Biruni

In developmentBuilt in Riyadh for MENA research

Run experiments in Arabic.Measure to the millisecond.

Biruni is the experiment platform for behavioral, cognitive, and psycholinguistic research in MENA. Build timed tasks without code, see live attrition and quality insights as sessions complete, and collect reaction-time data you can defend in peer review — with participant data that stays in the Kingdom.

Free to design and pilot. You only pay for completed sessions.

Example · Flanker

flanker.ar · attentional control
consentpracticematchconflict

Incongruent · trial 48150 ms

الحَرْف الأَوْسَط؟

ننبنن

F · بJ · ن

Runtime telemetry

Visual onset Δ0.3 ms
Stimulus duration150 ms
Dropped frames0
Response (key F)486 ms
Illustration of the Biruni Studio: a four-node experiment tree branching from consent through practice into match and conflict Flanker blocks, an Arabic selective-attention stimulus frame asking for the middle letter in a noon-flanked ba array, and a Runtime telemetry panel reporting sub-millisecond onset accuracy with no dropped frames.
audio–visual onset variance
≤10 ms
CONSORT attrition & quality dashboards
Live
Arabic words & nonwords with RT norms
20,000
data residency for GCC review boards
In-Kingdom

Why Biruni

Four things a MENA experiment platform has to get right

Generic survey tools miss these. Each one alone is enough to invalidate a study — or leave you blind until fieldwork is over.

  • 01

    Arabic that actually renders

    HarfBuzz-WASM shaping with full bidi resolution — correct ligatures, Tashkeel placement, and mixed Arabic/Latin/digit runs.

    Identical shaping in the builder, the preview, and the participant runner.

  • 02

    Timing you can defend in review

    Visual onset scheduled on VSYNC, responses stamped with performance.now(), audio scheduled on the Web Audio clock.

    Per-session calibration and per-trial frame telemetry export next to your response data.

  • 03

    Results and insights, out of the box

    Live dashboards for every survey and experiment — attrition, completion, RT distributions, accuracy by condition — without waiting to open a spreadsheet.

    CONSORT funnels, outlier flags, and long-format exports ready the moment sessions complete.

  • 04

    Data that never leaves the Kingdom

    Participant data is stored in AWS me-central-1, with PDPL/SDAIA consent, residency, and audit trails built into the schema.

    Generated PDPIA and IRB kits, plus a participant DSAR self-service portal.

The gap

Arabic-language research runs on tools that were never built for it

Every row below is a reason a MENA study gets delayed, rejected, or quietly compromised — and what Biruni does instead.

On general-purpose toolsLigatures corrupt, Tashkeel floats, and RTL bounding boxes misalign in reading tasks.

On BiruniReal text shaping with an RTL inspection overlay for shaped-glyph boxes.

On general-purpose toolsStimulus libraries are Western — no Arabic frequency, AoA, or concreteness norms.

On BiruniAraLex, Kalimah, the Arabic Lexicon Project, and Gulf/Levantine picture sets in the picker.

On general-purpose toolsParticipant data lands in EU/US regions, and GCC review boards reject the study.

On BiruniIn-Kingdom residency by default, with the compliance paperwork generated per study.

On general-purpose toolsTiming accuracy is a claim in a changelog you cannot verify for your own sample.

On BiruniCalibration and dropped-frame counts recorded per session and exported with the data.

On general-purpose toolsAnything beyond a survey means hand-writing and debugging JavaScript.

On BiruniA no-code trial builder, with a sandboxed JS escape hatch when you genuinely need one.

On general-purpose toolsResults stay locked in raw CSV until someone scripts R or SPSS — no live view while the study runs.

On BiruniOut-of-the-box dashboards: CONSORT attrition, RT and accuracy by condition, outlier flags, then one-click export.

Stimulus bank

Normed Arabic materials in the picker — not a translated English list

Frequencies, age of acquisition, picture-naming norms, and dialect filters sit next to the trial you are building. Ecological validity starts before the first participant.

  • AraLex Corpus40M tokens

    Root, pattern, and surface frequency · orthographic density

  • Kalimah Norms2,467 + 30k

    Age of acquisition · concreteness · lexical ambiguity

  • Arabic Lexicon Project20,000

    Empirical reaction times · error rates · reader splits

  • Gulf Arabic Picture Database460

    Name agreement · familiarity · visual complexity · AoA

Screen by dialect at recruitment

  • Najdi
  • Hejazi
  • Eastern Province
  • Qatari
  • Emirati
  • Egyptian
  • Levantine
  • Maghrebi
Lexical

Nonword · trial 84until key

كَلِمَة أَمْ غَيْر كَلِمَة؟

مُكَرْتَب

F · كَلِمَةJ · غَيْر
Illustration of a visual lexical-decision trial on an Arabic nonword token, wired to Arabic Lexicon Project–style norms in the picker, with Runtime response stamps.
Naming

Gulf picture · trial 12AoA linked

ما اسْم هذه الصُّورة؟

Typed / spoken response · name agreement logged

Illustration of a Gulf Arabic picture-naming trial linked to regional name-agreement norms, with Runtime onset and naming-latency telemetry.

Trial faces only — lexical decision against ALP-style norms, picture naming against Gulf sets.

Task examples

What a trial looks like in Arabic

Tabs on the left. Scroll — the studio preview and copy on the right step upward through each paradigm.

Attention

Arabic Flanker for attentional control

Run the classic flanker in Arabic — middle-letter judgments over shaped arrays, with the same VSYNC timing you defend in review.

Explore Academic suite
flanker.ar · attentional control
consentpracticematchconflict

Incongruent · trial 48150 ms

الحَرْف الأَوْسَط؟

ننبنن

F · بJ · ن

Runtime telemetry

Visual onset Δ0.3 ms
Stimulus duration150 ms
Dropped frames0
Response (key F)486 ms
Illustration of the Biruni Studio: a four-node experiment tree branching from consent through practice into match and conflict Flanker blocks, an Arabic selective-attention stimulus frame asking for the middle letter in a noon-flanked ba array, and a Runtime telemetry panel reporting sub-millisecond onset accuracy with no dropped frames.

See the full platform and template list →

Bring your next Arabic study to Biruni

We are building with a small group of MENA labs before general availability. If your work needs accurate timing, real Arabic rendering, and in-Kingdom data residency, we want to hear what your study requires.

What selected groups get
A complimentary 12-month Lab license: 5 author seats, 2,500 participant tokens, full norms access.
Who we're looking for
Research groups in cognitive psychology, linguistics, neuroscience, and education across Saudi Arabia, the UAE, and Qatar.
What we ask in return
Run one real study, publish one paradigm to Open Materials, and tell us candidly where we're wrong.