Currently, the benchmark standard for measuring everyday function generally takes the form of self-reporting or completing questionnaires. These methods are insufficient for a variety of reasons, including bias on the part of the individual reporting, the cognitive difficulties of those individuals who do have MCI, the unavailability of informants, a lack of objectivity and a lack of standardisation. It is therefore essential that a method of assessment that is ecologically valid, standardised and automated is developed. With that in mind, an international interdisciplinary collaboration of three computer engineers and a neuropsychologist from different laboratories have come together to overcome the barriers to understanding this decline. Dr Takehiko Yamaguchi is leading the computer engineering team from his base at the Virtual Reality and Data Science Laboratory, Suwa University of Science, Japan. The laboratory was established in 2017 and promotes research on a wide range of subjects, from basic to applied. As the name suggests, there is a focus on virtual reality and data science, with a view to disseminating research results and improving the quality of life of MCI patients. Yamaguchi is joined by Dr Tania Giovannetti, a neuropsychologist based at Temple University in Philadelphia where she specialises in cognitive aging and dementia, as well as two computer science experts, Professor Tetsuya Harada and Dr Hayato Ohwada, who are both based at the Tokyo University of Science, Japan. In order to identify meaningful markers, the team requires large samples and longitudinal studies, so they have developed an approach which uses sensitive measures of a wide range of cognitive abilities. Importantly, these are embedded within everyday tasks that are familiar to everyone and can be collected using a computer. This removes the need for humans to administer or score the tests, which lends a reliability to the results that are gathered. In addition, by relying on behaviour as opposed to measures of brain pathology the team can identify changes that are more likely to lead to functional problems. The importance of this is that different people can tolerate different levels of brain changes; it is well known that people who engage in cognitively challenging tasks throughout their life build a resilience to neuropathology. Thus, to quantify and understand an individual's quality of life, it is vital that someone's functioning can be measured, rather than the extent of their neuropathology. One of the team's latest projects involves the development of technology for early MCI screening based on the quantification of error and gaze behaviours. Their work has shown that cognitively healthy older adults without dementia make significantly more subtle errors in everyday tasks when compared to younger adults. "Micro-errors are significantly associated with measures of cognition. Also, in older they occur toward objects that are related to target objects and at choice-points in the task when there a larger range of possible next steps," says Giovannetti. "These findings suggest that contrary to popular belief, even relatively simple everyday tasks are vulnerable to the effects of cognitive ageing." Thus, the breakdown of everyday function in people with dementia does not appear suddenly but more likely reflects a gradual, slow decline that mirrors the slow decline in memory and other cognitive abilities.