Real-World Functional Skills, Measured and Strengthened

This study asked why the benefits of computerized functional skills training persist for some participants but not others. Eighty-four older adults with mild cognitive impairment (MCI) and 68 with normal cognition received remotely delivered FUNSAT™ training twice weekly for up to 12 weeks; half of the MCI group first received a burst of computerized cognitive training. Thirty days after training, the researchers used machine-learning models to predict whether participants retained at least 80% of their gains on the relatively easy ticket-purchase task and the more difficult internet-banking task.

FUNSAT™ produced substantial gains, with effect sizes above 0.75, and most participants retained those gains: 78% for ticket purchase and 65% for internet banking. The strongest predictors of durability came from the training process itself—such as improvement in completion time and end-of-training performance—rather than from static participant characteristics alone. Computerized cognitive training also improved durability for participants with MCI. These findings show a positive, precision-oriented use of FUNSAT™: its detailed performance data can help identify different kinds of poor outcomes and support more personalized training designed to improve both mastery and lasting benefit.

The current study examined training gains maintenance over time and the role of booster sessions on sustaining training gains for the technology-based Functional Skills Assessment and Training program (FUNSAT™) in older adults with and without Mild Cognitive Impairment (MCI).

The findings support the efficacy of the FUNSAT™ training program as most participants maintained or increased their training gains over time, and that the booster training was beneficial in terms of sustaining gains, indicating that booster sessions play a critical role in skills training.

Decentralized clinical trials are leading to rapid changes in assessment technology, including an expansion of interest in remote delivery. As technology changes, some of the updates include migration to fully cloud-based software and data management, with attendant differences in hardware, response modalities, and modifications in the level of tester engagement. We present comparative data on successive generations of two widely used functional capacity measures, the Virtual Reality Functional Capacity Assessment Tool (VRFCAT) and the Functional Capacity Assessment and Training System (FUNSAT).

Four samples of healthy older individuals completed either the original device-resident, computer-administered versions, or cloud-based, tablet delivered versions of these tasks. For the FUNSAT, performance and correlations with age and cognitive task scores were generally similar across versions for English Speakers, though there were some differences related to the testing language for the cloud-based version. These results support the feasibility of migrating digital assessments to cloud-based delivery and substantiate fundamental similarity across delivery strategies.

This narrative review surveys digital technologies designed to make cognitive and functional assessment more representative of everyday life in schizophrenia. It contrasts conventional tests with digitally delivered simulations and remote measures, then reviews the evidence for several contemporary approaches. A dedicated FUNSAT™ section describes six realistic technology-related functional tasks, including banking, ticket purchase, medication management, telephone menus, pharmacy website use, and online shopping, along with outcomes such as completion time, errors, and task mastery.

The review presents FUNSAT™ positively as a standardized, cloud-capable system with automated administration, scoring, and data transfer; three alternative forms support repeated assessment. It notes convergence with cognitive and legacy functional-capacity measures while preserving the distinction between cognition and functional performance. FUNSAT™ was also sensitive to remotely delivered intervention effects, with a reported composite effect size of 0.75, and post-training performance related to cognition, other functional-capacity measures, and real-world activity captured by ecological momentary assessment. Together, these features support FUNSAT’s usefulness for ecologically relevant assessment, remote research, and measurement of meaningful treatment response.

This commentary interprets the clinical and intervention implications of the FUNSAT™ mastery study by Harvey and colleagues. It focuses on the finding that errors made very early in training—especially on the first and least difficult task—were more informative about eventual mastery than completion time or a global cognitive screening score. The authors consider why such early errors may occur, including task familiarity, technology experience, motivation, and the fit between a participant’s abilities and the training demands.

The positive contribution of FUNSAT™ is its ability to generate precise, task-level information early enough to act on it. Rather than treating early mistakes as evidence that a participant cannot benefit, the commentary frames them as signals for additional orientation, feedback, smaller training steps, or an alternative training pathway. FUNSAT’s automated capture of errors and performance makes this type of rapid personalization feasible and creates an opportunity to improve engagement, reduce nonmastery, and help more older adults obtain tangible benefits from functional skills training.