

A cold probe is the first trial of a target in a session, presented before any teaching, model or prompt. The student's response is scored correct or incorrect. It shows whether the student can do the skill independently after time away from it.

How many cold probes show mastery?

Many programs use correct cold probes on 3 consecutive sessions, sometimes with two different adults or settings. The program or IEP should state the rule in advance so everyone applies it the same way.

Can cold probe data be used for IEP progress reports?

Yes, if the goal's measurement matches it. A goal stated as correct responses on probes is measured well by cold probes. The IEP must say how progress is measured, and the data method should match that statement.

Use it in IEP Atlas

Sources

34 CFR §300.320: Definition of individualized education program, Legal Information Institute, Cornell Law School
Progress Monitoring: Data & Goal Setting, National Center on Intensive Intervention

Written and checked by the IEP Atlas editorial team under our editorial policy. This guide describes general practice and federal rules. It is not legal advice; your state may add requirements, and the IEP team makes the decisions for each student.