The Internet-based Cognitive Assessment Tool (ICAT) is a web-based, self-administered test of cognitive function. It lets patients complete a validated cognitive screening at home or in the clinic, without a clinician reading the tasks aloud. ICAT is built on the Copenhagen Research Platform (CARP), which handles the study set-up, participants, and data.

About ICAT
Cognitive impairment is common in mood disorders such as depression and bipolar disorder, but in routine care it is rarely measured. Standard screening relies on paper-and-pencil tests read aloud by trained staff, which takes clinical time and requires a quiet room and a scheduled visit.
ICAT is a digital version of the Screen for Cognitive Impairment in Psychiatry (SCIP), a brief and validated screening test. The patient completes all tasks on their own in a web browser. Spoken answers are captured with automatic speech recognition, using Google’s speech-to-text service in the Google Chrome browser, and the tasks are scored automatically.
Cognitive Tasks
ICAT has six tasks, each measuring a cognitive domain. Four tasks are adapted from SCIP, and Letter–Number Sequencing is adapted from the Wechsler Adult Intelligence Scale (WAIS). The Face task was added after the first version. Spoken answers are captured with speech recognition, and all tasks are scored automatically. The table below shows the details of the six tasks:
| Task | Paper-based counterpart | What the participant does | Score |
|---|---|---|---|
| Face | Ekman faces emotion recognition, paper version | A photo of a face is shown briefly. The participant then picks the emotion the face showed. | Correctly recognized emotions (0–34) |
| List Learning | SCIP Verbal Learning Test, immediate (VLT-I) | A list of 10 words is played as audio. The participant recalls as many words as possible and says them aloud. This is repeated three times with the same list. | Correctly recalled words over 3 trials (0–30) |
| Consonant Repetition | SCIP Working Memory Test (WMT) | A sequence of letters is played as audio. The participant then sorts a set of numbers in descending order within a time limit, which delays the response. Finally, the participant types the letters heard earlier. | Correctly recalled letters (0–24) |
| Letter–Number Sequencing | WAIS-III Letter–Number Sequencing (LNS), adapted | Sequences of letters and numbers are shown on the screen and played as audio, one at a time. The participant sorts the numbers and letters of each sequence and types them. Starts with a practice set. | Correctly sorted sequences (0–21) |
| Delayed List Learning | SCIP Verbal Learning Test, delayed (VLT-D) | The participant recalls the words from List Learning and says them aloud, without hearing the list again. | Correctly recalled words (0–10) |
| Visuomotor Tracking | SCIP Processing Speed Test (PST) | A table pairs 6 letters with codes made of circles and asterisks. Within 30 seconds, the participant enters the matching letter for up to 30 random codes, one by one. Starts with a practice set. | Correct matching letters (0–30) |
A full session takes about 20–30 minutes.
How It Works
- Researchers set up the study in CARP. ICAT downloads its study configuration from CARP. The configuration specifies which tasks to include, which face images to use, and which words to use in the List Learning and Delayed List Learning tasks.
- Participants receive a link, log in, and complete the tasks in their web browser.
- The results are uploaded to CARP, where researchers can follow the study and download the data. This includes raw task scores, all answers, response timing, and technical details about the browser and operating system.
- Available in Danish and English, and prepared for localization, so other languages can be added if needed.
Updates Since the First Version
Since the first version was published in 2019, ICAT has been updated with:
- A redesigned and improved user interface.
- A new task for emotion recognition (Face).
- Full integration with CARP: study configuration is read from CARP and data is uploaded to CARP.
Evaluation
- Design and feasibility (2019): ICAT was developed through a user-centred design process. In healthy volunteers, it had high usability ratings and its scores correlated with the paper-based SCIP (Hafiz et al., 2019).
- Validation in bipolar disorder (2021): In patients with bipolar disorder and healthy controls, ICAT total scores correlated strongly with SCIP scores (r = .72), usability was rated high, and ICAT was at least as sensitive to cognitive impairment as SCIP (Miskowiak et al., 2021). See also our news post on the validation study.
Developed By
ICAT was developed at the Copenhagen Center for Health Technology (CACHET) and the Technical University of Denmark (DTU), in collaboration with Kamilla Miskowiak and Lars Vedel Kessing at the Copenhagen Affective Disorder Research Centre (CADIC), Psychiatric Centre Copenhagen. The first version was developed as part of Pegah Hafiz’s PhD project at DTU Compute (2017–2020), funded by the TEAM Innovative Training Network under the EU Horizon 2020 Marie Skłodowska-Curie programme (grant agreement No. 722561).
CARP Components Used
- CARP Web Services (CAWS) for study configuration, participant management, and data storage.
- The CARP TypeScript client (@carp-dk/client), which the ICAT web app uses to communicate with CAWS.
Publications
- Hafiz P, Miskowiak KW, Kessing LV, Jespersen AE, Obenhausen K, Gulyas L, Żukowska K, Bardram JE. The Internet-Based Cognitive Assessment Tool: System Design and Feasibility Study. JMIR Formative Research. 2019;3(3):e13898.
- Miskowiak KW et al. Internet-based cognitive assessment tool: Sensitivity and validity of a new online cognition screening tool for patients with bipolar disorder. Journal of Affective Disorders. 2021;289:125–134.
- Hafiz P. Pervasive Computing Technologies for Ambulatory Cognitive Assessment. PhD thesis, DTU Health Technology, 2020.