Disentangling verbal learning performance in multiple sclerosis: Dissociating strategy initiation and maintenance.

Objectives: Performance in verbal learning tests conflates two distinct executive mechanisms: the ability to generate a semantic strategy and the ability to maintain it over time. Deconstructing verbal learning performance may help in the design of personalized cognitive rehabilitation interventions. Methods: We analyzed verbal learning performances of 558 people with MS and 501 healthy controls. The initiation index (first successful semantic cluster) and the cessation index (trial where clustering was permanently abandoned) were calculated. Linear and multinomial logistic regressions adjusted for age, sex and education were used to map learning trajectories. Results: People with MS were less likely to employ semantic clustering (OR = 0.26, 95% CI 0.13-0.51), were significantly more likely to delay strategy initiation (Trial 5 vs. Trial 1: OR = 4.46, 95% CI 2.16-9.20, p < 0.001) and to abandon strategies prematurely (Trial 2 vs. Trial 5: OR = 34.83, 95% CI 4.65-260.74, p < 0.001). People with RRMS often demonstrated successful initiation followed by rapid strategy decay, whereas people with the progressive forms exhibited a primary failure to generate strategies. Higher education provided a protective effect on strategy initiation, supporting the cognitive reserve model. Conclusion: A distinct dissociation emerged, indicating that verbal learning deficits in MS may be driven by a failure of endurance or a failure of acquisition. Cognitive rehabilitation could be enriched with cognitive endurance training, mainly applicable in early disease stages, and compensatory scaffolding is often needed in people with -progressive forms.