Clinical usability of the Story-Based Empathy Task (SET) in non-demented ALS patients

Background: This study aimed at assessing the clinical usability of the Story-Based Empathy Task (SET) in non-demented amyotrophic lateral sclerosis (ALS) patients. Methods: N = 106 non-demented ALS patients and N = 101 healthy controls (HCs) were administered the SET, which includes three subtests assessing Emotion Attribution (SET-EA), Intention Attribution (SET-IA) and causal inference (SET-CI) — the latter being a control task. Patients also underwent the Reading the Mind in the Eyes Test (RMET), the Edinburgh Cognitive and Behavioural ALS Screen (ECAS) and a thorough behavioural and motor-functional evaluation. The diagnostics of the SET-EA and -IA were tested against a defective performance on the RMET. The association between SET subtests and cognitive/behavioural outcomes was examined net of demographic and motor-functional confounders. Case-control discrimination was explored for each SET subtest. Results: Demographically adjusted SET-EA and -IA scores accurately detected defective RMET performances at the optimal cutoffs of <3.04 (AUC =.84) and <3.61 (AUC =.88), respectively. By contrast, the SET-CI performed poorly in doing so (AUC =.58). The SET-EA converged with the RMET, as well as with ECAS-Executive and -Memory scores, whilst the SET-IA was unrelated to cognitive measures (including the RMET); the SET-CI was related to the ECAS-Language the ECAS-Executive. SET subscores were unrelated to behavioural outcomes. Only the SET-EA discriminated patients from HCs. Conclusions: The SET as a whole should not be addressed as a social-cognitive measure in this population. At variance, its subtest tapping on emotional processing — i.e., the SET-EA — is recommended for use as an estimate of social-cognitive abilities in non-demented ALS patients.