Enhanced clinical evaluation of Alzheimer disease through rapid automatized naming of emotional pictures

Abstract

Purpose: Naming tests are commonly incorporated as short versions into clinical evaluation of Alzheimer disease (AD), but evidence suggests they may overestimate semantic abilities. This study investigated whether a more challenging naming task could distinguish cognitively healthy individuals from those with amnestic mild cognitive impairment (aMCI) and mild AD, while more effectively capturing lexical-semantic deficits. Method: In this cross-sectionally study, 15 aMCI patients, 15 mild AD patients, and 20 cognitively healthy controls completed a rapid automatized naming (RAN) task using emotionally balanced object and action pictures. Regression analyses and receiver operating characteristic (ROC) curves were used to evaluate diagnostic accuracy and identify influencing factors. Results: Controls significantly outperformed aMCI and AD groups in response times and naming accuracy (all p-values <0.001). Negative pictures elicited slower responses and higher error rates in controls. While the “negative delay” effect was attenuated in patient groups, negative stimuli remained cognitively demanding, as indicated by increased errors. Emotional valence enhanced naming performance for action pictures in the control group, a pattern that diminished or was absent in the patient groups. Regression analyses showed significant associations between response times/error rates and group membership (both p-values <0.001), independent of age, sex, education, anxiety, and depression. ROC analysis revealed moderate diagnostic accuracy for classifying mild AD (AUC of 0.907 for response times; AUC of 0.861 for error rates), but not for aMCI. Conclusion: The RAN task effectively identifies early stages of AD and naming impairments across different stages of cognitive decline. The inhibitory effect of negative valence on lexical retrieval underscores the importance of incorporating emotional valence into naming assessments by further considering emotional reactivity (arousal) and different word classes.