Cluster Analysis in Neuropsychological Research

Allen, Daniel N.; Goldstein, Gérald

doi:10.1007/978-1-4614-6744-1

Cited by 32 publications

(7 citation statements)

References 150 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The direction of the z-score was adapted to facilitate interpretation, with a positive value reflecting a higher performance. Outliers with a z-score beyond 3 standard deviations were curtailed to a value of ± 3 (e.g., Renwick et al, 2015 ), as clusters are sensitive to extreme values ( Milligan and Cooper, 1987 ; Allen and Goldstein, 2013 ).…”

Section: Methodsmentioning

confidence: 99%

“…As variables must be selected based on the problem of interest ( Bratchell, 1989 ), three variables assessing verbal skills that are not part of the main cognitive areas of interest of this study (i.e., verbal short-term memory, similarities, and vocabulary; Wechsler, 2016 ) were kept aside and used to assess the external validity of the clustering solution. Variables containing insufficient information to contribute to the clustering solution ( Milligan and Cooper, 1987 ) or bringing redundant information ( Allen and Goldstein, 2013 ) were identified through Pearson’s correlations between the 18 variables of interest. Three variables were excluded from clustering analysis: motor inhibition had low correlation coefficient with the rest of the variables (i.e., insufficient information; r < 0.30), while copying figures and attentional vigilance were highly correlated with others (i.e., redundant information; r > 0.70).…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Cognitive, perceptual, and motor profiles of school-aged children with developmental coordination disorder

et al. 2022

View full text Add to dashboard Cite

Developmental coordination disorder (DCD) is a heterogeneous condition. Besides motor impairments, children with DCD often exhibit poor visual perceptual skills and executive functions. This study aimed to characterize the motor, perceptual, and cognitive profiles of children with DCD at the group level and in terms of subtypes. A total of 50 children with DCD and 31 typically developing (TD) peers (7–11 years old) underwent a comprehensive neuropsychological (15 tests) and motor (three subscales of the Movement Assessment Battery for Children-2) assessment. The percentage of children with DCD showing impairments in each measurement was first described. Hierarchical agglomerative and K-means iterative partitioning clustering analyses were then performed to distinguish the subtypes present among the complete sample of children (DCD and TD) in a data-driven way. Moderate to large percentages of children with DCD showed impaired executive functions (92%) and praxis (meaningless gestures and postures, 68%), as well as attentional (52%), visual perceptual (46%), and visuomotor (36%) skills. Clustering analyses identified five subtypes, four of them mainly consisting of children with DCD and one of TD children. These subtypes were characterized by: (i) generalized impairments (8 children with DCD), (ii) impaired manual dexterity, poor balance (static/dynamic), planning, and alertness (15 DCD and 1 TD child), (iii) impaired manual dexterity, cognitive inhibition, and poor visual perception (11 children with DCD), (iv) impaired manual dexterity and cognitive inhibition (15 DCD and 5 TD children), and (v) no impairment (25 TD and 1 child with DCD). Besides subtle differences, the motor and praxis measures did not enable to discriminate between the four subtypes of children with DCD. The subtypes were, however, characterized by distinct perceptual or cognitive impairments. These results highlight the importance of assessing exhaustively the perceptual and cognitive skills of children with DCD.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Cognitive, perceptual, and motor profiles of school-aged children with developmental coordination disorder

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Ward’s method of cluster analysis was utilized as it is consistent with past methodology of cluster analysis studies in this area (Allen, Leany, et al, 2010; Mottram & Donders, 2006; Ringdahl et al, 2019) and as it has a number of advantages over other methods, including that it provides consistent results with other agglomerative clustering methods and is less affected by outliers (Cross, 2013; Morris et al, 1981). Squared Euclidian distance was utilized as the dissimilarity measure as it is sensitive to pattern and level of performance differences and is consistent with previous TBI research (Allen, Leany, et al, 2010; Allen & Goldstein, 2013). In order to identify the most appropriate number of clusters at each time point, stability, parsimony, and internal and external validity of the cluster solutions were examined (Aldenderfer & Blashfield, 1984; Cross, 2013).…”

Section: Methodsmentioning

confidence: 99%

Cognitive profiles following sport-related concussion in high school athletes.

Maietta¹,

Kuwabara²,

Keene³

et al. 2022

Neuropsychology

View full text Add to dashboard Cite

Objective: Heterogeneity is common following traumatic brain injury (TBI) with important implications for clinical outcome. Research in moderate to severe TBI identifies differentiated cognitive profiles. There is little research investigating potential clusters following sport concussion. Methods: Cluster analysis was used to determine patterns of Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT) performance at three time points (72 hr, 4-7 days, and 8-30 days postconcussion). Participants included 1,817 high-school athletes (M age = 15.5; 37.9% female) who completed ImPACT following concussion. Results: Separate cluster analyses of cognitive scores were conducted at each postconcussion timepoint. Results indicated three clusters of cognitive performance within 1 week postinjury, while a two-cluster solution was identified at longer intervals (8-30 days). Clusters differed primarily by level of performance and also exhibited stratified differences in symptom severity and magnitude of change in cognitive function and symptom-reporting from pre-to postconcussion. The lowest performing cluster in each PC group was Mildly-to-Moderately Impaired and exhibited the highest rates of complicated recovery, suggesting an association between the current clusters and protracted recovery. Conclusions: Results provide preliminary evidence that ImPACT can identify clusters of athletes based on cognitive performance postconcussion that differ in clinically meaningful ways, including symptoms, magnitude of change from baseline, and concussion recovery outcomes. Discriminant functions can classify athletes into clusters based on postconcussion scores and a supplemental excel calculator is provided for such purposes. Absence of full demographic data regarding race/ethnicity and socioeconomic status may limit generalizability. Future work should investigate whether these clusters may assist in return-to-play decision-making by identifying at-risk athletes who may benefit from targeted intervention. Key PointsQuestion: Can unique clusters of athletes be identified at different time points after concussion based on cognitive performance?. Findings: Two to three clusters of athletes were identified at varying postconcussion assessment time points that differed in severity of cognitive deficits, symptomreporting, magnitude of change from pre-to postconcussion, and recovery outcomes. Importance: Further identifying and understanding the heterogeneity in postconcussion cognitive and symptom profiles can assist with specifying clusters of athletes who demonstrate more severe deficits and reported symptoms following concussive injury. Next Steps: Future work should investigate whether the clusters identified here can be useful in identifying at-risk athletes for targeted treatment following concussion.

show abstract

“…The heterogeneity of ABI is extensively considered as one of the most significant barriers to finding effective therapeutic interventions (9). Identifying subgroups of patients who have distinguishable cognitive profiles that, in turn can assist in treatment planning and patient care, is crucial.…”

Section: Introductionmentioning

confidence: 99%

Neuropsychological Assessments of Patients With Acquired Brain Injury: A Cluster Analysis Approach to Address Heterogeneity in Web-Based Cognitive Rehabilitation

et al. 2021

View full text Add to dashboard Cite

We aimed to (1) apply cluster analysis techniques to mixed-type data (numerical and categorical) from baseline neuropsychological standard and widely used assessments of patients with acquired brain injury (ABI) (2) apply state-of-the-art cluster validity indexes (CVI) to assess their internal validity (3) study their external validity considering relevant aspects of ABI rehabilitation such as functional independence measure (FIM) in activities of daily life assessment (4) characterize the identified profiles by using demographic and clinically relevant variables and (5) extend the external validation of the obtained clusters to all cognitive rehabilitation tasks executed by the participants in a web-based cognitive rehabilitation platform (GNPT). We analyzed 1,107 patients with ABI, 58.1% traumatic brain injury (TBI), 21.8% stroke and 20.1% other ABIs (e.g., brain tumors, anoxia, infections) that have undergone inpatient GNPT cognitive rehabilitation from September 2008 to January 2021. We applied the k-prototypes algorithm from the clustMixType R package. We optimized seven CVIs and applied bootstrap resampling to assess clusters stability (fpc R package). Clusters' post hoc comparisons were performed using the Wilcoxon ranked test, paired t-test or Chi-square test when appropriate. We identified a three-clusters optimal solution, with strong stability (>0.85) and structure (e.g., Silhouette > 0.60, Gamma > 0.83), characterized by distinctive level of performance in all neuropsychological tests, demographics, FIM, response to GNPT tasks and tests normative data (e.g., the 3 min cut-off in Trail Making Test-B). Cluster 1 was characterized by severe cognitive impairment (N = 254, 22.9%) the mean age was 47 years, 68.5% patients with TBI and 22% with stroke. Cluster 2 was characterized by mild cognitive impairment (N = 376, 33.9%) mean age 54 years, 53.5% patients with stroke and 27% other ABI. Cluster 3, moderate cognitive impairment (N = 477, 43.2%) mean age 33 years, 83% patients with TBI and 14% other ABI. Post hoc analysis on cognitive FIM supported a significant higher performance of Cluster 2 vs. Cluster 3 (p < 0.001), Cluster 2 vs. Cluster 1 (p < 0.001) and Cluster 3 vs. Cluster 1 (p < 0.001). All patients executed 286,798 GNPT tasks, with performance significantly higher in Cluster 2 and 3 vs. Cluster 1 (p < 0.001).

show abstract

Cluster Analysis in Neuropsychological Research

Cited by 32 publications

References 150 publications

Cognitive, perceptual, and motor profiles of school-aged children with developmental coordination disorder

Cognitive, perceptual, and motor profiles of school-aged children with developmental coordination disorder

Cognitive profiles following sport-related concussion in high school athletes.

Neuropsychological Assessments of Patients With Acquired Brain Injury: A Cluster Analysis Approach to Address Heterogeneity in Web-Based Cognitive Rehabilitation

Contact Info

Product

Resources

About