RSRobert J. Spencer

Robert J. Spencer

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Robert J. Spencer has a total of 40 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • The Montreal Cognitive Assessment: Norms and Reliable Change Indices for Standard and MoCA-22 Administrations

    by Lauren N Ratcliffe, Andrew C Hale, Taylor McDonald, Kelsey C Hewitt, Christopher M Nguyen, Robert J Spencer, David W Loring

    Objective: The Montreal Cognitive Assessment (MoCA) is among the most frequently administered cognitive screening tests, yet demographically diverse normative data are needed for repeated administrations. Method: Data were obtained from 18,410 participants using the National Alzheimer’s Coordinating Center Uniform Data Set. We developed regression-based norms using Tobit regression to account for ceiling effects, explored test–retest reliability of total scores and by domain stratified by age and diagnosis with Cronbach’s alpha, and reported the cumulative change frequencies for individuals with serial MoCA administrations to gage expected change. Results: Strong ceiling effects and negative skew were observed at the total score, domain, and item levels for the cognitively normal group, and performances became more normally distributed as the degree of cognitive impairment increased. In regression models, years of education was associated with higher MoCA scores, whereas older age, male sex, Black and American Indian or Alaska Native race, and Hispanic ethnicity were associated with lower predicted scores. Temporal stability was adequate and good at the total score level for the cognitively normal and cognitive disorders groups, respectively, but fell short of reliability standards at the domain level. Conclusions: MoCA total scores are adequately reproducible among those with cognitive diagnoses, but domain scores are unstable. Robust regression-based norms should be used to adjust for demographic performance differences, and the limited reliability, along with the ceiling effects and negative skew, should be considered when interpreting MoCA scores.


  • A quantitative review of competing learning slope metrics: effects of age, sex, and clinical diagnosis

    by Robert J. Spencer, Trevor F. Williams, Victoria M. Kordovski, Sarah D. Patrick, Ketrin Lengu, Brian D. Gradwohl, Dustin B. Hammers

    IntroductionIn learning and memory tests that involve multiple presentations of the same material, learning slope refers to the degree to which examinees improve performances over successive learning trials. We aimed to quantitatively review the traditional raw learning slope (RLS), and the newly created learning ratio (LR) to understand the effects of demographic variables and clinical diagnoses on learning slope (e.g., limited improvement over multiple trials), and to develop demographically sensitive norms.MethodA systematic literature search was conducted to evaluate the potential for these aims to be examined across the most popular contemporary multi-trial learning tests. Two databases were searched. Following this, hierarchical linear modeling was used to examine how demographic variables predict learning slope indices. These results were in turn used to contrast the performance of clinical groups with the predicted performance of demographically similar healthy controls. Finally, preliminary normative estimates for learning slope indices were presented.ResultsA total of 82 studies met criteria for inclusion in this study. However, the Rey Auditory Verbal Learning Test (RAVLT) was the only test to have sufficient trial-level learning and demographic data. Fifty-eight samples from 19 studies were quantitatively examined. Hierarchical linear models provided evidence of sex differences and a curvilinear decline in learning slope with age, with strongest and most consistent effects for LR relative to RLS. Regression-based norms for demographically corrected RLS and LR scores for the RAVLT are presented. The effect of clinical diagnoses was consistently stronger for LR, and Alzheimer's disease had the strongest effect, followed by invalid performances, severe traumatic brain injury, and seizures/epilepsy.ConclusionOverall, LR enjoys both conceptual and demonstrated psychometric advantages over RLS. Replication of these findings can be completed by reanalyzing existing datasets. Further work may focus on the utility of using LR in diagnosis and prediction of clinical prognosis.


  • Reliability and Factor Structure of the Saint Louis University Mental Status (SLUMS) Examination

    by Emily T. Noyes, Saudia Major, Addie M. Wilson, Elizabeth B. Campbell, Lauren N. Ratcliffe, Robert J. Spencer

    Compared to its alternatives (e.g., Mini-Mental State Examination [MMSE] and Montreal Cognitive Assessment [MoCA]), little is known about the psychometric properties and factor structure of the Sai...


  • Systematic Review of the Psychometric Properties of the Saint Louis University Mental Status (SLUMS) Examination

    by Robert J. Spencer, Emily T. Noyes, Jessica L. Bair, Michael T. Ransom

    This systemic review qualitatively synthesizes existing psychometric support for the Saint Louis University Mental Status (SLUMS) Examination, a cognitive screening measure which presents as a free...


  • Using supplemental memory measures to refine interpretation of the repeatable battery for the assessment of neuropsychological status

    by Brian D. Gradwohl, Ryan W. Mangum, Emily T. Noyes, Robert J. Spencer

    The Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) is frequently used as a cognitive screening measure or as part of a comprehensive neuropsychological battery. Augmenting the current memory subtests of the RBANS to reflect clinically meaningful distinctions in memory performance may improve its clinical utility, allowing users to generate additional hypotheses and refine clinical interpretations. We pilot four supplementary memory measures to use with the RBANS, adapted from paradigms commonly used by other neuropsychological tests. We also provide several new theoretically derived memory indices to supplement the standard Delayed Memory Index. The new subtests correlated significantly with the standard memory measures, and the new indices demonstrated good reliability and diagnostic accuracy. This study provides preliminary support for supplementing the RBANS to allow for more nuanced interpretations of memory performance.


  • The Savage Organizational Scoring System for the Rey-Osterrieth Complex Figure Test: Norms and New Data From Veterans in a Traumatic Brain Injury Clinic

    by Ryan W. Mangum, Kathryn A. Tolle-Fu, Andrew C. Hale, Brian D. Gradwohl, Robert J. Spencer

    The Rey-Osterrieth Complex Figure Test (RCFT) permits quantifying diverse cognitive abilities, including executive function (EF). We evaluated the psychometric properties of a scoring procedure for the RCFT, the Savage Organizational Scoring System (SOSS), that awards points for drawing the largest structural elements of the figures as continuous wholes. This was a two-phase study: first, we conducted a systematic literature search for studies using the SOSS, and aggregated previously published data for healthy controls to create a normative database; second, we observed performances from veterans evaluated for traumatic brain injury (TBI), examining the reliability of their SOSS scores, the SOSS correlations with two EF measures and the participants’ self-reported cognitive functioning, and then compared their mean scores to normative expectations. Across our literature-derived normative database, the aggregated mean SOSS score was 4.12 (SD = 1.72), which was marginally higher than that of our veteran participants evaluated for TBI, 3.72 (SD = 1.79). The SOSS had modest internal consistency (α = .59). Unlike the criterion EF measures, the SOSS was not significantly related to self-reported cognitive functioning. The SOSS shared a small, significant correlation with Trails B and Shipley Abstraction; but RCFT Copy scores were more strongly related to these tests, and the SOSS added no significant incremental predictive value beyond the RCFT Copy score. However, SOSS scores did predict RCFT Recall beyond RCFT Copy scores. We conclude that the SOSS has modest reliability and is predictive of RCFT Recall scores, but it is not strongly correlated with other EF measures, and it is only minimally affected by mild TBI.


  • Open Access

    Demographically-corrected normative data for the RBANS learning ratio in a sample of older adults

    by Dustin B. Hammers, Kevin Duff, Robert J. Spencer

    A novel learning slope score – the Learning Ratio (LR) – has recently been developed that appears to be sensitive to memory performance and AD pathology more optimally than traditional learning slo...


  • Open Access

    Relationship between a novel learning slope metric and Alzheimer’s disease biomarkers

    by Dustin B. Hammers, Kayla Suhrie, Ava Dixon, Brian D. Gradwohl, Zane G. Archibald, Jace B. King, Robert J. Spencer, Kevin Duff, John M. Hoffman

    The Learning Ratio (LR) is a novel learning score examining the proportion of information learned over successive learning trials relative to information available to be learned. Validation is warranted to understand LR’s sensitivity to Alzheimer’s disease (AD) pathology. One-hundred twenty-three participants across the AD continuum underwent memory assessment, quantitative brain imaging, and genetic analysis. LR scores were calculated from the HVLT-R, BVMT-R, RBANS List Learning, and RBANS Story Memory, and compared to total hippocampal volumes,18F-Flutemetamol composite SUVR uptake, and APOE ε4 status. Lower LR scores were consistently associated with smaller total hippocampal volumes, greater cerebral β-amyloid deposition, and APOE ε4 positivity. This LR score outperformed a traditional learning slope calculation in all analyses. LR is sensitive to AD pathology along the AD continuum – more so than a traditional raw learning score – and reducing the competition between the first trial and subsequent trials can better depict learning capacity.


  • Open Access

    Validation of HVLT-R, BVMT-R, and RBANS Learning Slope Scores along the Alzheimer’s Continuum

    by Dustin B Hammers, Kayla Suhrie, Ava Dixon, Brian D Gradwohl, Kevin Duff, Robert J Spencer

    Background: The learning ratio (LR) is a novel learning slope score that has been developed to reduce the inherent competition between the first trial and subsequent trials in traditional learning slopes. In essence, the LR is the number of items learned after the first trial divided by the number of items yet to be learned. Criterion and convergent validation of this LR score is warranted to understand its sensitivity along the Alzheimer’s disease (AD) continuum. Method: The LR metric was calculated for 123 participants from standard measures of memory, including the Hopkins Verbal Learning Test—Revised, Brief Visuospatial Memory Test—Revised, Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) list learning, and RBANS story memory. All participants were categorized as normal cognition, mild cognitive impairment (MCI), or AD. LR performances were compared between groups, among other standard memory measures, and with regards to how well they discriminated cognitively impaired from unimpaired samples—and within diagnostic subgroups. Results: Lower LR scores were observed for the MCI and AD groups than the normal cognition group, with the AD group performing worse than the MCI group for several slope calculations. Lower LR scores were also consistently associated with poorer performances on traditional memory measures. LR scores further displayed excellent receiver operator characteristics when differentiating those with and without cognitive impairment—and MCI from normal cognition. Overall, LR scores consistently outperformed traditional learning slope calculations across all analyses. Conclusions: This LR score is sensitive to memory dysfunction along the AD continuum, and results offer criterion and convergent validity for use of the LR metric to understand learning capacity.


  • Teleneuropsychology practice survey during COVID-19 within the United States

    by Amber D. Rochette, Annalise Rahman-Filipiak, Robert J. Spencer, David Marshall, Julija E. Stelmokas

    Identify factors influencing service delivery changes during COVID-19 and examine barriers and provider satisfaction with teleneuropsychology (teleNP). Licensed clinical neuropsychologists within t...