The Role of Entropy in Construct Specification Equations (CSE) to Improve the Validity of Memory Tests: Extension to Word Lists

Abstract

Metrological methods for word learning list tests can be developed with an information theoretical approach extending earlier simple syntax studies. A classic Brillouin entropy expression is applied to the analysis of the Rey’s Auditory Verbal Learning Test RAVLT (immediate recall), where more ordered tasks—with less entropy—are easier to perform. The findings from three case studies are described, including 225 assessments of the NeuroMET2 cohort of persons spanning a cognitive spectrum from healthy older adults to patients with dementia. In the first study, ordinality in the raw scores is compensated for, and item and person attributes are separated with the Rasch model. In the second, the RAVLT IR task difficulty, including serial position effects (SPE), particularly Primacy and Recency, is adequately explained (Pearson’s correlation R=0.80) with construct specification equations (CSE). The third study suggests multidimensionality is introduced by SPE, as revealed through goodness-of-fit statistics of the Rasch analyses. Loading factors common to two kinds of principal component analyses (PCA) for CSE formulation and goodness-of-fit logistic regressions are identified. More consistent ways of defining and analysing memory task difficulties, including SPE, can maintain the unique metrological properties of the Rasch model and improve the estimates and understanding of a person’s memory abilities on the path towards better-targeted and more fit-for-purpose diagnostics.

Traceability and comparability through crosswalks with the NeuroMET Memory Metric

“…For the total NMM only three items showed significant DIF with regards to sub-groups but for the individual test RAVLT as many as five of fifteen items showed DIF with regards to sub-groups. This issue has, however, been studied in detail in one of our previous papers 25 . We concluded that this may be due to dimensionality issues but refrained from separating groups of items into subtests (for RAVLT for primacy and recency items) in an attempt to improve the unidimensionality as the measurement uncertainties become too large with only a handful of dichotomous items.…”

“…However, only one item showed a significant x 2 , and removing the items did not significantly affect the model fit. In separate papers we report further on the issue of the so-called serial position effect that occurs in the word learning list tests for the items in the beginning (primacy) and the end (recency) of the lists 25,36,37 .…”

“…The set of items also had to have associated construct specification Eqs. 11,25 , which provide an comprehensive empirical understanding of how the collection of items works together; what is being measured; and how validity is ensured. Short-term memory items from the CBT forward sequence (n = 14), DST forward sequence (n = 12), the RAVLT first trial A-list (n = 15), and the WLL CERAD first trial (n = 10) were included.…”

Application of the Rasch measurement model in rehabilitation research and practice: early developments, current practice, and future challenges

“…It is an empirical matter if the traceability extends to other frames of reference. Elsewhere, one study identified the role of a construct specification equation as a recipe for producing “certified reference materials” for calibration of both the task difficulty and a person's ability ( 139 ).…”

Medical Applications of Nonadditive Entropies

“…The concept of entropy has more than once been useful in connection with medical applications (see, for instance, [ 7 , 8 , 9 ]). In particular, nonadditive entropies have been extensively used in image and signal processing in order to improve speed and clarity.…”