Researchers at Aalto University in Finland and Japan's Kochi University of Technology have developed an artificial intelligence (AI) that takes individual patient difference into account to optimize a user interface. Findings were published March 15 in IEEE Pervasive Computing Journal.
"The majority of available user interfaces are targeted at average users. This "one size fits all" thinking does not consider individual differences in abilities—the ageing and disabled users have a lot of problems with daily technology use, and often these are very specific to their abilities and the circumstances," said postdoctoral researcher Jussi Jokinen at Aalto University. "There are ways to automatically optimize the user interface, but this is efficient only if we have a realistic model of the user. Previously, designers did not have detailed models that are based on psychological research and can be used to predict, how different individuals perform in interactive tasks.”
The AI uses a predictive model of interaction to determine an individual’s ability in text entry on a touch screen. By combining psychological research on finger pointing and eye movements, the AI can predict text entry speed, number of errors and proofreading.
"After this prediction, we connected the text entry model to an optimizer, which iterates through thousands of different user interface designs. No real user could of course try out all these designs,” said Jokinen. “For this reason, it is important that we could automatize the evaluation with our computational model.”
In this study, researchers evaluated the AI in its ability to predict the needs of an individual with essential tremor. Following the optimization of the text entry interface for individuals with essential tremor by the AI, users were able to input text with almost no errors.
"We started with text entry, which is an everyday task,” concluded Jokinen. “We chose to simulate and optimize for essential tremor, because it makes text entry very difficult. Now that we have confirmed the validity and usefulness of the model, it can be extended to other use cases and disabilities. For example, we have models for simulating how being a novice or an expert with an interface impacts user's performance. We can also model how memory impairments affect learning and everyday use of interfaces. The important point is that no matter the ability or disability, there must be a psychologically valid theory behind modelling it. This makes the predictions of the model believable, and the optimization is targeted correctly.”
Cara joined TriMed Media in 2016 and is currently a Senior Writer for Clinical Innovation & Technology. Originating from Detroit, Michigan, she holds a Bachelors in Health Communications from Grand Valley State University.