Differences in the way healthy and cognitively impaired individuals used their smartphones were enough to tell them apart.
How they did it: Apple researchers monitored the app usage of 113 adults between the ages of 60 and 75 over 12 weeks. Thirty-one of them had clinically diagnosed cognitive impairment; 82 were healthy. For every session—from the moment users unlocked their phones to the moment they locked them again—the researchers logged the sequences of apps used and categorized the sessions into different types. The data was used to train a machine-learning model.
The results: The model was able to distinguish healthy from cognitively impaired users roughly 80% of the time, or 30% more often than chance. The results also showed that the context in which apps were used was important to the model’s prediction. An app like Messages used alone, for example, was strongly associated with a healthy individual, but used together with Mail was more strongly associated with a cognitively impaired individual.
Why it matters: Around 15% to 20% of people 65 and older suffer from mild cognitive impairment, which affects memory and thinking skills and increases the risk of developing Alzheimer’s. Early diagnosis and treatment could help reverse the symptoms.
Future work: Apple has been conducting similar studies for a while now, but it’s still unclear how they intend to apply the results. This study in particular has been conducted on a very limited number of users and considers simple statistics on app usage. The researchers hope to incorporate richer data into their analysis, including the order in which apps are used, the time of day, and the users’ movements.
Geoffrey Hinton tells us why he’s now scared of the tech he helped build
“I have suddenly switched my views on whether these things are going to be more intelligent than us.”
ChatGPT is going to change education, not destroy it
The narrative around cheating students doesn’t tell the whole story. Meet the teachers who think generative AI could actually make learning better.
Deep learning pioneer Geoffrey Hinton has quit Google
Hinton will be speaking at EmTech Digital on Wednesday.
We are hurtling toward a glitchy, spammy, scammy, AI-powered internet
Large language models are full of security vulnerabilities, yet they’re being embedded into tech products on a vast scale.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.