In its clinical form, the Attention Training Technique is homework. The manual asks for practice twice a day at approximately twelve minutes a session[1], and uses an ATT summary sheet to monitor that practice between appointments[1]. A practice record is simply that monitoring, kept by you.
What a record is for
It answers one question: is the practice actually happening on schedule?
That question matters more than it looks, because the manual is candid about the gap between the ask and the reality — patients are asked to practise twice a day, and in reality most only manage to do it once[1]. A record is how you find out which of those you are doing, rather than estimating it from memory.
What belongs on it
Short, and mostly factual:
| Field | Why |
|---|---|
| Date and time | The protocol is about regular repetition on a schedule[1]; the timestamp is the whole record |
| Completed / not completed | A session that was started and abandoned is useful information about conditions, not a moral failing |
| Anything that got in the way | Practical, not emotional — no quiet place, no time, fell asleep |
That is close to all of it. The temptation is to add more.
The one thing to be careful about
Grading your sessions is the part to think twice about.
ATT is meant to be attentionally demanding, and to stay that way rather than becoming easier with practice[1]. So a rating of “how well it went” has no stable meaning: a hard session is the exercise working as designed. Wells also notes the specific misconception that intrusive thoughts mean the practice was ineffective[1] — exactly the judgement a self-rating invites you to make.
There is a deeper reason. The pattern the exercise targets is self-focused, threat-monitoring attention[1]. A daily habit of inspecting your own performance and mood after every session is a way of practising that pattern immediately after practising against it. If you record anything subjective, keep it to whether you kept bringing your attention back — which is what the script actually asks of you[1].
A record is not a progress chart
It is worth being explicit, because the instinct to plot a line is strong. Nothing in the protocol offers a metric that should go up. The exercise does not get easier by design, the manual’s success criterion is the returning of attention rather than the holding of it[1], and checking for results puts attention back where the exercise is trying to move it from.
A practice record tells you whether you practised. That is a smaller claim than a progress chart, and a more honest one.
A minimal template
Date Time Done? Notes (practical only)
2026-09-07 08:10 yes —
2026-09-07 19:40 no no quiet room
2026-09-08 08:05 yes —Three columns and an optional fourth. If you use an app that logs sessions, that log is already this record — the value is in looking at it weekly and seeing the pattern, not in the recording itself.
For what the schedule is meant to be, see how often to practise ATT. For the misreadings a record can encourage, see are you doing ATT right?.
Questions and answers
- Should I track my ATT practice?
- The published protocol treats ATT as scheduled homework and uses a summary sheet for monitoring it, so recording that a session happened is part of how the technique is delivered.
- What should I write down?
- That the session happened, and when. Wells' protocol asks for practice on a schedule; a record's job is to show whether the schedule is being kept.
- Should I rate how well it went?
- Be careful here. Checking a session for results turns attention back onto your own state, which is the pattern the exercise is meant to interrupt. Recording that you practised is safer than grading how well you practised.
References
- Wells, A. (2009). Metacognitive Therapy for Anxiety and Depression. Guilford Press. Chapter 4: Attention Training Techniques.
Published by the makers of Heed, an app for practicing ATT-style sessions. iPhone · Android. Heed is a self-practice tool inspired by ATT research — not therapy, and not a medical device.