Aphantasia test: method

What it measures

Self-rated vividness of voluntary visual imagery (the headline score), plus single-item vividness ratings for auditory, olfactory, gustatory and tactile imagery. It records how people describe their imagery; it does not measure imagery objectively.

Procedure

14 original items in a fixed order: 10 visual scenes, then one item each for sound, smell, taste and touch. Instruction: 'Close your eyes if it helps. Try to bring each thing to mind, then choose how clear it was.' One item at a time; the visitor taps an answer (or presses keys 1–5), the answer is highlighted briefly and the next item follows; 'Back' (or Backspace) returns to change an answer. Untimed; the time to each final answer is recorded only for quality flags. Answer scale: 1 'Nothing: I know what it is, but I don't see, hear or feel it', 2 'Faint and vague', 3 'Moderately clear', 4 'Clear and lively', 5 'As clear as the real thing'.

Parameters

Scoring

Visual score = mean of the 10 visual items (1–5, shown to one decimal); higher means more vivid. The four other senses are shown one by one and as their mean. Descriptive band: 'no pictures' when all 10 visual answers are 1; 'faint' when the visual mean is 2.0 or lower; 'very vivid' when it is 4.7 or higher; otherwise 'in the middle'. These cut-offs copy the usual VVIQ groupings (16 = none, 17–32 = weak, 75–80 = very vivid; Dance et al. 2022; Wright et al. 2024) applied to an item average, but they are not validated for these items.

Valid-run rules

Every item must be answered exactly once, with the item-set version the page received; otherwise the run is rejected. Two quality flags are stored with the run and never shown or used to block it: 'straight_lining' (all 14 answers identical) and 'too_fast' (at least half of the answers given less than 700 ms after the item appeared). Flagged runs can be left out of aggregates.

Differences from the classic task

The VVIQ (Marks, 1973) has 16 visual items in four scenes and is often given with eyes open and eyes closed. This check uses 10 original, unrelated scenes, one rating pass, its own answer wording, and single items for the other senses (the Psi-Q is not used or adapted). Scores are on their own scale and cannot be converted to VVIQ totals.

Limitations

Self-report only: people mean different things by 'clear', and the result depends on how each person reads the scale. No objective check of imagery (such as binocular-rivalry priming) is included. The bands are not validated cut-offs. Translated item sets have no local validation or norms. Not a diagnosis; aphantasia is not an illness.

Languages and review status

English original items; translations into 99 further languages (machine-assisted translation with automated script and spelling checks). Item IDs, order and scale length are fixed across languages; each language has its own item-set version so answers to different wordings are never mixed.