Neuropsychology
Assessment Battery Builder
A referral question, ninety minutes and seventeen task types. Choose what to administer, then find out what your battery could and could not conclude.
Simulated - generic task descriptions, fictional referrals, no real tests and no scoring of anybody
Learning objective
By the end you should be able to select tasks from a referral question, and say which measures make the rest interpretable.
About 30 minutes across three referrals. Nothing you do here is saved or sent anywhere.
Not a battery, and not a clinical tool
Every task here is described generically - "learning and recalling a word list", not any published test. No real test, subtest, item, stimulus or norm is reproduced, no score is produced for anybody, and the durations are round teaching numbers rather than published administration times.
The three referrals are invented. Real assessment involves a history, an interview, informant report, observation across the session and a formulation, none of which a task list can carry. This activity is about the reasoning behind a selection, not about how assessment is actually done.
- Read the referral question. It is the only thing that makes a task relevant or irrelevant.
- Tick tasks. The panel on the right updates as you go: time used, what the battery can support, and what it cannot.
- Submit the battery for a verdict, then try the other two referrals - the right answer changes completely.
First, a judgement
You have ninety minutes and a referral asking whether the memory problems a fictional person reports show up on testing. You have already chosen one good memory test.
The builder
Key terms
- Referral question
- What this assessment is being asked to answer. It is what makes a task relevant here, rather than how good the task is.
- Comparison domain
- An area you expect to be unaffected, tested so that a low score elsewhere has something to be low against.
- Premorbid estimate
- An estimate of where this person was functioning beforehand. Without one, a current score has no baseline.
- Performance validity measure
- A task included to check whether the other scores can be taken at face value. It is not an accusation about the person.
- Confound measure
- Something measured because it could lower scores for reasons outside the referral question — mood, pain or sleep, say.
Build a battery for this referral
Goal
This battery, against this referral
What this battery could say
The tasks you have chosen
| Task | Domain | What it measures | Minutes |
|---|
Every task on the menu
| Task | Domain | Minutes | Note |
|---|
Durations are round teaching numbers. No published administration time, norm or item is reproduced, and the task descriptions are generic on purpose.
Challenge - a conclusion to weigh
What this demonstrates
A battery is a set of inferences, and the referral decides which ones are needed. That is why the tasks that answer nothing matter: a domain you expect to be intact, and a validity check, are what let you say a low score means what you think it means. Longer is not better, and two measures of the same thing are not waste.
For teaching elsewhere: take this activity as one self-contained block of HTML, on the clipboard or as a file. Either way it is styled so that it will not disturb the page you put it into.