A causal discovery framework for digital phenotyping
Ahmed Ibrahim · 2026
Read A causal discovery framework for digital phenotyping by Ahmed Ibrahim (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13473514.txt.
Description
Ahmed Ibrahim proposed in 2026 a shift in digital phenotyping, the tracking of behaviour moment by moment through phones and sensors. Prediction models in this field have hit a plateau: they can be accurate, yet they reveal little about why wellbeing rises or falls. The paper argues for moving from classification to causal modelling, so researchers can see which behaviours plausibly lead to which states.
The two-stage framework first learns daily behaviour summaries from mixed sensor data, then applies neuro-symbolic causal discovery to draw graphs of behaviour and psychology over time. In tests, even deep models barely beat chance at predicting stress, at a best area under the curve of 0.532. The causal stage instead flagged time-lagged hints, with lower sleep activity and reduced mobility often preceding stress episodes. The author stresses these are candidate markers for hypothesis testing, not proven causes.
Catalog record
| CREATOR | Ahmed Ibrahim |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 80 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |