Effects of music listening on the relationship between P300 and anxiety: an auditory oddball paradigm study

Anxiety can impair the efficiency of attentional and cognitive processes, also affecting brain responses related to stimulus processing, such as the P300 component of Event-Related Potentials (ERPs). This component reflects mechanisms of selective attention and working memory updating, which are often altered in individuals with high levels of trait anxiety. Music, through its effects on the limbic system and on brain areas involved in emotional regulation, may act as a tool capable of modulating emotional states and, consequently, cognitive functioning. This study investigates how music listening may influence the P300 component in relation to individual differences in trait anxiety. Twelve healthy participants, with varying anxiety levels assessed using the State-Trait Anxiety Inventory (STAI), took part in an experiment in which has been evaluated the elecetronencephalographic activity. One participant was excluded from statistical analyses due to poor signal quality. The P300 was analyzed through an auditory oddball paradigm, administered both before and after listening to ten music tracks differing in rhythm and emotional content. Each session involved the random presentation of frequent (n = 160) and deviant (n = 40) tones, played respectively at 1000 Hz and 500 Hz, to which participants responded by pressing two different keys on a keyboard. EEG signals were preprocessed through high-pass filtering at 0.1 Hz and low-pass filtering at 20 Hz using a 4th-order Butterworth filter, interpolation of noisy channels, artifact removal via Independent Component Analysis (ICA), and rereferencing to the average potential across electrodes. Epochs were extracted from -200 to 600 ms relative to stimulus onset, and average event-related potentials were computed for each condition. The P300 peak was identified as the maximum positive value of responses at Cz and Pz electrodes. Participants were divided into high- and low-anxiety groups based on the 50th percentile of STAI-T scores, considering only trait anxiety. Before music listening, high-anxiety individuals showed significantly reduced P300 amplitude at Cz and shorter latency at Pz compared to low-anxiety individuals. After music listening, these group differences were no longer statistically significant, suggesting a potential normalization effect. Pearson correlations revealed a significant negative association between trait anxiety and P300 amplitude at Cz in both conditions, while the same pattern was found between trait anxiety and latency at Pz only before listening. Although no significant group-level effect of music on amplitude was found, a general trend toward increased P300 amplitude, mainly at Pz, emerged post-music. Spearman correlations performed on post–pre differences revealed significant positive associations at both Cz and Pz electrodes: higher trait anxiety was associated with widespread increased P300 latency after music, while individuals with low anxiety exhibited decreased latency. This indicates a differential modulation of cognitive processing speed depending on anxiety levels. These findings support the presence of subtle cognitive impairments linked to trait anxiety even in healthy individuals, particularly reflected in reduced P300 amplitude, and highlight the role of music as a potential neuromodulatory tool for attentional and emotional regulation.