NINDS CDE Notice of Copyright
Lumos Labs NeuroCognitive Performance Tests (NCPT)
Lumos Labs NeuroCognitive Performance Tests (NCPT)
Please visit this website for more information about the instrument: Lumos Labs NeuroCognitive Performance Tests
Lumos Labs licenses the online NCPT platform and the cognitive training product Lumosity®
Exploratory: Parkinson's Disease (PD)
|Short Description of Instrument||
Purpose: The Lumos Labs NCPT is a brief, repeatable, web-based platform of cognitive assessments.
Overview: It is intended to measure functioning across several cognitive domains including working memory, visuospatial memory, psychomotor speed, fluid and logical reasoning, response inhibition, numerical calculation, and selective and divided attention. The platform includes 18 subtests that are online adaptations of widely used conventional neuropsychological tests. The NCPT platform is modular, and the subtests can be arranged into customized batteries. The NCPT is optimized for administration in an unsupervised environment on desktop or laptop computers with internet connectivity. Internet connectivity is required for subtest loading and data transmission, but not for the active test taking. For each subtest, users must successfully complete a tutorial and practice session before they are able to move on to the assessment to ensure they understand the task requirements (Morrison et al., 2015).
Lumos Labs also created the Human Cognition Project, which is an online collaborative research platform that grants qualified researchers' free access to cognitive training tasks and research tools, and limited access to data on cognitive task performance.
The completion time varies. Completion of the 8 subtests in the validation study took about 20-30 min (Morrison et al., 2015).
The Lumos Lab NCPT platform is intended to be used as a cognitive assessment tool applicable to a broad population. It may also be used as a screening tool to evaluate potential clinical trial participants with specific cognitive profiles. An example of this potential was demonstrated in a study using the Brain Health Registry (BHR) and associated online platforms to characterize the cognitive profiles of 634 self-reported PD patients (Cholerton et al., 2019):
The BHR is an online registry and database with the overall goal to accelerate the development of effective treatments and preventative interventions for Alzheimer's disease and other brain disorders. The BHR recruits, screens, and longitudinally monitors cognition and function in participants and additionally gathers information from their study partners and caregivers. The website, www.brainhealthregistry.org, was established in early 2014 (Weiner et al., 2018). Participants are directed to online neuropsychological tests administered by Cogstate, Lumos Labs, and MemTrax. The go/no-go, trail making A and B, and forward/reverse memory span subtests of the Lumos Labs NCPT; detection, identification, one-card learning, and one-back subtests of the Cogstate Brief Battery; and one-minute memory screening test of the MemTrax were used in the study. Compared with the self-reported non-PD group (N=11,179), the self-reported PD group showed significantly worse performance on the one-card learning (visual learning and memory), identification (visual attention), detection (processing speed), MemTrax memory (recognition memory and response time), and trail making A and B (spatial working memory and divided attention) tests after correction for multiple comparisons. These results suggest that an online registry may be practical to screen and identify trial-ready adults with self-reported PD, and the online cognitive measures can distinguish them from those without self-reported PD.
|Scoring and Psychometric Properties||
Scoring: Each NCPT subtest is scaled following a percentile rank-based inverse normal transformation applied to a large dataset of raw scores, and normative tables are created. For each task, the empirical percentile for each baseline raw score is then mapped to its corresponding place on the normal distribution with 100 representing the mean and 15 representing one standard deviation (Sternberg et al., 2013).
In the validation study (Morrison et al., 2015), the normative data were collected online from 130,140 individuals aged 13-89 years, who were generally healthy based on self-report. These subjects took 8 subtests of the NCPT battery at least once (arithmetic reasoning, digit-symbol coding, forward/reverse memory span, grammatical reasoning, progressive matrices, trail making A and B). A subset of 35,779 individuals from the normative sample took the NCPT battery for the second time on average 78.8 (range 29-235) days later. This was called the pre-post sample, and their data were used to assess test-retest reliability. The pre-post sample was free to play Lumosity games during the test-retest interval. In a separate study with 73 young healthy individuals aged 21-43 with 16.95 ± 2.0 years of education, the concurrent validity for 5 (digit-symbol coding, forward/reverse memory span, trail making A and B) of the 8 NCPT subtests to their corresponding pencil-paper neuropsychological tests was evaluated. Normative tables were created for each subtest and used to transform the individual raw scores to scaled scores. Performance on the NCPT subtests was sensitive to age, gender, and education. The NCPT subtest scores showed strong correlations with each other (p<0.0001 for all correlations, range 0.14-0.52), but weaker with progressive matrices, yet captured distinct cognitive domains demonstrating good construct validity. The pre-post sample data showed that the test-retest reliability (i.e., Pearson correlations) was significant (p<0.001) for all subtests and ranged from 0.388 to 0.738. The number of games played during the test-retest interval had a small but significant effect on these correlations. Correlations between the 5 NCPT subtests and their paper-pencil counterparts were also significant (p<0.05-0.01, range 0.47-0.71) indicating good concurrent validity.
-Easy to self-administer
-Dynamically generated content creates multiple versions allowing for repeatability
-Concurrent validity estimate of the subtests was based on a group of young and highly educated adults.
-Health status is determined based on self-report
-Unsupervised self-administration of the tests may affect data quality and reliability.
-Psychometric properties have not been validated for use in PD
Morrison GE, Simone CM, Ng NF, Hardy JL. Reliability and validity of the NeuroCognitive Performance Test, a web-based neuropsychological assessment. Front Psychol. 2015 Nov 3;6:1652.
An updated list of all publications can be found on Lumos Labs webpage: https://www.lumosity.com/hcp/research/completed
Cholerton B, Weiner MW, Nosheny RL, Poston KL, Mackin RS, Tian L, Ashford JW, Montine TJ. Cognitive Performance in Parkinson's Disease in the Brain Health Registry. J Alzheimers Dis. 2019;68(3):1029-1038.
Sternberg DA, Ballard K, Hardy JL, Katz B, Doraiswamy PM, Scanlon M. The largest human cognitive performance dataset reveals insights into the effects of lifestyle factors and aging. Front Hum Neurosci. 2013 Jun 20;7:292.
Weiner MW, Nosheny R, Camacho M, Truran-Sacrey D, Mackin RS, Flenniken D, Ulbricht A, Insel P, Finley S, Fockler J, Veitch D. The Brain Health Registry: An internet-based platform for recruitment, assessment, and longitudinal monitoring of participants for neuroscience studies. Alzheimers Dement. 2018 Aug;14(8):1063-1076.
Document last updated August 2022