A Blueprint for Digital Tools in Early Detection of Cognitive Impairment: The Development and Validation of the Spatial Performance Assessment for Cognitive Evaluation (SPACE)
Bibliographic record
Abstract
As the global population ages, the prevalence of dementia continues to rise, creating an urgent demand for tools that enable early detection and intervention. Alzheimer’s disease (AD), the most common form of dementia, is typically preceded by a stage known as Mild Cognitive Impairment (MCI), which is characterised by subtle but measurable declines in cognitive function that do not yet interfere with daily activities. Among other regions, MCI and AD primarily affect the medial temporal lobe, including the hippocampus and entorhinal cortex, which are central to spatial coding and position tracking during navigation. The early degeneration of regions supporting spatial navigation provides a strong theoretical basis for targeting navigation abilities in the early detection of cognitive impairment. Yet, traditional cognitive screening assessments widely used in clinical practice often neglect spatial navigation abilities and instead focus on memory, executive function, and language. These tools may lack sensitivity to early cognitive changes, require trained administrators, and are susceptible to biases related to education, language, and culture. As a consequence, many individuals are only diagnosed after time- intensive neuropsychological assessments that require specialised personnel and often involve costly and invasive procedures such as neuroimaging or biomarker testing. By then, symptoms are usually apparent, and interventions are less effective. Digital technologies offer a means to deliver standardised, scalable, and sensitive assessments, including spatial navigation tasks, to better detect early cognitive impairment. Although several digital tools targeting navigation have demonstrated potential for detecting cognitive impairment, few have undergone the rigorous validation needed for clinical or public health use. This thesis presents the development and validation of SPACE, a tablet-based tool designed to assess spatial navigation deficits as early markers of cognitive impairment. Validation was conducted through a structured multimodal pathway integrating usability, cognitive performance, neural correlates, and diagnostic evaluation. The first step of the validation pathway emphasises iterative usability studies as a key component of the validation process, enabling continuous optimisation of the digital tool through both subjective and objective measures. Article I tested design features of SPACE across three studies to enhance accessibility, including a simplified control interface to reduce motor and cognitive demands, a rotational aid to support heading estimation, and a simplified spatial configuration of the tasks. These adaptations improved accessibility for older adults while maintaining sensitivity to cognitive variation and established a user-centred foundation for digital cognitive assessments intended across broad age groups. Article II assessed whether SPACE tasks predicted cognitive impairment as measured by the Montreal Cognitive Assessment (MoCA), while accounting for age, gender, and modifiable dementia risk factors. Two tasks in SPACE (i.e., pointing and perspective taking) predicted MoCA scores, while other navigation-based tasks captured variance independent of MoCA. A further analysis revealed a dissociation, as some participants performed well on MoCA but poorly on navigation tasks, and vice versa, suggesting that SPACE taps into complementary cognitive domains not fully captured by conventional screening tools. Building on these findings, Article III extended the validation to neural correlates of spatial navigation by examining whether SPACE performance explained variance in hippocampal volume beyond full neuropsychological assessment, including the MoCA. Here, the performance of two navigation-based tasks in SPACE (i.e., path integration and mapping) showed significant associations with hippocampal volume, providing anatomical validation that SPACE targets brain systems known to be affected early in AD progression. While the previous studies focused on healthy participants, Article IV evaluated the diagnostic classification performance of SPACE across clinical stages of cognitive impairment, as defined by the Clinical Dementia Rating (CDR) scale. SPACE demonstrated high diagnostic accuracy even when discriminating between adjacent, subtler stages, while maintaining sensitivity and specificity comparable to, or exceeding, those of existing digital tools. Moreover, a shortened version, sSPACE, maintained robust classification performance while reducing administration time, offering a format more suitable for in-clinic use and with potential applicability to large-scale screening efforts. Taken together, the work presented in this thesis establishes SPACE as a psychometrically robust, biologically anchored, and clinically promising tool for the early detection of cognitive impairment. By targeting spatial navigation deficits and adopting a multimodal validation framework, this thesis offers a methodological blueprint for the responsible development and validation of future digital cognitive assessments aimed at earlier identification of individuals at risk of cognitive impairment, when interventions may still provide meaningful benefit.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.068 | 0.119 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.006 | 0.003 |
| Science and technology studies | 0.001 | 0.008 |
| Scholarly communication | 0.006 | 0.008 |
| Open science | 0.004 | 0.009 |
| Research integrity | 0.006 | 0.012 |
| Insufficient payload (model declined to judge) | 0.004 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".