P300 Responses to Speech and Music Stimuli: Effect of Musical Training

Music is the art of combining different sounds to express ideas and emotions through elements such as pitch, timbre, and rhythm. The processing apparatus required for music involves sound encoding at higher cognitive levels, including skills such as memory, sequencing, and learning.1 The neurological processes responsible for music perception and production are also utilized in non-musical endeavours, such as auditory discrimination and temporal processing. Specializations associated with music perception are also seen in areas of the brain that are more popularly and exclusively known for language processing.2 Experienced musicians can discern and identify the fine-grained acoustics of musical elements such as pitch, timbre, and rhythm,3 as well as individual notes of a melody as compared with non-musicians. Musicians have a symmetrical distribution in both hemispheres for music processing whereas non-musicians have a clear right-hemispheric dominance.4 The inter-hemispheric neural connectivity is found to be greater in musicians due to the neuroplasticity-induced changes produced through musical training. Musical experience and training produces changes in the auditory processing of linguistic and non-linguistic sounds (cross-domain plasticity), both at the cortical and subcortical levels.5 Music training enhances higher cognitive skills which ultimately help to improve speech and language processing.1 Interestingly, differences in cortical auditory processing can be seen as early as one-year post commencement of training.6 Apart from the structural changes, different patterns of neural activation have been observed in musicians, such as heightened responses to simple artificial tones and stronger reactions to the stimuli resembling sounds of their musical instruments.6 The responses to speech and music stimuli are extremely dependent on auditory learning and exposure, specifically on the amount of practice, age at which training began, nature of training, practice methods, and behavioural relevance of the stimuli.3 Electrophysiological testing procedures form a fundamental part in the understanding of -auditory function and neuroplastic changes induced by the process produced by learning.7Table 1: Protocol Used for Recording of Cortical Auditory Evoked Potentials.Figure 1: Clustered bar graph representing the amplitude (in µV) obtained in the right ear for both the groups using speech stimuli.Figure 2: Clustered bar graph representing the amplitude (in µV) obtained in the left ear for both the groups using speech stimuli.Figure 3: Clustered bar graph representing the latency (in millisecond) obtained in the right ear for both the groups using speech stimuli.Figure 4: Clustered bar graph representing the latency (in millisecond) obtained in the left ear for both the groups using speech stimuli.Figure 5: Clustered bar graph representing the amplitude (in µV) obtained in the right ear for both the groups using non-speech stimuli.Figure 6: Clustered bar graph representing the amplitude (in µV) obtained in the left ear for both the groups using non-speech stimuli.Figure 7: Clustered bar graph representing the latency (in millisecond) obtained in the right ear for both the groups using non-speech stimuli.Figure 8: Clustered bar graph representing the latency (in millisecond) obtained in the left ear for both the groups using non-speech stimuli.A P300 response is an endogenous evoked cortical potential, or an extended auditory late response (ALR) component recorded within its time frame under special stimulus conditions. The response to the infrequent stimuli of P300, known also as the P3 wave, is seen as a positive deflection in the latency region of 300 ms8 and is highly dependent on the subject's attention to the auditory stimulus.9 Researchers have shown that superior information processing is directly correlated with shorter latencies and larger amplitudes of P300 which receives contributions from the thalamus, auditory cortex (posterior superior temporal plane), temporal parietal cortical areas, and portions of the frontal lobe.10 It is employed in the current study to investigate the effects of musical training on cortical auditory processing. This study reports the advantages of musical training in not just auditory perception but also the cognitive processing of different auditory stimuli and hemispheric specialization of the same. This study is directed towards investigation of the contribution of attention and cognitive mechanisms cultivated due to long-term musical training that are employed during a P300 response paradigm to obtain electrophysiological indications of auditory plasticity effects of musical training. METHODS Twenty adults between the ages of 18-30 years with normal hearing sensitivity were recruited for the study using purposive sampling. Informed consent was obtained from all participants before testing in accordance with the ethical rules followed at Bharati Vidyapeeth School of Audiology and Speech-Language Pathology. Participants were divided into two age-and-gender-matched groups: musicians (n=11) with a mean age of 23.09 years and non-musicians (n=9) with a mean age of 22.67 years. Pure tone audiometric thresholds for octave frequencies between 0.25k Hz to 4k Hz were <25 dBHL for air conduction and bone conduction for all participants. The group of musicians reported having had professional musical training for a minimum of three years with five to 20 hours of practice per week. The non-musicians reported having had no professional training in music across their lifetime and did not play any musical instrument. Two types of stimuli, speech sounds and non-speech musical chords, were employed for the recording of cortical auditory evoked potentials (CAEP) responses. Speech stimuli used were consonant-vowel pairs /da/ and /ba/ (duration=90 ms) recorded in a male voice in a sound-treated room using Adobe Audition software. In the P300 paradigm, /da/ was used as the standard or frequent stimulus and /ba/ as the rare stimulus. Non-speech stimuli used were E major (frequent stimulus) and F major (infrequent stimulus) chords recorded on a piano (duration=200 ms). Major chords have a distinctly different tone eliciting differential cortical activation in comparison with minor chords.11 Hence, closely spaced major chords commonly occurring in contemporary music were chosen for the study. The sound files were processed to equalize loudness, remove unwanted noise, and to ensure equivalent amplitude for all four stimuli. CAEPs P1, N1, P2, N2, and P300 were recorded in a quiet, electrically shielded room using Bio-Logic Navigator Pro AEP system via insert earphones. Gold-plated disc electrodes were used for response acquisition. The participants were seated in a comfortable position in a well-lit room and instructed to count the number of infrequent stimuli. The subjects were told to fixate their gaze or watch a silent video to control ocular artifacts. Multiple traces of 80-120 sweeps for frequent stimuli were recorded and the grand average waveforms were analysed for latency and amplitude. Table 1 delineates the protocol used for the recording. RESULTS The study was conducted on a total of 20 participants divided into two groups, musicians and non-musicians, containing 11 and nine individuals respectively. The statistical analysis was done using SPSS Version 20. The Shapiro-Wilk test of normality revealed that the data followed a normal distribution. Hence, parametric tests were used for further analyses. Results of the Independent sample t-test revealed that the latencies and amplitudes of P1, N1, and N2 in both ears did not differ significantly (P<0.05) for speech and chords between both groups. Musicians, however, were observed to have a significantly greater P2 amplitude in the left ears (P=0.046) as compared to non-musicians. However, no statistically significant difference was observed for the P2 obtained from right ears across the two groups. A statistically significant difference was seen in the amplitude of P300 between groups for both stimuli. Using speech as the stimulus, P300 amplitude showed a marked difference in both the right and left ear, between the two groups, with the difference being highly significant in the left ear (Right ear P=0.029 and Left ear P=0.001). Figures 1 and 2 show clustered bar graphs representing the median amplitude, while Figures 3 and 4 represent median latency of responses obtained in right ear and left ear, respectively, for both groups using speech stimuli. Using musical chords as non-speech stimuli, a highly significant difference was observed in the P300 amplitude for both ears (Right: P=0.004 and Left: P=0.001) as well as in its latency in the left ear (P=0.031) using t-test. Figures 5 and 6 show clustered bar graphs representing the median amplitude, while Figures 7 and 8 represent median latency of responses obtained in right ear and left ear, respectively for both groups using non-speech stimuli. In our study, we observed that the musician group displayed shorter mean latencies of P300 as compared to the non-musician group for both stimuli. However, this latency difference was not found to be statistically significant. DISCUSSION The results indicate a congruence with the findings across previous studies. The P2 component of the late latency response (LLR) which receives a contribution from the auditory association areas was seen to have higher amplitudes in the left ears of the musician's group which may be attributed to music -specialization in the association areas of the right hemisphere, remaining in agreement with a previous study done by Joseph in 1998.12 The musician group had notably higher -amplitudes and shorter latencies of P300 CAEP, suggestive of greater neuronal contribution of auditory resources including attention, and deviance detection for auditory stimuli compared to the non-musician group. This is an indication of neuro-plastic changes that occur in musicians on account of musical training. The difference that is obtained in the right ear demonstrates that music training not only causes plastic changes in the right hemisphere (dominant for music) but also has its transfer effects in the left hemisphere (non-dominant), heightening the neural connections in the hemisphere which is -commonly considered to be dominant for language processing as opposed to music processing alone. Stronger neural connections and an increase in neural synchrony are implied by changes in the morphology of the waves of CAEPs (in terms of decrease in latency and increase in amplitude) commonly seen due to neuroplastic changes resulting from musical training.13 Literature on P300 in musicians demonstrates that musicians have an advantage over non-musicians in the cognitive and sensory domains, especially in terms of auditory processing.1 Interestingly, the enhanced P300 responses found for speech sounds and music found in this study indicate that -music training could have induced transfer effects causing cross-domain plastic changes for speech processing, as well. Higher amplitudes for P300 elicited by both chords and speech, seen in musicians, suggests that they have a better processing of pitch (for speech as well as music) as compared to non-musicians.1 In this comparison of musicians and non-musicians, changes in cortical processing are evident in the auditory evoked potentials obtained in P300 responses which receive contributions from attention, cognition, and linguistic networks. Further examinations of CAEPs as well as behavioural measures for speech and music stimuli are needed to better understand music perception and the effect of long-term auditory training. Music education involves not only listening training, but refinement of motor skills concepts which is equivalent to learning a new language. By adopting music stimuli into CAEP testing, audiologists could differentially examine aspects of audition such as perception of pitch, rhythm, timbre, and auditory object perception in various types of musical instruments such as strings, keys, percussion, and vocals. Incorporating these aspects of auditory learning may translate into clinical utility in cases of central auditory processing disorder, developmental language disorder, dyslexia, Alzheimer's disease, Parkinson's disease, schizophrenia, as well as age-related hearing loss where plasticity effects play an important role in further communication outcomes.1 However, the involvement of other factors such as non-auditory sequencing skills, improved attention, and pre-existing structural and functional differences should be assessed minutely before concluding how auditory training, specifically, music training, and language skills are interlinked.14 Longitudinal studies in children of the same age may eliminate the effects of predisposing factors from the effects of music training.15 The -results of this study align with previous research that indicates a positive effect of musical training on auditory language processing abilities3 and may provide a direction for specific examination of neuroplasticity using a larger sample size. In summary, a statistically significant difference was found for both speech and non-speech stimuli in P300 amplitude with the difference being greater in the left ear indicating positive effects of musical training. A large-scale study of age-related -neuroplastic changes in the P300 responses is still needed. Future fMRI findings, MEG studies, and related evoked -potentials measured at multiple electrode sites may aid in the understanding of the exact site of generation of the LLR and P300 responses.