Abstract

Background: The auditory verbal learning test (AVLT) procedure is routinely deployed in neuropsychological investigations to examine learning and memory status in research and clinical cohorts. Concerns however have been raised regarding the susceptibility of the standard AVLT procedure to ceiling effects, which may have adverse consequences for psychometric properties and result in an underestimation of true potential and differences between normal and abnormal scores.
Methods: We examined the performance of patients with temporal lobe epilepsy (TLE; n = 40) who had completed a standard 15-item AVLT and compared a group of TLE patients (n = 12) with healthy controls (n = 12) who completed an extended 24-item AVLT, which was designed to minimise the probability of ceiling scores.
Results: Ceiling effects on at least one trial (≥ 14) was achieved by 33% of patients on the 15-item test, with 60% of patients scoring within or above the average list learning total score. Increasing the list length to 24-items reduced the percentage of TLE patients scoring within the normal range to 42%. In addition, no patients but 25% of control participants achieved a maximum score on trial A5. The performance of controls was superior to patients for the best learning trial, learning rate and total learning score.
Conclusions: Increasing the list length to 24-items eliminated ceiling scores in all TLE patients and most controls and allowed the true magnitude in difference between the groups to be observed. These findings have implications for decisions relating to optimal AVLT list lengths that might be deployed for memory assessment in TLE.