Introduction: Music is increasingly used to improve cognition in clinical settings. However, it remains unclear whether its use as a mnemonic strategy is effective in Alzheimer's disease (AD). The present study aimed at determining whether a musical mnemonic might mitigate patients' learning of new verbal information and at exploring the effect of factors such as retention delay and emotional valence of the musical excerpt used. Method: 13 patients with AD and 26 healthy comparisons (HC) with a low musical expertise were included. They learned texts about everyday life themes that were either set to familiar instrumental music, which was positively- or negatively-valenced, or spoken only. Immediate and delayed recalls (after 10 min and 24 hours) were measured. Results: Main results showed that (i) HC demonstrated better verbal episodic memory performance than participants with AD; (ii) participants with AD encoded texts paired with positively-valenced music better than texts paired with negatively-valenced music; (iii) participants with AD recalled sung texts better than spoken texts (after 10 min and 24 hours), regardless of musical valence while HC displayed better recall for texts paired with positively-valenced music. Conclusions: Musical mnemonics may help people with AD learn verbal information that relates to their daily life, regardless the musical expertise of the patients. This result gives promising clinical insights showing that music processing is robust to brain damage in AD. Possible hypotheses explaining the effectiveness of musical mnemonics in AD regardless the musical valence are discussed (e.g., different processing between musical and spoken conditions; disappearance of the positivity bias and implications with respect to the underlying socio-emotional selectivity theory).
Keywords: Mnemonics; positivity bias; recall; socio-emotional selectivity theory; verbal learning.