About OCTAL
OCTAL (Oxford Cognitive Testing Portal) is a browser-based platform for remote digital cognitive assessment, developed by the Cognitive Neurology Research Group at the University of Oxford. It comes in two versions, designed for different research contexts.
| OCTAL | MiniOCTAL | |
|---|---|---|
| In one line | Comprehensive cognitive battery for clinical research and patient characterisation. | Quick, mobile-first assessment for scalable remote data collection. |
| Tasks | 11 | 5, plus an optional speech module (verbal fluency and language production) |
| Time | From 5 minutes; 20 minutes for the core subset | About 10 minutes; 16 with the speech module |
| Device | Tablet or computer | Smartphone, tablet or computer |
| Languages | English, Simplified Chinese | 44, more on request with native-speaker proofreading |
| Memory | Rey-Osterrieth Complex Figure (copy and recall) · Object-in-Scene · Oxford Memory Test · Verbal Learning Memory · Corsi-Block | Digit Memory (copy, immediate and delayed recall) · Object-in-Scene (immediate and delayed recall) · Tangram (delayed recall) |
| Speed and attention | Simple Reaction Time · Digit Symbol Substitution | Simple Reaction Time · Digit Symbol Substitution |
| Executive function | Trail-Making · Tower of London · Go-NoGo · Option Generation | |
| Visuospatial | Rey-Osterrieth Complex Figure (copy) | Tangram (copy, with motor trace analysis) |
| Validation | 1,664 participants, from healthy adults across the lifespan to memory clinic patients. A 5-minute screen separated Alzheimer's disease from subjective cognitive decline as well as the ACE-III (AUC 0.92); a 20-minute subset did better (AUC 0.97). Reliable over six months. npj Digital Medicine, 2026. | More than 1,000 people, including patients with depression, mild cognitive impairment, Alzheimer's disease, Lewy body dementia and Parkinson's disease. Over 75% of patients with dementia completed it. In a memory clinic it detected ACE-III-defined impairment with excellent accuracy and predicted the ACE-III total to within 6.4 points. Publication in preparation. |
| Best suited to | Studies that need a comprehensive cognitive profile or target a very specific effect. OCTAL needs more maintenance and analysis than MiniOCTAL; contact us to discuss suitability before applying. | Large population studies, multilingual cohorts, remote screening, epidemiological research and participant-friendly protocols. |
Cognitive profile
Participants open a link on their own phone and complete the tasks in about ten minutes; there is no app to install and no account to create. Within minutes of each session you have the raw data, the processed scores and a profile like these. Each spoke is one measure, compared with healthy people of the same age. Hover over a name, a dot or a coloured band for an explanation.
- Executive function
- Visuospatial
- Short-term memory
- Long-term memory
How to read it. Each dot is one measure, compared with healthy people of the same age. The further out it sits, the better. The colours are a traffic light: green is the normal range, orange is below average, red is well below.
The bands are set in standard deviations, the usual unit for how far a score is from the average. Scores of healthy people spread out in a bell shape. Most, about two in three, fall within one standard deviation either side of the average. A score one standard deviation below means that about 84 in 100 peers do better; two below, about 98 in 100.
Research applications
OCTAL and MiniOCTAL have been used across a range of clinical and population studies. For the latest findings and the full publication list, see News & Publications; for where the platform is running, see the map on the home page.
Large-scale normative data across the adult lifespan, with cohorts established in the UK, China and Gaza.
Alzheimer's disease · Mild cognitive impairment · Subjective cognitive decline · Parkinson's disease · Dementia with Lewy bodies · Limbic encephalitis · Frontotemporal dementia
Cognitive consequences of SARS-CoV-2, including processing-speed deficits in long COVID and attentional changes in acute infection survivors.
Depression and other psychiatric cohorts.
Our approach
For elderly participants and patients with cognitive difficulties, the assessment experience must be as frictionless as possible. No app downloads, no account creation, no login screens. All OCTAL tasks are entirely web-based: with an internet connection and a common browser such as Chrome, Firefox or Safari, it works.
Every task is calibrated to avoid floor effects in impaired populations while staying sensitive enough that healthy or high-functioning participants do not hit the ceiling. Both versions of the platform have been validated and their reliability tested, though only part of our data has been published so far, with more on the way.
Anonymisation is built into data collection: no personally identifiable information is stored or transmitted during a study. The platform is hosted on European servers operated by Open Science Tools, the company behind Pavlovia, which holds ISO 27001 certification. Data are stored on encrypted, access-controlled systems available only to the researchers named on each study. Details are on the data hosting page.
Interested in using OCTAL? Academic licences are issued through Oxford University Innovation and start with a conversation with us.
Apply for a licenceoctal@psy.ox.ac.uk