The retooling of old cognitive tests as an interim step on the path to validating a next generation of neuropsychological paradigms and assays

Just as one brief example, the Trailmaking Test, which is still very commonly relied on as a measure of simple attention (Trails A) and divided attentional controls (Trails B), was developed during World War II as a measure of general intelligence for use by the U.S. Army. Psychologists Efforts to design and validate neuropsychological tools for the purpose of detecting or tracking change over time for patients experiencing the cognitive sequelae of Alzheimer’s disease (AD) may be described as a recurring pattern of inventing and reinventing the wheel. We have access to a plethora of episodic verbal learning tests, paired associate learning tests, and reasoning/problem-solving tests from which to choose, and they all rely on essentially the same small set of cognitive test paradigms. Despite decades of clinical trials and methods validation studies, nearly all of these instruments have demonstrated limited ability to detect change in the very earliest stages of the disease (prior to behavioral/clinical impairment) [1]. As our understanding of the early, preclinical manifestations of the disease has increased in recent years (cf [2]), there is an unmet demand for sensitive cognitive assays that can track very subtle change in presymptomatic individuals. Given this shortage of such cognitive assays to identify or detect reliably change in very early AD, early clinical identification is relying increasingly on more expensive and often invasive disease-related biomarkers [3,4] such as positron emission tomographic scanning of amyloid burden [4], the measurement of cerebrospinal fluid (CSF) levels of amyloid beta (Ab) and p-tau [5,6], and the measurement of structural brain changes to track disease progression [6,7]. Previous comparisons between biomarker measures of disease progression and cognitive and/or clinical measures suggest that CSF-based or neuroimaging biomarkers of disease progression are predictive of later, downstream cognitive decline [8]. In fact, Jack and colleagues [1] have proposed a now widely cited hypothetical model, suggesting that CSF measures of Ab and tau, and changes in brain structure, are all appropriate early measures of pathophysiological changes in AD, whereas impaired memory and clinical functions are identified as changes that occur later in the disease course and are of less relevance for the targeted identification of patients prior to a diagnosis of mild cognitive impairment (MCI). However, the specific cognitive measures (used in the Alzheimer’s Disease Neuroimaging Initia-