Enhanced Complex Mismatch Negativity and Mnemonic Representations in Older Adult Musicians

Enhanced Complex Mismatch Negativity and Mnemonic Representations in Older Adult Musicians | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 30 September 2025 V1 Latest version Share on Enhanced Complex Mismatch Negativity and Mnemonic Representations in Older Adult Musicians Authors : Jennifer Bugos 0000-0001-8061-3752 [email protected] , Ricky Chow 0000-0002-7977-656X , Shimin Mo , R. Shayna Rosenbaum , and Claude Alain Authors Info & Affiliations https://doi.org/10.22541/au.175926445.57433481/v1 277 views 220 downloads Contents Abstract Supplementary Material Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Instrumental music performance is associated with enhanced perceptual processing as evidenced by auditory discrimination and speech-in-noise perception. However, little is known about the extent to which auditory perceptual processes support cognition in aging. We investigated whether music training enhances precision in perception in 26 older amateur and professional musicians (62–85 years, 13 females) and 25 older non-musicians (61–82 years, 16 females). Participants completed a novel paradigm of auditory mnemonic discrimination while electroencephalography (EEG) was recorded. The mismatch negativity (MMN), an event-related potential of change detection, was measured during a passive auditory oddball paradigm with standard and deviant pure-tone sequences differing in pitch contour. Participants subsequently completed an incidental memory test for oddball stimuli (i.e., targets) amongst similar lure sequences (matched for frequency but differing in contour) and dissimilar foil sequences (differing in frequency and contour), as well as a back-to-back perceptual discrimination task. Musicians showed enhanced amplitudes, left-lateralized source activity of the MMN, and increased memory discriminability for targets compared to lures and foils, which was not explained by perceptual discrimination ability or MMN amplitude. No group differences were found for neural or behavioural measures on a mnemonic discrimination task for visual everyday objects. Our results clarify the role of music training on precision in perception and auditory memory in older adult musicians compared to non-musicians. Our findings underscore the contribution of musical engagement to perception and memory to the development of cognitive reserve in aging. [1]¿p#1 Introduction Musical performance requires integrating sensorimotor processing with continuous auditory feedback. Cross-sectional studies show that musicians, compared to non-musicians, demonstrate enhanced perceptual processing and executive functions (Alain et al., 2018; Strong & Mast, 2019). For instance, musicians outperform non-musicians on measures of central auditory processing such as figure-ground perception in speech recognition tasks (Mankel & Bidelman, 2018; Yoo & Bidelman, 2019), gap detection tasks (Zendel & Alain, 2012) and mistuned harmonic detection tasks (Zendel & Alain, 2013). Studies also suggest benefits of musicianship on speech-in-noise processing (Parbery-Clark et al., 2011; Zendel et al., 2019; Zendel and Alain, 2012; but see Whiteford et al., 2025) and have shown correlations between musicianship duration and speech-in-noise performance (Zendel & Alain, 2012). Notably, studies of speech perception found that neural activity in older musicians corresponds to perceptual functions and faster speech processing as found through early cortical responses (Bidelman & Alain, 2015), speech categorization (Bidelman & Walker, 2019), and perceptual learning paradigms (MacLean et al., 2024). Yet, little is known about the relationship between perceptual processing and episodic memory in older adult musicians. This study aimed to examine the role of extensive musical engagement on early sensory processing in older adults and determine the extent to which differences in perceptual abilities between older musicians and non-musicians relate to changes in memory precision. Research shows that music training can benefit perceptual and cognitive processing in older adults. Cross-sectional music training studies comparing older adult musicians with extensive experience to age-matched non-musicians showed enhanced scores on measures of naming, executive functions, and auditory attention (Amer et al., 2013; Hanna-Pladdy & MacKay, 2011; Strong & Mast, 2019). Longitudinal studies of group piano or percussion interventions for older adult novices revealed benefits to executive functions in cognitive flexibility (Bugos, 2019; Mack et al., 2025), working memory (Degé & Kerkovius, 2018), inhibitory control (Bugos, 2010; Seinfeld et al., 2013), and verbal fluency (Bugos, 2010; Bugos & Wang, 2022). A possible reason for these enhancements relates to the ability to segment auditory information into sequential units or patterns of sound sequences that may transfer to higher levels of memory precision (Bugos & Wang, 2022). The benefits of music training also extend to the visual domain. For instance, musicians demonstrated enhanced selective attention compared to non-musicians, which was not explained by increased sensorimotor integration (Rodrigues et al., 2013). Another study showed enhanced visuospatial working memory and inhibition in late middle-aged adults musicians (Amer et al., 2013), suggesting far-transfer effects from extensive musical engagement in a non-auditory domain. Older adult non-musicians tend to experience declines in sensory-perceptual and memory precision (Pishdadian et al., 2020; Stark et al., 2013; Yassa et al., 2011). Even within the auditory domain, it is unclear if benefits of musical experience on sensory precision in aging extends and translates to memory precision. Perception and Memory Precision in Aging Aging is often accompanied by declines in the specificity and precision of episodic memory (i.e., mnemonic discrimination; Korkki et al., 2020; Pishdadian et al., 2020; Salthouse, 2010; Stark et al., 2013; Yassa et al., 2011). One proposed mechanism for this decline is impaired pattern separation—the neural process that enables the encoding of distinct representations for similar or overlapping events (Rolls, 2013, 2016). For instance, studies using the Mnemonic Similarity Task (MST, Stark et al., 2013), a task that places demands on pattern separation, have shown age-related deficits in distinguishing between previously studied everyday objects (targets) and highly similar but new items (lures), with intact performance for clearly dissimilar, unstudied items (foils) (Bowman et al., 2019; Huffman & Stark, 2017; Pishdadian et al., 2020; Stark et al., 2013, 2015; Stark & Stark, 2017; Toner et al., 2009; Yassa et al., 2011). Age - related memory impairment may reflect deficits in encoding (i.e., sensory acquisition) and retrieval (Craik & Rose, 2012; Langnes et al., 2019). Research reveals age-related differences across sensory modalities when encoding simple stimuli (Alain et al., 2022; Reuter et al., 2013). These findings suggest that perceptual declines could contribute to memory impairments, with difficulties in perceptual discrimination at encoding potentially leading to less distinct memory representations (Clinard et al., 2010; Gellersen et al., 2021; Roberts & Allen, 2016). Yet, little is known about memory precision or the role of sensorimotor integration experience gained through music training on encoding in aging. To our knowledge, no study has examined the effects of sensorimotor expertise in aging instrumental musicians on sensory processing and memory precision. Sensory Processing and the MMN Electroencephalography (EEG), specifically, event-related potentials (ERPs), provides a sensitive and temporally precise method for examining the neural mechanisms underlying sensory processing. While neuroimaging research showed neural activity in visual cortex during encoding predicted memory precision in young adults (Wing et al., 2020) and older adults (Bowman et al., 2019), fMRI studies do not contain the temporal resolution necessary to evaluate perceptual processing, which is influenced by the brain’s ability to make predictions. A well-established neural correlate of change detection in perception is the mismatch negativity (MMN), an auditory ERP component that reflects pre-attentive, automatic detection of deviations from expected sensory input or a bottom-up neural index of sensory prediction errors (Bartha-Doering et al., 2015; Garrido, Kilner, Stephan, et al., 2009; Näätänen et al., 2007; Sussman et al., 2003). The MMN is typically elicited in oddball paradigms where a repetitive, standard stimulus (e.g., a pure tone) is occasionally replaced by a deviant stimulus differing in a perceptual feature (e.g., frequency, intensity, or duration). The repeated standard stimulus generates a prediction, while the deviant violates this prediction, triggering a prediction error response (Barron et al., 2020; Garrido, Kilner, Stephan, et al., 2009). The MMN is calculated as the difference waveform between ERPs to deviant and standard stimuli, typically manifesting as a fronto-central negative deflection peaking between 100–250 ms post-stimulus. The MMN component is particularly valuable for probing early sensory encoding and auditory sensory memory independent of attentional confounds. The amplitude and latency of the MMN are linked to perceptual discrimination processes and differences between musicians and non-musicians: larger amplitudes and shorter latencies are typically observed when the physical difference between standard and deviant stimuli is more salient, and earlier MMN onsets predict faster behavioral responses to changes in the auditory environment (Näätänen & Alho, 1997; Schröger et al., 1992; Tiitinen et al., 1994). Several studies suggest that the MMN could be a valid measure of incidental learning based on a memory trace of the standard stimulus (see Paavilainen, 2013). More recently, computational modelling by Teichert et al. (2025) showed that the MMN can index deviance detection based on information that persists on timescales beyond auditory sensory memory. Recent work has shown that age attenuations in MMN amplitudes can partially explain age declines in subsequent episodic memory for oddball stimuli (Chow et al., 2025). Thus, the MMN may rely on adaptation processes as the signal that include reduced responses to repetitive standard sounds, and increased responses to sound violations of predicted sounds or deviance detection. These findings highlight the MMN as a versatile marker of both basic, low-level and more complex abstract prediction mechanisms in auditory processing. Musicians and MMN Research suggests differences between musicians and non-musicians in incidental learning through novel manipulations of musical elements such as pitch, melody, and dynamics. When compared to non-musicians, musicians typically demonstrate increased MMN amplitudes, including violations to higher-order, abstract regularities of auditory representations for tonal patterns (Alain et al., 1999; Fujioka et al., 2004; Horváth & Winkler, 2004; Macdonald & Campbell, 2011; Tervaniemi et al., 2006). Tonal patterns in experiments with up to five pitches, regardless of key presentation, presented between 400 -1000 ms in length, elicited an MMN in adults (Alain et al., 1999; Habermeyer et al., 2009; Tervaniemi et al., 2006). Research examining the sensory memory trace for tones in musicians and non-musicians found a strong musician advantage for detecting deviants in the MMN paradigm, which remained enhanced despite pattern length (Boh et al., 2011; Herholz et al., 2009). In addition, research with musically naive participants suggests that the brain may automatically model pitch relationships based upon musical scale structures (Brattico et al., 2006). However, limited research has examined the MMN in aging musicians, with only one study to our knowledge showing earlier MMN latencies to harmonic tone pitch deviants in a relatively small sample of older musicians than non-musicians (aged 55 to 70 years), though no difference in MMN amplitude was found (O’Brien et al., 2015). In another study, the MMN of older adult choral members showed more robust encoding of complex pitch and location features when compared to older non-musicians (Pentikäinen et al., 2022). While evidence indicates enhanced neural representations of complex auditory stimuli in musicians, it is unclear how these advantages are preserved in older adults. Therefore, the present study examines the extent to which older musicians show enhanced indices of precision in auditory perception (as demonstrated by the MMN) and mnemonic discrimination in both auditory and visual domains. A passive oddball paradigm was used to elicit a pattern-deviant MMN. We hypothesized that musicians compared to non-musicians would demonstrate enhanced encoding and memory performance of perceptually similar sound objects as compared to non-musicians. To examine whether group differences in the pattern-deviant MMN predicted memory precision, subsequent memory for oddball stimuli was tested against similar lures and dissimilar foils. Compared to their non-musician peers, we predicted that 1) older adult musicians would demonstrate enhanced MMN amplitudes to pitch contour deviants (i.e., temporal pattern and arrangement of pitched tones in sequence), and 2) enhanced subsequent memory for oddball stimuli. To examine the extent to which perceptual advantages from musicianship extend to memory precision in the visual domain, neural and behavioural measures of mnemonic discrimination for everyday objects were also tested between groups using the MST. Given prior research demonstrating far-transfer effects, we predicted 3) higher lure discrimination on the visual MST compared to older adult non-musicians. Materials and Methods Participants Older musicians were recruited from community outreach (e.g., online advertisements, musician network mailing lists, talks) in the Greater Toronto Area. Non-musicians were recruited from the Rotman Research Institute participant database. Participants were classified as non-musicians if they reported at least five cumulative years of formal music training (operationalized as individualized private lessons on an instrument beyond general academic curricula or extracurriculars, such as secondary school music classes) and were concurrently playing or performing on an instrument at the time of testing. Participants were classified as non-musicians if they reported fewer than four cumulative years of formal music training and were not concurrently learning or playing any musical instrument at the time of testing. Data from five older adults who did not fit the above criteria (e.g., reported between two and five years of formal music training and reported concurrently actively practicing or learning a musical instrument at the time of testing) were excluded from the study. All participants reported fluency in English, and no history of neurological conditions (e.g., stroke, transient ischemic attack, traumatic brain injury) or formal diagnoses of a mood disorder, substance use disorder or learning disabilities. Participants were excluded if they reported a history of chemotherapy or radiation therapy to the head or neck, or if they were concurrently taking medications known to substantially impact cognitive functioning (e.g., antidepressants or antipsychotics). Additionally, participants were excluded if they reported regular use of hearing aids or self-reported hearing difficulties. No participant underwent a neuropsychological evaluation within the six months prior to testing. No participant scored below the cut-off on the modified version of the Telephone Interview for Cognitive Status (TICS-m). A pure-tone audiogram of octave frequencies between 250 to 8000 Hz was administered to ensure age-appropriate hearing thresholds. Data from one older non-musician were excluded due to moderate hearing loss, operationalized as average thresholds exceeding an average of 35 decibels (dB) hearing level (HL) in both ears; therefore, data from this participant were used in analyses of visual tasks only. Data from one musician and one non-musician were excluded due to cognitive status based on neuropsychological assessment (see below). All participants reported normal or correct-to-normal visual acuity. The final sample of participants comprised 26 older adult musicians (62-85 years, 16 females) and 25 older adult non-musicians (61-82 years, 13 females). A chi-square test showed no significant difference in sex distribution across groups, χ²(1, N = 51) = .47, p = .492. No group difference was found for years of education, t (49) = 1.62, p = .111, d = .455, B 10 = .82; the sample was highly educated, with on average 17.73 years ( SD = 2.51) of education for the musician group and on average 16.60 years ( SD = 2.47) of education for the non-musician group. All participants provided written informed consent and received monetary compensation for participation. The experimental protocol was approved by the Research Ethics Board at the Rotman Research Institute at Baycrest Centre and York University. Musicians reported practicing a variety of instruments (i.e., piano, clarinet, flute, oboe, French horn, saxophone, and violin). All musicians reported training in the classical style and proficiency in reading standard staff notation. The musician sample reported practicing or performing on average 9.13 cumulative hours ( SD = 3.96 hours) per week on an instrument. Eleven of the 26 older musicians reported proficiency on two instruments, and 10 reported proficiency on three or more instruments. Eighteen reported completing a professional degree in music performance (i.e., Bachelor’s degree or higher). Twenty-three musicians started music lessons on at least one instrument before the age of 18, and the other three musicians began their training between 30 to 50 years of age. Nineteen musicians reported playing any single instrument for 10 or more years. Thirteen reported not having taken formal music lessons for 15 years or more. At the time of testing, nine reported concurrently engaging in formal music lessons, 25 reported playing regularly in an ensemble, and three musicians reported engaging in music composition. One musician reported absolute pitch ability (i.e., the ability to label the pitch of a note without an external reference). Eight of the 25 non-musicians reported having received private music lessons in their lifetime; this subset of participants reported on average 2.00 years of consistent formal music education ( SD = 1.31). Only two non-musicians received four years of formal music education in their adolescence and reported no formal music education in the past 50 years. [1]¿p#1 Neuropsychological Assessment All older adult participants were administered the Montreal Cognitive Assessment (MoCA; Nasreddine et al., 2005) as a brief test of global cognitive function. Data from one musician and one non-musician were excluded, given a score below a revised, less conservative cut-off (i.e., below 23, Thomann et al., 2020). The Vocabulary and Matrix Reasoning subtests from the Wechsler Abbreviated Scale of Intelligence (Wechsler, 2008) were administered as estimates of crystallized and fluid intelligence, respectively. The Digit Symbol Coding subtest from the Wechsler Adult Intelligence Scale, 3rd Edition (WAIS-III) was administered to measure processing speed. The Trail-Making Test (Forms A and B) assessed psychomotor speed and cognitive flexibility (Reitan & Wolfson, 1985). Episodic memory performance was assessed using the 3 rd edition of the California Verbal Learning Test (CVLT-3; Delis et al., 2000), along with the Incidental and Free Recall trials of the WAIS-III Digit Symbol Coding Test. The WAIS-IV Digit Span subtest (Wechsler, 2008) was administered to assess auditory attention and working memory span. The Verbal Fluency subtest from the Delis-Kaplan Executive Function System (D-KEFS; Delis et al., 2012) was administered, including trials for measuring phonemic fluency, semantic fluency, and cognitive flexibility. The Musical Ear Test (Wallentin et al., 2010) was administered to test musical aptitude in melodic and rhythmic domains. To assess depressive and anxious symptoms, all participants completed the Generalized Anxiety Disorder 7-item scale (GAD-7; Spitzer et al., 2006) and the Patient Health Questionnaire 8-item scale (PHQ-8; Kroenke et al., 2009). Procedure and EEG Acquisition All participants completed two computerized tasks that examined mnemonic discrimination, the ability to precisely differentiate similar events or objects in episode memory. Participants first completed the Mnemonic Similarity Task (MST), a well-established task that measures mnemonic discrimination in the visual domain using images of everyday objects. This task is thought to place demands on pattern separation by testing episodic memory for old target objects against perceptually similar lures and dissimilar foils (Stark et al., 2013). Subsequently, participants completed a novel task of auditory memory—the Auditory MMN Memory Task (AMMT)—which evaluates mnemonic discrimination in the auditory domain using novel, pre-experimentally unfamiliar tone sequences lacking musicality (i.e., frequencies comprising tone sequence stimuli were constructed irrespective of musical scale). This task, previously called the Auditory Pattern Separation Task (Herman et al., 2023), was administered following the MST. Participants were offered an optional break between the two tasks. To reduce fatigue, older adults (aged 60 and above) completed the memory tasks on a separate day from their neuropsychological assessments. All older participants completed both sessions within two months, and no participant withdrew from the study. All participants completed the MST and the AMMT while seated in a recliner in a double-walled, sound-attenuated booth. EEG data were recorded using 66 Ag/AgCl scalp electrodes with a BioSemi ActiveTwo system (BioSemi V.O.F., Amsterdam, Netherlands), following the 10–20 international layout of electrode placement. A common mode sense and driven right leg electrode served as references. Ten additional electrodes were placed below the hairline to monitor ocular activity and enhance coverage, including both mastoids, pre-auricular points, lateral points neighbouring the outer canthi of the eyes, inferior to each eye, and two frontotemporal locations. EEG data were continuously digitized at a 512 Hz sampling rate (DC–100 Hz bandwidth) and stored offline for analysis. EEG data for the MST are reported in Supplementary Table 1. Auditory MMN Memory Task (AMMT) The AMMT comprised three stages: the oddball phase (i.e., incidental encoding phase), a surprise recognition test, and a same-different perceptual discrimination task. Detailed methods for stimulus construction can be found in Herman et al. (2023). Six distinct tone sequences were created, each consisting of five 100 ms pure tones with varying pitch contours. During the oddball phase, only two of these sequences (i.e., a standard and a deviant), with all six sequences used in the latter two stages. Auditory stimuli were presented at 85 dB SPL through Etymotic ER-3A insert earphones. Oddball Phase Participants were presented with 1,000 auditory sequences (700 presentations of the standard, 300 presentations of the deviant) in randomized order over a 25-minute session while watching a muted movie. They were instructed to attend the movie, and compliance was monitored via webcam. Each sequence lasted 500 ms, followed by a jittered ISI between 900 and 1150 ms (in 50 ms steps). Standard and deviant sequences differed in pitch contour (while matched in average frequency across tones) to elicit the MMN. [1]¿p#1 Recognition Test Phase After encoding, participants completed a surprise recognition test for the tone sequences presented during the oddball phase. Test probes included the original standard and deviant sequences (targets), three novel lure sequences with pitch contours similar to targets, and one foil sequence differing in contour and average frequency. To reduce contributions from short-term and working memory, the test phase began after a 2–3-minute break during which instructions were provided. Participants identified each sequence as “old” or “new” in the context of the encoding phase using a keyboard button press. Each of the six sequence types (two targets, three lures, one foil) was presented ten times in randomized order, totaling 60 trials (20 targets, 30 lures, 10 foils). The task was self-paced, with a 500 ms delay between each sequence and the response prompt. Perceptual Discrimination Task In the final phase, participants completed a same-different perceptual discrimination task to ensure they could distinguish among the sequences presented earlier. While not central to the current study, this phase assessed auditory discrimination abilities. Each trial involved two sequences played back-to-back with a 500 ms ISI. Participants then judged whether each pair was identical or different. All possible pairings were tested in both presentation orders, yielding 84 trials. This self-paced task used a forced-choice format, with responses recorded via keyboard button press. Mnemonic Similarity Task All participants completed the MST as a task of visual mnemonic discrimination that placed demands on pattern separation for everyday objects. It included two phases: an incidental encoding phase and a recognition test. During encoding, participants viewed 128 images of everyday objects and categorized each as “indoor” or “outdoor.” In the subsequent test phase, participants were presented with 192 images and identified each as “old,” “similar,” or “new” in the context of the encoding phase. These images comprised previously viewed targets, perceptually similar lures, and new foils. Each image was displayed for two seconds, with a 500 ms inter-stimulus interval (ISI). Participants were randomly assigned to one of two equivalent image sets (Set C or Set D). Responses were made via button press on a computer keyboard. EEG was recorded for both phases; results of analyses from only the test phase are reported in the present paper. Data Preparation EEG Preprocessing Preprocessing of ERPs was conducted using BESA Research software (version 7.1; MEGIS GmbH, Gräfelfing, Germany). Continuous EEG recordings were re-referenced to the average reference and filtered with a 0.53 Hz high-pass filter (forward, 6 dB/octave) and a 40 Hz low-pass filter (zero-phase, 24 dB/octave). Electrodes demonstrating substantial artifacts or drift (for example, due to head or body movements) were interpolated using spherical spline interpolation (Picton et al., 2000); no more than 10% of channels were interpolated per participant. To correct for ocular artifacts (such as eye blinks and horizontal eye movements), a spatial filtering approach was used based on Berg and Scherg (1994) and Ille et al. (2002); spatial topographies that best accounted for blinks and vertical/lateral eye movements for each participant were marked and then subtracted from the continuous EEG recording. AMMT Oddball. For each participant, EEG data from the passive oddball paradigm were segmented into 1000-ms epochs with a 200-ms pre-stimulus baseline. These epochs were additionally screened for artifacts with those demonstrating peak-to-peak fluctuations exceeding ±60 μV for a single channel, which were excluded from further analysis. For musicians, this excluded on average 9.45% ( SD = 7.95%) of standard trials and 7.48% ( SD = 6.91%) of deviant trials per participant. For non-musicians, this excluded on average 9.06% ( SD = 8.14%) of standard trials and 7.17% (6.25%) of deviant trials per participant; this did not differ between groups, F (1, 49) = .89, p = .351, η²G = .018. The remaining epochs were then averaged separately according to standard and deviant conditions, and baseline-corrected using the 200-ms pre-stimulus interval. To visualize the MMN component, the difference waveform was computed for each participant by subtracting the averaged standard waveform from the deviant waveform. Based on prior investigations using the same paradigm (Chow et al., 2025; Herman et al., 2023), the late discriminative negativity (LDN) component was also expected to be observed as a late deviance-related potential that indexed higher-order, auditory gestalt representations based on higher-order abstract violations to pitch contour. To investigate group differences in neural generators underlying the MMN, source activity was modelled using an iterative application of low-resolution electromagnetic tomography (LORETA), termed classical LORETA analysis recursively applied (CLARA). This imaging approach provides focal localizations of the source activity by weighting images with a reduced source space in each iteration. Two iterations were performed; in the second iteration, the image was spatially smoothed with a half-width of one voxel. The regularization parameters were set with a single value decomposition cutoff at 0.01%, and voxels with amplitudes in the lowest 10% of the first image were set to zero to reduce the effective source image. This thresholded image was then used to define a voxel-wise spatial weighting term for the next iteration, and a new LORETA image was computed incorporating this weighting. The voxel size in Talairach space was set to 7 mm. A four-shell ellipsoidal head model with a head radius of 85 mm, and thicknesses for scalp, skull, and cerebrospinal fluid were set at 6, 7, and 1 mm, respectively. The relative conductivities were 0.33, 0.33, 0.0042, and 1 S/m for brain, scalp, skull, and cerebrospinal fluid, respectively. To ensure source images represented deviance-related activity from oddball detection, CLARA images were computed on difference waveforms generated from the subtraction between standard and deviant trials (rather than standard and deviant waveforms separately). Source reconstructions were then compared between groups using a two-tailed independent-samples permutation t-test. MST Test Phase. EEG data from the test phase were segmented into 1500-ms epochs for each participant, including a 200-ms pre-stimulus baseline. The epochs were then averaged according to condition (i.e., Target, Lure, and Foil); these epochs included for averaging only contained correct responses (i.e., “old” for Target, “similar” for Lure, and “new” for Foil). Epochs were also averaged to a fourth condition for lure false alarms (FA Lure condition, i.e., lures incorrectly classified as “old”) to examine false recognition of lures. To ensure adequate number of trials per condition, a closer inspection of trial counts revealed inadequate trial accounts for the correct Lure condition (on average 18.40 trials ( SD = 7.66) for musicians, on average 17.00 trials ( SD = 9.85) for non-musicians, with only 30 participants with 15 or more trials with correct responses to lures) and were deemed not suitable for analysis due to low statistical power. Epochs were then screened for additional artifacts and those exceeding ±120 μV peak-to-peak amplitude were excluded from further analysis. The percentage of trials excluded did not significantly differ across groups ( F (1, 48) = .105, p = .748, η²G = .002). For musicians: 8.89% ( SD = 8.50%) for Target, 9.04% (SD = 10.07%) for Foil, and 9.40% ( SD = 8.37%) for the FA Lure condition; for non-musicians: 8.09% ( SD = 8.76%) for Targets, 9.22 ( SD = 8.93%) for Foil, and 7.85% ( SD = 7.60%) for the FA Lure condition). The remaining conditions were comprised of moderate trial counts (for musicians: on average 48.04 trials ( SD = 6.95) for Target, 44.64 ( SD = 8.45) for Foil, and 31.6 ( SD = 6.42) for the FA Lure condition; for non-musicians: on average 46.48 trials ( SD = 9.13) for Target, 42.00 ( SD = 10.73) for Foil, and 31.96 ( SD = 8.53) for the FA Lure condition), which did not differ between groups, F (1, 48) = 1.51, p = .225, η²G = .014. Epochs were then baseline-corrected using the 200-ms pre-stimulus interval. From these data, the FN400 (as an index of familiarity, Curran, 2000) was expected to be visible as a frontocentral modulation between waveforms from correctly identified targets compared to correctly rejected foils, with a greater negativity for foils than targets. Similarly, the FN400 was also expected to be visible between waveforms from falsely recognized lures compared to correctly rejected foils. A greater FN400 in this contrast would signify greater false recognition of lures and worse mnemonic discrimination. Therefore, two difference waveforms were computed for each participant. To examine group differences in the old-new effect in episodic memory, the old-new difference waveform was created by subtracting the averaged Foil from Target conditions. Another difference waveform was created by subtracting the Foil condition from the FA Lure condition to examine group differences in false lure recognition. Given only moderate trial counts for the condition and to reduce the variability of individual participant waveforms, these difference waveforms were additionally filtered using a 20 Hz low-pass filter (zero-phase, 24 dB/octave). Behavioural Measures AMMT. Sensitivity metrics (i.e., d -primes and response bias) were computed for the test phase: target–foil discriminability as a measure of the conventional old–new recognition performance in episodic memory, defined as z (old | targets) – z (old | foils); and target–lure discriminability as a measure of mnemonic discrimination, calculated as z (old | targets) – z (old | lures). As a measure of response bias, the criterion c was calculated per participant for each of target–lure and target–foil discriminability. Each measure was scaled by its corresponding number of trials (20 target trials, 30 lure trials, and 10 foil trials). This yielded a criterion c value for each of the two contrasts that reflected the tendency to respond ”old” regardless of stimulus type, with positive values indicating a conservative bias and negative values indicating a liberal bias. Both d -prime and c measures were based on loglinear-adjusted d -prime values, correcting for extreme response rates using the method outlined by Hautus (1995) using the psycho R package (Makowski, 2018). While standard tone sequences occurred more frequently than deviant ones during encoding (at a seven to three ratio), Wilcoxon signed-rank tests indicated no significant differences in hit rates between the two target types across the sample ( W = 486.0, z = 1.34, p = .180, r rb = .25), and no significant interaction between group and target subtype ( F = 1.27, p = .265, η²G = .009). Data from one musician was excluded on this task due to near-ceiling response bias for “new” across the task. For the perceptual discrimination task, hit rates were computed for three conditions: correctly identifying identical pairs as same , identifying similar pairs (i.e., target-lure or non-identical lure pairs) as different , and identifying dissimilar pairs (i.e., foil sequence paired with either target or lure sequence) as different . Data from one non-musician was excluded on this task due to not understanding instructions. MST. Data from one non-musician were excluded due to colour-blindness that was self-reported after study participation, and data from another non-musician were excluded due to falling asleep during the majority of the test phase; data were included from one non-musician who did not participate in the AMMT due to mild hearing loss (total of n = 24 non-musicians). Trials from the test phase were excluded from analysis if responses occurred outside the 2500 ms response window. To control for accidental button presses, responses that occurred faster than 200 ms were excluded from accuracy calculations. On average, 1.18% ( SD = 1.62%) of trials were excluded per participant in the musician group, and 3.80% ( SD = 4.62%) for the non-musician group, with a greater proportion of trials trimmed for non-musicians ( W = 163.5, z = 2.92, p = .004, r rb = .476). Calculation of hit rates were performed for each of the three response categories, i.e., correctly identifying targets as “old”, lures as “similar”, and foils as “new”. Old-new recognition performance was computed by subtracting the rate of false alarms to foils from the hit rate to targets (i.e., old | target – old | foil). Mnemonic discrimination was calculated as the difference between the proportion of lures correctly identified as “similar” and the proportion of foils incorrectly identified as “similar” (i.e., similar | lure – similar | foil). Data Analysis Event-Related Potentials ERPs were analyzed using a data-driven approach using cluster-based permutation testing identifying statistically significant spatiotemporal clusters reflecting amplitude differences between conditions or groups. For the AMMT Oddball, a two-tailed paired-samples permutation t -test between standard and deviant waveforms was conducted within each group to confirm the presence of the MMN. An independent samples permutation t-test was conducted to examine group effects on the difference waveforms. ERP analyses were performed using BESA Statistics (version 2.1; MEGIS GmbH, Gräfelfing, Germany), which performs cluster-based permutation testing and topographic ANOVAs that account for all electrodes and time points within an epoch, with robust correction for multiple comparisons. Each permutation test proceeded in two stages. In the first stage, parametric t-tests were computed at every electrode and time point: paired-samples t-tests for amplitude waveforms (e.g., AMMT Oddball standard versus deviant trials; for the MST test phase, Target vs. Foil and FA Lure vs. Foil contrasts) and independent-samples t-tests for group comparisons of difference waveforms and CLARA source images (for the AMMT, the standard-deviant difference; for the MST, the difference waveform between Target and Foil conditions, and between FA Lure and Foil conditions). Data points with p -values below an initial cluster-forming alpha threshold were grouped into spatiotemporal clusters. For the AMMT Oddball, the cluster alpha was set at 0.01 given the focal nature of the effects; for the MST Test phase, this was set at 0.05 given more distributed old-new effects. The t -values within each cluster were then summed to generate a cluster-level test statistic. In the second stage, Monte Carlo permutation testing was performed by randomly shuffling condition or group labels across 5000 permutations to create a null distribution of maximum cluster statistics. The observed cluster-level statistics were then compared to this distribution to compute Monte-Carlo p -values, defined as the proportion of permutations yielding cluster values greater than or equal to the observed cluster value; for further details on cluster-based permutation testing, see Maris and Oostenveld (2007). Permutation testing was followed by supplementary analyses confirming the presence of the MMN and LDN between groups. Permutation testing was followed up by conventional ROI-based approaches using a specified cluster of electrodes. For the AMMT Oddball phase, MMN amplitudes were extracted from seven frontocentral electrodes between 200-400ms, and LDN amplitudes were extracted from 550-700 ms following Chow et al. (2025). To explore whether the sensory components of ERPs differ between groups, latencies and amplitudes of early auditory-evoked responses (i.e., N1 and P1 components) from the AMMT Oddball were also examined only from standard trials. Peak latencies were measured as the maximum positivity or negativity within a specific time window. Upon visual inspection of grand mean waveforms, both P1 and N1 components were maximal at frontal and frontocentral sites. P1 peak latencies were exported from each participant at the latency of the maximal positive-going peak between 20 to 80 ms. N1 peak latencies were measured for each participant between 60 to 140 ms. As for amplitudes, P1 mean amplitudes were derived from the same electrode cluster between 30 and 70 ms, and N1 mean amplitudes between 80 and 120 ms. Time windows for peak latencies were liberal to capture individual variability in latency and skew, whereas time windows for mean amplitudes were more conservative to account for individual variability in the width of P1 and N1 components. For the MST Test phase, permutation testing was conducted within each group between Target and Foil conditions (i.e., Target–Foil contrast) to verify the presence of the FN400 and between groups to examine whether this component was modulated by musicianship. Permutation testing was also conducted on the difference between the FA Lure condition and the Foil condition (i.e., Lure–Foil contrast) to examine the group difference in the FN400 for false recognition of lures. Behavioural Measures AMMT Test Phase Accuracy. A generalized linear mixed-effects model (GLMM) using a binomial logistic link was used to predict trial-level accuracy on the test phase of the AMMT. Fixed effects included Group (Musician, Non-Musician), Condition (Target, Lure, Foil), Trial (i.e., trials 1 through 60), Age, and Group by Condition and Group by Trial interaction terms. For the Group factor, the non-musician group was set as the reference level; for Condition, Target was used as the reference level, such that all effects for Condition compared hit rates of foils or lures relative to target trials. The fixed effects of Trial and Age were each standardized to facilitate convergence. Variance inflation factors (VIFs) were inspected to assess multicollinearity, and all adjusted VIF values (GVIF^(1/(2×Df))) for fixed effects were below 2.5, indicating no serious multicollinearity among predictors. The initial model included random intercepts for Participant; a second model included additional uncorrelated random slopes for Condition to account for individual differences in response accuracy between stimulus types (e.g., some participants showing greater difficulty with lures than others). The bobyqa optimizer further facilitated convergence with an increased iteration limit of 200000. Given that an exploratory model that included correlated random slopes for Condition failed to converge, uncorrelated slopes were specified to allow participant-level variability in condition effects without estimating their correlation with the intercept. The model with uncorrelated random slopes for Condition provided a significantly better fit to the data than the intercept-only model, χ2(6) = 242.84, p < .001. Model fit indices also supported the second, more complex model (AIC = 3206.0, BIC = 3302.1) over the simpler intercept-only model (AIC = 3436.9, BIC = 3496.9), and the log-likelihood improved from -1708.4 (intercept-only model) to -1587.0 (uncorrelated random slopes model). Post-hoc comparisons were conducted using estimated marginal means with Tukey adjustment for multiple comparisons. Simple effects analyses were performed to probe significant Group by Condition interactions. AMMT Cumulative Hit Rates. To further examine the effects of learning and interference for each stimulus type and between groups, linear mixed-effects modelling was conducted, including fixed effects of Group, stimulus type (i.e., Standard, Deviant, Lure1, Lure2, Lure3, Foil), and Probe Position (i.e., from first to tenth position of each stimulus type across time on task) on accuracy. To account for individual differences in trial-wise performance across participants, model comparison was conducted to determine whether including random slopes for Probe Position significantly improved model fit. The initial model included random intercepts for each participant, while a more complex model included both random intercepts and random slopes for Probe Position within participants, specified as uncorrelated random effects. The more complex model with random slopes showed better fit than the intercept-only model, χ²(1) = 106.69, p < .001; in other words, accounting for individual differences in learning trajectories significantly improved the model’s explanatory power. Simple slopes were used to examine significant three-way interactions. In this context, steeper slopes denote a greater rate of cumulative accuracy across the 10 probe repetitions. Sensitivity Metrics. For the AMMT Test phase, d -primes and criterion c values for each of the target-lure and target-foil contrasts were subjected to one-way ANCOVAs between groups with age and sex as covariates. Partial generalized eta-squared was reported as a measure of effect size. Hit rates from all three conditions of the AMMT Perceptual Discrimination Task violated normality due to ceiling effects (identical pairs: W = .69, p < .001; for similar pairs: W = .87, p < .001; for dissimilar pairs: W = .51, p < .001). Therefore, non-parametric Mann-Whitney U-tests were used to analyse auditory discrimination abilities between groups, with rank-biserial correlations reported as effect size measures. As for the MST Test phase, LDI and old-new recognition performance were each subjected to the same aforementioned ANCOVA structures with age and sex as covariates. Correlation and Mediation Analyses Bivariate two-tailed Pearson correlations with age as covariates were conducted to examine the relationship between event-related potentials and behavioural performance (i.e., MMN with AMMT d -primes). Analyses were run with 1000 bootstrap samples, and bias-corrected and accelerated (BCa) confidence intervals are reported. For variables that violated normality, Spearman rho correlations were also run while controlling for age, with 95% confidence intervals reported. Mediation models using the R lavaan package were tested to examine whether MMN amplitude or perceptual discrimination ability mediated the effect of musicianship on AMMT target-lure and target-foil d- primes. A series of parallel mediation models were specified where Group (Musician, Non-musician) was entered as the exogenous predictor, d -prime was entered as the outcome variable, and MMN amplitude and perceptual discrimination hit rates were independent mediators. Specifically, for the target-lure discrimination model, the mediator was the hit rate for correctly identifying similar pairs; for the target-foil discrimination model, the mediator was the hit rate for correctly identifying dissimilar pairs. Models were estimated using full information maximum likelihood, and standard errors for indirect effects were estimated using 5000 bootstrap samples. All preprocessing and statistical analyses for behavioural measures were conducted using JASP (version 0.19.3) and R software (version 4.2.3). Neuropsychological Measures Demographic and neuropsychological data for each group are presented in Table 1. Both groups were, on average, highly educated and showed average to high scores on general intellectual functioning, episodic memory, language, and executive functioning. For neuropsychological tests with scaled scores, non-parametric tests of significance (i.e., Bayesian Mann-Whitney U tests) were conducted. Analyses showed weak evidence of higher scaled scores on a test of processing speed (i.e., WAIS-III Digit-Symbol) for musicians than non-musicians, U = 440.0, p = .029, r = .354, B 10 = 1.37). Analyses on all other neuropsychological measures showed weak or null evidence of group differences. Bayesian independent-samples t -tests showed strong evidence for a group difference on the MET. As expected, musicians showed higher scores for the MET total score than non-musicians, indicative of higher musical aptitude ( t (49) = 3.20, p = .002, d = .90, B 10 = 15.11). Musicians also scored higher for the melodic subdomain ( t (49) = 3.29, p = .002, d = .92, B 10 = 18.41). As for the rhythmic subdomain, analyses showed weak evidence for a group difference ( t (49) = 2.28, p = .027, d = .67, B 10 = 2.24). [1]¿p#1 Event-Related Potentials AMMT – Oddball Paradigm P1 and N1 Components. Standard and deviant grand-averaged waveforms for both groups and the resulting difference waveforms and scalp topographies are displayed in Figure 1. As expected, the P1-N1-P2 complex generated by the first tone of the five-tone sequence was visible and is a hallmark of the frontocentral auditory-evoked potential (Picton, 2010). The N1 corresponding to each of the four subsequent tones was also visible in both standard and deviant waveforms across musician and non-musician groups. The ANCOVAs on early sensory components of the AEP (P1 and N1 of the P1-N1-P2 complex) with age and sex as covariates revealed no group difference for P1 latencies, F (1, 47) = 2.26, p = .139, η²G = .056, or for N1 latencies, F (1, 47) = .20, p = .659, η²G = .011. Similarly, analyses for mean amplitudes showed no group difference for either the P1 ( F (1, 47) = 1.08, p = .305, η²G = .021 ) or N1 deflections ( F (1, 47) = .51, p = .477, η²G = .013). MMN and LDN Components. The presence of the MMN and LDN for each group was statistically verified via a data-driven approach, followed by traditional analyses on mean amplitudes over a specified cluster of nine electrodes. The MMN was identified in both musician and non-musician groups as an early frontal-frontocentral negativity (approximately 200–400ms) in deviant trials compared to standard trials. Permutation testing verified the presence of the MMN at frontal and frontocentral electrodes, as well as its polarity reversal that spanned lateral and temporal electrodes. The LDN was also visualized via a later frontal negativity (~500–750ms) in deviant trials compared to standard trials. Similarly, permutation testing statistically quantified the presence of the LDN at frontal and frontocentral electrodes and its polarity reversal that spanned lateral and temporal electrodes. To statistically quantify the group difference in MMN amplitude, permutation testing was also conducted on the difference waveforms between musician and non-musician groups. Analyses showed larger MMN amplitudes in musicians than non-musicians ( p = .044) at frontal and frontocentral electrodes. No clusters identified a statistically significant difference between groups for the LDN. Cluster-based statistics for the within-group and between-group comparisons are shown in Table 2. Results of permutation testing were verified by an ANCOVA between groups for MMN amplitude values used in the correlation (i.e., values averaged over 200–400 ms time window, across seven frontal and frontocentral electrodes) with age and sex as covariates. As expected, musicians showed larger MMN amplitudes than non-musicians, F (1, 47) = 5.57, p = .022, η²G = .098). An ANCOVA with the same covariates was performed for LDN amplitudes but showed no group difference, F (1, 47) = 0.25, p = .622, η²G = .003. CLARA MMN source reconstructions were averaged over a 250-350 ms time window representing most of the deviant-related response. Both groups showed focal bilateral source activation spanning primary auditory, superior temporal, and inferior frontal regions. Permutation testing revealed one cluster (maximum t- value = 4.79, p = .006, statistical extrema at Talariach coordinates x = -31.5, y = -2.93, z = -4.34) indicating greater source activity spanning the left A1, superior temporal gyrus, and inferior frontal gyrus (IFG) for older musicians compared to non-musicians (Figure 2). MST Test Phase Event-related potentials of the test phase of the MST, including the resulting difference waveforms and scalp topographies, are displayed in Figure 3. Upon visual inspection, the P1-N1 complex of the visual-evoked potential was visible in both groups as expected. In the Target–Foil contrast, a subsequent old-new modulation was also visible in both groups over frontal and frontocentral scalp regions, with a greater negativity for foil than target trials, and was therefore identified as the FN400. Similarly, in the Lure–Foil contrast, the FN400 was also visible in both groups over frontal and frontocentral sites, with a greater negativity for foil than falsely recognized lures. Within each group, cluster-based permutation testing confirmed the presence of the FN400 in both contrasts (see Supplementary Tables). However, no group difference was found in the FN400 modulation for either the Target–Foil or Lure– Foil contrast. Behavioural Measures AMMT Test Phase Accuracy. Figure 4A shows hit rates per condition and group on the subsequent test phase. A generalized linear mixed model (GLMM) with a binomial distribution was used to examine trial-level accuracy, with random intercepts and uncorrelated random slopes for Condition included for each participant; from the model, Table 3 shows a summary of model parameters. A significant Group by Condition interaction was found for foil hit rates compared to targets, b = 3.73, SE = .98, z = 3.79, p < .001, indicating that musicians were more accurate than non-musicians for foil hit rates. However, the Group by Condition interaction for lure versus target hit rates was not significant, b = 0.23, SE = .51, z = .45, p = .65, indicating that the reduction in accuracy for lures (relative to targets) did not differ between groups. As for Condition effects, the analysis revealed worse accuracy for lures compared to those for targets as expected, b = -2.08, SE = .36, z = -5.82, p < .001, with no significant difference between foil and target hit rates, b = .02, SE = .56, p = .97. Post-hoc comparisons revealed high foil accuracy for musicians than non-musicians, b = -3.91, SE = .98, z = –3.99, p < .001. No group difference was observed for lures, b = -0.41, SE = .30, z = -1.37, p = .17, or for targets, b = -0.19, SE = .30, z = -0.61, p = .54. Within older non-musicians, accuracy for lures was lower than that of both targets ( b = 2.08, SE = .36, z = 5.82, p < .001) and foils ( b = 2.10, SE = .61, z = 3.45, p = .002), with no difference between target and foil accuracy ( b = -0.02, SE = .56, z = -0.04, p = .999). Similarly, within musicians, accuracy for lures was lower than targets ( b = 1.85, SE = .36, z = 5.13, p < .001) and foils accuracy ( b = 5.60, SE = .92, z = 6.08, p < .001); additionally, accuracy was higher for foils than targets ( b = -3.75, SE = 0.89, z = -4.21, p < .001). The main effect of stimulus type reached significance, indicating a slight decline in accuracy over time on the task, b = -.14, SE = .06, p = .021, though the Group by Trial was not significant, b = -0.14, SE = .09, z = -1.48, p = .14. The effect of Group was non-significant, b = .19, SE = .30, p = .54, and the effect of Age was also not significant, b = 0.07, SE = 0.09, p = .45. In summary, hit rates for lures were significantly worse than for targets or foils within each group, as expected. Musicians demonstrated less interference from foil probes than non-musicians; no group difference was observed for target or lure hit rates. Cumulative Hit Rates Per Probe Type. Table 4 shows a summary of model parameters from the linear mixed model (LMM) that examined cumulative accuracy for each of the six memory probes (i.e., Stimulus Type) and its relative position in the task (i.e., Probe Position from 1 to 10) as a measure of resistance to interference over time. As expected, the main effect of Probe Position was significant, b = .78, SE = .04, t = 19.86, p < .001, indicating that cumulative accuracy improved overall across trials. The model also showed a significant two-way interaction between Probe Position and Stimulus Type; compared to the Standard probe, the rate of cumulative hits over time on task was slower for all three lure types, as expected (Lure 1, b = -.50, p < .001; Lure 2, b = -.44, p < .001; and Lure 3, b = -.49, p < .001). Notably, the model revealed a significant three-way interaction between Group, Stimulus Type, and Probe Position for cumulative accuracy for the Foil probe (relative to Standard), b = .24, SE = .066, t = 3.59, p < .001. Simple slopes analyses revealed, for the Foil probe, steeper cumulative accuracy rates for older musicians compared to non-musicians, b = .30, SE = .06, t (397) = 5.45, p < .001. No other group differences in cumulative accuracy rates reached significance (relative to Standard, Deviant: b = -.07, p = .230; Lure 1: b < .01, p = .971; Lure 2: b = -.04, p = .595; Lure 3: b = .03, p = .608). In summary, the model demonstrated that cumulative accuracy increased with probe repetition. As expected, the model showed slower cumulative hit rates over time for each lure probe relative to the Standard probe. A three-way interaction showed steeper cumulative hit rates for foils (compared to Standard) for musicians than non-musicians, showing greater resistance to interference in memory discriminability. Sensitivity Metrics . Results from the ANCOVA on d-primes showed that musicians outperformed non-musicians on target-lure discriminability, F (1, 46) = 4.27, p = .045, η²G = .076 (see Figures 4B and 4C). Furthermore, musicians also outperformed non-musicians on target-foil discriminability, F (1, 46) = 13.87, p < .001, η²G = .219. Additionally, a group difference was observed for response bias ( c ) for target-foil discriminability, F (1, 46) = 12.40, p < .001, η²G = .007, whereby older musicians adopted a more conservative decision criterion (greater values for c ) compared to non-musicians (i.e., older non-musicians were more likely to incorrectly endorse the foil probe as “old,” indicative of greater false alarm rates). No group difference was found for response bias for target-lure discriminability, F (1, 46) = .37, p = .546, η²G = .200. AMMT Perceptual Discrimination Task Hit rates for the Perceptual Discrimination Task by group are shown in Figure 5. Mann-Whitney U tests revealed no group difference on hit rates for identifying identical pairs ( U = 272.5, z = -.89, p = .374, r = .127). In contrast, musicians showed on average greater accuracy for correctly discriminating similar pairs ( U = 108.5, z = -3.97, p < .001, r = .652) and for discriminating between dissimilar pairs ( U = 215.0, z = -2.21, p = .027, r = .312). MST Test Phase As shown in Figure 6, the group difference for LDI score did not reach significance, F (1, 46) = 4.16, p = .625, η²G = .004, BF₁₀ = .387 . Likewise, no group difference was shown for old-new recognition performance, F (1, 46) < .01, p = .997, η²G = .001, and moderate evidence for a null difference between groups (BF₁₀ = .281). Correlation Analyses Partial correlations controlling for the effects of age and sex showed that the relationship between MMN amplitude and target-lure d-primes did not reach significance within musicians, ( r (22) = -.003, p = .988, BCa CI [-.262, .298]), within non-musicians ( r(22) = -.260, p = .209, BCa CI [-.553, .115]), or for the two groups combined ( r(47) = -.089, p = .539, BCa CI [-.278, .127]). The relationship between MMN amplitude and target-foil d-primes was also non-significant for within musicians ( r(22) = -.088, p = .674, BCa CI [-.444, .264]), within non-musicians ( r(22) = -.319, p = .120, BCa CI [-.579, .023]), or for the two groups combined ( r(47) = -.104, p = .472, BCa CI [-.373, .173]). In other words, individual differences in MMN amplitude in older adult musicians and non-musicians did not explain sensitivity metrics in memory performance on the AMMT. Additional post-hoc correlation analyses with measures on the Perceptual Discrimination Task showed that, across the sample, greater MMN amplitude was associated increased hit rates for discriminating similar pairs ( ρ (47) = .29, p = .040, 95% CI [.04, .532]) while controlling for age, while no significant association was found for hit rates for the other two conditions (dissimilar pairs: ρ (47) = -.13, p = .384, 95% CI [-.408, .175]; for identical pairs: ρ (47) = -.13, p = .377, 95% CI [-.416, .156]). However, this association between MMN amplitude and discriminating similar pairs was not significant when split by group (Musicians: ρ (23) = .09, p = .653, 95% CI [-.300, .496]; Non-Musicians: ρ (21) = .21, p = .337, 95% CI [-.219, .607]). Mediation Analyses Mediation models were run to examine whether enhanced target-lure and target-foil discriminability in older musicians were explained by enhanced MMN amplitudes or perceptual discrimination abilities. A series of parallel mediation models were specified where Group (Musician, Non-musician) was entered as the exogenous predictor, and d -prime was entered as the outcome variable, and MMN amplitude and perceptual discrimination hit rates were independent mediators (see Figure 7 for a schematic of the model and results of model parameters). For the target-lure discrimination model, Group significantly predicted MMN amplitude, B = −.20, SE = .09, p = .021, accounting for 10% of the variance ( R ² = .10). Group also significantly predicted discrimination of similar pairs, B = −.17, SE = .05, p < .001, explaining 22% of the variance ( R ² = .22). The direct effect of Group on d’ (T,L) was not significant, B = −.30, SE = .21, p = .157, with MMN amplitude ( B = −.33, SE = .30, p = .276) and discrimination of similar items ( B = .88, SE = .60, p = .142) also not significantly predictive of d’ (T,L) values. Importantly, the Group effect on d’ (T,L) was not significantly mediated by MMN amplitude, B = .065, SE = .07, p = .369, or through perceptual discrimination for similar pairs, B = −.154, SE = .12, p = .192. The total effect of Group on d’ (T,L) reached significance, B = −.39, SE = .20, p = .047, R ² = .13. As for the target-foil discrimination model, Group significantly predicted MMN amplitude, B = −.20, SE = .09, p = .022, accounting for 10% of the variance ( R ² = .10). Group also significantly predicted discrimination of dissimilar pairs, B = −.04, SE = .02, p = .036, R ² = .09. The direct effect of Group on d’ (T,F) remained significant, B = −.97, SE = .37, p = .010. However, MMN amplitude ( B = .19, SE = .36, p = .590) and ability to discriminate dissimilar pairs ( B = 2.42, SE = 2.55, p = .343) did not significantly predict d’ (T,F). As for indirect effects, Group effects on d’ (T,F) were not mediated by MMN amplitude, B = −.04, SE = .08, p = .617, or through discrimination of different foils, B = −.10, SE = .09, p = .236. The total effect of Group on d’ (T,F) was significant, B = −1.11, SE = .31, p < .001, R ² = .23. Summary of Results On the AMMT Oddball, older musicians showed enhanced MMN amplitudes and greater source activity spanning left IFG and auditory areas compared to non-musicians. On the subsequent test phase, both groups were more likely to falsely endorse lures that were “old” compared to targets and foils in both groups, as expected. Sensitivity metrics revealed enhanced target-lure and target-foil d -primes in musicians, indicative of enhanced memory precision for oddball stimuli and resistance to interference from similar lures and dissimilar foils. Non-musicians showed greater false alarm rates for foils compared to targets (i.e., more likely to incorrectly endorse the foil probe as “old”) compared to musicians. When examining each probe type, results further showed a) increased interference across repetitions for each lure type across groups, and b) greater interference from repetition of the foil probe over time in non-musicians than musicians. No correlation was found between MMN and subsequent memory performance, though MMN amplitude across the older adult sample was correlated with the ability to discriminate similar tone sequences while accounting for age. Although musicians showed better perceptual discrimination between similar and dissimilar pairs of tone sequences, mediation analyses showed that this benefit did not explain the group differences on target-lure and target-foil d- primes. No group differences were found for early auditory P1 or N1 components, or for the LDN. As for the MST, no group difference was observed for the FN400 component or for behavioural measures of lure discrimination or old/new recognition. Discussion Our results suggest that extensive musical engagement is associated with enhanced precision in perception and memory in older adults. We observed enhanced MMN amplitudes for older musicians compared to non-musicians, which was accompanied by enhanced source activity in left auditory and frontal regions for oddball detection. The MMN was selectively enhanced in older musicians as no group differences were observed in earlier sensory components (i.e., P1 and N1) or in later deviance-related potentials (i.e., the LDN). Notably, older adult musicians performed with higher accuracy on auditory lure and foil discrimination than their non-musician counterparts. Furthermore, older musicians showed enhanced back-to-back perceptual discrimination abilities for similar and dissimilar tone sequences, which did not explain enhanced lure and foil discrimination. Although MMN of older adults was related to perceptual discrimination of similar tone sequences, this relationship did not vary by group. Overall, these findings provide neural and behavioural evidence for enhanced sensory processing in aging musicians compared to non-musicians, consistent with prior research in other auditory abilities such as speech-in-noise processing (Bidelman & Krishnan, 2010; Parbery-Clark, Skoe, & Kraus, 2009; Parbery-Clark, Skoe, Lam, et al., 2009; Zendel & Alain, 2014). Additionally, findings indicate enhanced perceptual and mnemonic representations in older adult musicians that may contribute to cognitive reserve in aging. [1]¿p#1 Enhanced MMN and Auditory Prediction in Older Musicians Within the predictive coding framework (Friston & Kiebel, 2009), perception is thought to rely on the brain’s ability to generate and update predictions about incoming sensory input. Deviations from these predictions elicit prediction error signals, which are central to updating internal models of the environment (Friston, 2005; Rao & Ballard, 1999). Based upon predictive coding theories of auditory processing, researchers suggested that MMN reflects prediction error responses when incoming sensory input is compared and weighed to internally-generated models of upcoming sensory events (Garrido, Kilner, Stephan, et al., 2009; Heilbron & Chait, 2018; Wacongne et al., 2012). Our results suggest that long-term engagement of musical activities is associated with enhanced predictive coding in auditory sensory memory (Alain et al., 1998; Doeller et al., 2003; Garrido, Kilner, Kiebel, et al., 2009), with effects observed even in older musicians. Results extend prior models postulating greater predictive processing in musicians (i.e., predictive coding of music model, see Vuust et al., 2022), given our results of enhanced prediction error signals even in older musicians and in a passive setting, minimizing attention. Other work has also shown that rhythmic features in musical performance offer metrical groupings that serve as anticipatory structures from which the brain automatically derives what is heard for rhythm and metrical expectancies (Vuust & Witek, 2014). Similarly, predictions can guide our perception of auditory sequences in the frequency domain (i.e., melodies, Vuust et al., 2022). Notably, source activity underlying enhanced MMN amplitude was greater in older musicians than non-musicians, and was left lateralized to auditory and inferior frontal regions. The conventional network for MMN generation has involved the right rather than the left IFG, in addition to superior temporal gyri (Garrido et al., 2008; Garrido, Kilner, Kiebel, et al., 2009; Opitz et al., 2002). However, more recent investigations have implicated the left IFG in the network underlying generation of MMN to abstract, complex irregularities, such as temporal irregularities or omissions (Chennu et al., 2016; Phillips et al., 2015, 2016; see Heilbron & Chait, 2018), which are in line with pattern-deviant pitch contour violations used in the present study. In the predictive coding framework, greater source activity in frontal regions in musicians may reflect stronger violations from top-down comparisons between incoming and predicted auditory inputs, and may indicate more robust internal predictive models of upcoming sensory events, due to stronger sensory memory traces of the standard stimulus (Garrido, Kilner, Kiebel, et al., 2009; Garrido, Kilner, Stephan, et al., 2009). Regarding left-lateralization of source activity, results also extend prior work demonstrating left-lateralized MMN responses in young adult musicians, particularly with complex deviants such as abstract violations of tone patterns (Herholz et al., 2009), timbre violations (Vuust et al., 2012), meter and rhythmic deviants (Vuust et al., 2005, 2009). Furthermore, the MMN to lexical tones was found to be right lateralized for pitch deviants and left lateralized for pitch contour deviants (Wang et al., 2013), consistent with oddball stimuli in the present study. The group difference in the older musician sample was distributed across auditory and frontal areas, consistent with prior studies positing a network of frontal and auditory sources for MMN generation (Doeller et al., 2003; Tsolaki et al., 2017). Furthermore, our results are consistent with studies that examine long-term musical enculturation effects (Haumann et al., 2018), showing that long-term listening to musical patterns may contribute to left lateralization akin to language-based studies (Pulvermüller et al., 2006). In addition, multisensory studies show that musicians demonstrate higher MMN source activity in the left superior temporal gyrus during audiovisual non-symbolic congruent trials and lower activation for incongruent trials compared to non-musicians (Chalas et al., 2022). Furthermore, shifting of neural sources has been a hallmark of impaired sensory processing in many populations, particularly in auditory processing relevant to musical training (e.g., mild cognitive impairment, auditory processing traits; Yukhnovich et al., 2025). Collectively, these studies suggest that the MMN source generation are subject to neuroplastic changes based upon active or passive music exposure. Our study is the first to our knowledge to show heightened MMN amplitudes in aging adult musicians. These findings are notable, particularly given robust age-related declines in MMN amplitude between young and older adults (Cheng et al., 2013). Research further demonstrates that MMN amplitude is further attenuated for individuals with mild cognitive impairment (i.e., the prodromal stage of dementia) and early stages of Alzheimer’s disease compared to healthy older adults (Lindín et al., 2013; Papadaniil et al., 2016; Tsolaki et al., 2017), with smaller MMN amplitudes related to worse neuropsychological performance for verbal learning (Mowszowski et al., 2012). Given that enhancements in MMN responses may be an indicator of preserved sensory processing in aging, music training may insulate age-related decreases in sensory processing through preserved fidelity of encoding auditory information. Results did not show a correlation between the MMN amplitude and subsequent memory performance. There are many potential explanations for these findings. First, while expertise in auditory tasks such as playing a musical instrument are associated with perceptual sensitivity (Candidi et al., 2014; Groussard et al., 2014; Pantev et al., 1998; Proverbio & Bellini, 2018), our musicians reported a wide range of musical experiences. Each musical instrument requires different perceptual and motor abilities. Despite essential differences in sensorimotor performance, many studies, including the current study, place instrumentalists in one category, mixing professional string and wind players (Rodrigues et al., 2013); amateur instrumentalists (Bidelman & Alain, 2015); or a mix of professional, amateur, and school-recruited musicians (Yamashita et al., 2022). While most studies report that their sample of musicians were actively playing an instrument at the time of testing, the degree to which these previous musical experiences dilute or confound perceptual processing is unknown. In addition, some musicians in the current study and other studies (Yamashita et al., 2022) reported practicing multiple instruments, which may increase variability in musical engagement activities that may underscore enhanced precision in perception. The age at which training commenced, type of training (method or approach) received (e.g., Suzuki Method, Orff Schulwerk approach), intensity of the training, and genre of specialized training (e.g., jazz, classical) could also affect perceptual processing (Bianco et al., 2018). Auditory cortical representations and responses have been associated with the years of music training and the age at which training commenced (Pantev et al., 1998). Additionally, the types of training employed and gene-environment interactions may account for auditory abilities (e.g., the degree to which home and academic/work environments reinforced the development of musical skills). Measures such as the Gold-MSI only partially account for the home listening environment role on perceptual processing; thus, there is a need for robust objective measures of musical experiences. Further studies examining perceptual processing and memory precision in musicians may consider including musical achievement measures to account for variability in musicianship (e.g., dabbler, amateur, professional). Mnemonic Representations in Aging Musicians Based on sensitivity metrices (i.e., d -primes), the present study found evidence for enhanced mnemonic representations for oddball stimuli in older musicians when presented with distractor lures and foils. These findings were notable, particularly given evidence for null differences between groups on baseline neuropsychological measures of episodic memory and auditory attention. Though these findings could be attributed to enhanced auditory discrimination abilities, we view this possibility as unlikely, given that these effects in memory were not explained by perceptual discrimination ability, as shown through mediation models. These enhancements to mnemonic representations were limited to the auditory domain, as no group differences on neural or behavioural measures on the visual MST was found in the present study. These domain-specific effects may be related to extensive deliberate practice of cognitive abilities in the auditory domain involved in music training (i.e., effortful attention, inhibitory control, auditory working memory) throughout years of instrumental tuning, performance monitoring and correcting patterns of rhythmic, melodic, and harmonic sequences, facilitating transpositions, and aural and written music theory. Complex sensorimotor experiences, such as learning a musical instrument, has been shown to increase preservation of sensorimotor regions of the brain, resulting in enhanced speech-in-noise processing (Alain et al., 2014; Zhang et al., 2021, 2023, 2025). Research in pattern separation with older adults suggests age-related declines in fidelity of representing highly similar details belonging to separate, yet overlapping items or events at encoding, as primarily demonstrated in tasks of perceptual and mnemonic discrimination in the visual domain (Davidson et al., 2019; Gellersen et al., 2024; Stark et al., 2015; Stark & Stark, 2017). Animal models suggest that pattern separation processes are attributed to the dentate gyrus, perirhinal cortex, and lateral perirhinal gyrus (Lim & Lee, 2024; Suter et al., 2019). Future work may investigate whether older musicians show preserved structural or functional changes in such hippocampal areas necessary for pattern separation, and whether these abilities generalize across sensory domains. Although the hippocampus, is critical for pattern separation (in particular the dentate gyrus, see Baker et al., 2016), extrahippocampal regions as proposed by the cortico-hippocampal pattern separation framework (ChiPS; Amer and Davachi, 2023). This framework posits that pattern separation is contributed by neural networks beyond those of the hippocampus (e.g., frontoparietal control network) that aid to memory specificity by resolving competition between interfering stimuli in early sensory regions, contributing to hippocampal input for learning and long-term retention. Given that music training may enhance inhibitory control in aging (Bugos, 2010; Seinfeld et al., 2013), musicians may demonstrate strengthened fronto-parietal neural networks, thereby increasing resiliency to resolving interference in aging (Amer & Davachi, 2023). Indeed, cross-sectional fMRI data showed that older musicians demonstrate connectivity transitions between prefrontal and somatomotor regions at short functional distances that may account for automaticity in aging (Guo et al., 2023). As the ChiPS framework posits, interference control contributes to pattern separation by resolving interference of similar, competing sensory inputs, which enables stronger mnemonic representations in episodic memory. Indeed, performance on the test phase of the AMMT placed demands on pattern separation by testing participants’ resistance to interference from similar lure and dissimilar foil sequences, which was shown to increase over time on task. Additionally, older musicians adopted a more conservative response criterion threshold (i.e., less likely endorsing any probe as “old”) suggestive of greater inhibitory control and response suppression when discriminating targets from distractor lures and foils. Interestingly, non-musicians showed greater variability in foil hit rates than musicians and did not show clear ceiling effects as expected; this finding may be due to large interference effects for the foil probe over time in addition to less conservative response thresholds. Therefore, musicianship may serve as a protective factor, providing less susceptibility to interference from similar, competing auditory stimuli at both encoding and retrieval. Future work may examine the extent to which enhanced interference control from extensive musical training mediates perceptual and mnemonic discrimination, and whether these effects generalize across auditory and visual domains. Conclusion Overall, our findings suggest enhanced perceptual processing in aging adult musicians compared to older non-musician counterparts. Data suggest that enhancements may be related to attenuation of sensory deficits that typically occur in neurotypical aging adults. We show older adult musicians with long-term musical engagement may have strengthened internal representations of auditory sensory memory, in addition to more robust mnemonic representations for complex auditory stimuli and reduced susceptibility to interference. As benefits to mnemonic discrimination and old/new recognition were not simply explained by perceptual discrimination ability, findings highlight the role of music training and engagement in supporting cognitive reserve and protecting against age-related declines in auditory encoding and memory precision. Disclosure Statement The authors report no conflict of interest. Funding This work was supported by the Canada First Research Excellence Fund, Vision: Science to Applications program, York Research Chair (to R.S.R.); the GRAMMY Museum Grant (to J.A.B. and C.A.); and the Natural Sciences and Engineering Research Council (RGPIN- 2021-02721 to C.A. and RGPIN-04238-2015 to R.S.R). Acknowledgements We thank Stevenson Baker, Deena Herman, Brian Lau, William Ji, Sabrina Labardo, Aivien Shudimai Mangahas, Mary O’Neil, Shaya Samet, and Jasmine Zhai for assistance in participant recruitment, data collection, and data preparation. Data Accessibility Statement The data that support the findings of this study will be made publicly available at https://osf.io/tzy9n/ References Alain, C., Achim, A., & Woods, D. L. (1999). Separate memory-related processing for auditory frequency and patterns. Psychophysiology , 36 (6), 737–744. https://doi.org/10.1111/1469-8986.3660737 Alain, C., Chow, R., Lu, J., Rabi, R., Sharma, V. V., Shen, D., Anderson, N. D., Binns, M., Hasher, L., Yao, D., & Freedman, M. (2022). Aging enhances neural activity in auditory, visual, and somatosensory cortices: The common cause revisited. Journal of Neuroscience , 42 (2), 264–275. https://doi.org/10.1523/JNEUROSCI.0864-21.2021 Alain, C., Khatamian, Y., He, Y., Lee, Y., Moreno, S., Leung, A. W. S., & Bialystok, E. (2018). Different neural activities support auditory working memory in musicians and bilinguals. Annals of the New York Academy of Sciences , 1423 (1), 435–446. https://doi.org/10.1111/nyas.13717 Alain, C., Woods, D. L., & Knight, R. T. (1998). A distributed cortical network for auditory sensory memory in humans. Brain Research , 812 (1), 23–37. https://doi.org/10.1016/S0006-8993(98)00851-8 Alain, C., Zendel, B. R., Hutka, S., & Bidelman, G. M. (2014). Turning down the noise: The benefit of musical training on the aging auditory brain. Hearing Research , 308 , 162–173. https://doi.org/10.1016/j.heares.2013.06.008 Amer, T., & Davachi, L. (2023). Extra-hippocampal contributions to pattern separation. eLife , 12 , e82250. https://doi.org/10.7554/eLife.82250 Amer, T., Kalender, B., Hasher, L., Trehub, S. E., & Wong, Y. (2013). Do older professional musicians have cognitive advantages? PLOS ONE , 8 (8), e71630. https://doi.org/10.1371/journal.pone.0071630 Baker, S., Vieweg, P., Gao, F., Gilboa, A., Wolbers, T., Black, S. E., & Rosenbaum, R. S. (2016). The human dentate gyrus plays a necessary role in discriminating new memories. Current Biology , 26 (19), 2629–2634. https://doi.org/10.1016/j.cub.2016.07.081 Barron, H. C., Auksztulewicz, R., & Friston, K. (2020). Prediction and memory: A predictive coding account. Progress in Neurobiology , 192 , 101821. https://doi.org/10.1016/j.pneurobio.2020.101821 Bartha-Doering, L., Deuster, D., Giordano, V., am Zehnhoff-Dinnesen, A., & Dobel, C. (2015). A systematic review of the mismatch negativity as an index for auditory sensory memory: From basic research to clinical and developmental perspectives: MMN as an index for auditory sensory memory. Psychophysiology , 52 (9), 1115–1130. https://doi.org/10.1111/psyp.12459 Berg, P., & Scherg, M. (1994). A multiple source approach to the correction of eye artifacts. Electroencephalography and Clinical Neurophysiology , 90 (3), 229–241. https://doi.org/10.1016/0013-4694(94)90094-9 Bianco, R., Novembre, G., Keller, P. E., Villringer, A., & Sammler, D. (2018). Musical genre-dependent behavioural and EEG signatures of action planning. A comparison between classical and jazz pianists. Neuroimage , 169 , 383–394. Bidelman, G. M., & Alain, C. (2015). Musical training orchestrates coordinated neuroplasticity in auditory brainstem and cortex to counteract age-related declines in categorical vowel perception. Journal of Neuroscience , 35 (3), 1240–1249. https://doi.org/10.1523/JNEUROSCI.3292-14.2015 Bidelman, G. M., & Krishnan, A. (2010). Effects of reverberation on brainstem representation of speech in musicians and non-musicians. Brain Research , 1355 , 112–125. https://doi.org/10.1016/j.brainres.2010.07.100 Bidelman, G. M., & Walker, B. (2019). Plasticity in auditory categorization is supported by differential engagement of the auditory-linguistic network. NeuroImage , 201 , 116022. https://doi.org/10.1016/j.neuroimage.2019.116022 Boh, B., Herholz, S. C., Lappe, C., & Pantev, C. (2011). Processing of complex auditory patterns in musicians and nonmusicians. PLOS ONE , 6 (7), e21458. https://doi.org/10.1371/journal.pone.0021458 Bowman, C. R., Chamberlain, J. D., & Dennis, N. A. (2019). Sensory representations supporting memory specificity: age effects on behavioral and neural discriminability. The Journal of Neuroscience , 39 (12), 2265–2275. https://doi.org/10.1523/JNEUROSCI.2022-18.2019 Brattico, E., Tervaniemi, M., Näätänen, R., & Peretz, I. (2006). Musical scale properties are automatically processed in the human auditory cortex. Brain Research , 1117 (1), 162–174. https://doi.org/10.1016/j.brainres.2006.08.023 Bugos, J. A. (2010). The benefits of music instruction on processing speed, verbal fluency, and cognitive control in aging. Music Education Research International , 4 (1), 1–9. Bugos, J. A. (2019). The effects of bimanual coordination in music interventions on executive functions in aging adults. Frontiers in Integrative Neuroscience , 13 . https://doi.org/10.3389/fnint.2019.00068 Bugos, J. A., & Wang, Y. (2022). Piano training enhances executive functions and psychosocial outcomes in aging: results of a randomized controlled trial. The Journals of Gerontology: Series B , 77 (9), 1625–1636. https://doi.org/10.1093/geronb/gbac021 Candidi, M., Maria Sacheli, L., Mega, I., & Aglioti, S. M. (2014). Somatotopic mapping of piano fingering errors in sensorimotor experts: TMS studies in pianists and visually trained musically naïves. Cerebral Cortex , 24 (2), 435–443. https://doi.org/10.1093/cercor/bhs325 Chalas, N., Karagiorgis, A., Bamidis, P., & Paraskevopoulos, E. (2022). The impact of musical training in symbolic and non-symbolic audiovisual judgements of magnitude. PLOS ONE , 17 (5), e0266165. https://doi.org/10.1371/journal.pone.0266165 Cheng, C.-H., Hsu, W.-Y., & Lin, Y.-Y. (2013). Effects of physiological aging on mismatch negativity: A meta-analysis. International Journal of Psychophysiology , 90 (2), 165–171. https://doi.org/10.1016/j.ijpsycho.2013.06.026 Chennu, S., Noreika, V., Gueorguiev, D., Shtyrov, Y., Bekinschtein, T. A., & Henson, R. (2016). Silent expectations: Dynamic causal modeling of cortical prediction and attention to sounds that weren’t. Journal of Neuroscience , 36 (32), 8305–8316. https://doi.org/10.1523/JNEUROSCI.1125-16.2016 Chow, R., Baker, S., Mo, S., Bugos, J. A., Alain, C., & Rosenbaum, R. S. (2025). Mismatch negativity predicts age-related declines in memory precision. Neurobiology of Aging , 147 , 32–48. https://doi.org/10.1016/j.neurobiolaging.2024.11.012 Clinard, C. G., Tremblay, K. L., & Krishnan, A. R. (2010). Aging alters the perception and physiological representation of frequency: Evidence from human frequency-following response recordings. Hearing Research , 264 (1), 48–55. https://doi.org/10.1016/j.heares.2009.11.010 Craik, F. I. M., & Rose, N. S. (2012). Memory encoding and aging: A neurocognitive perspective. Neuroscience & Biobehavioral Reviews , 36 (7), 1729–1739. https://doi.org/10.1016/j.neubiorev.2011.11.007 Curran, T. (2000). Brain potentials of recollection and familiarity. Memory & Cognition , 28 (6), 923–938. https://doi.org/10.3758/BF03209340 Davidson, P. S. R., Vidjen, P., Trincao-Batra, S., & Collin, C. A. (2019). Older adults’ lure discrimination difficulties on the mnemonic similarity task are significantly correlated with their visual perception. The Journals of Gerontology: Series B , 74 (8), 1298–1307. https://doi.org/10.1093/geronb/gby130 Degé, F., & Kerkovius, K. (2018). The effects of drumming on working memory in older adults. Annals of the New York Academy of Sciences , 1423 (1), 242–250. https://doi.org/10.1111/nyas.13685 Delis, D. C., Kaplan, E., & Kramer, J. H. (2012). Delis-Kaplan Executive Function System [Dataset]. https://doi.org/10.1037/t15082-000 Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (2000). California Verbal Learning Test—Second Edition [Dataset]. https://doi.org/10.1037/t15072-000 Doeller, C. F., Opitz, B., Mecklinger, A., Krick, C., Reith, W., & Schröger, E. (2003). Prefrontal cortex involvement in preattentive auditory deviance detection: Neuroimaging and electrophysiological evidence. NeuroImage , 20 (2), 1270–1282. https://doi.org/10.1016/S1053-8119(03)00389-6 Friston, K. (2005). A theory of cortical responses. Philosophical Transactions of the Royal Society B: Biological Sciences , 360 (1456), 815–836. https://doi.org/10.1098/rstb.2005.1622 Friston, K., & Kiebel, S. (2009). Predictive coding under the free-energy principle. Philosophical Transactions of the Royal Society B: Biological Sciences , 364 (1521), 1211–1221. https://doi.org/10.1098/rstb.2008.0300 Fujioka, T., Trainor, L. J., Ross, B., Kakigi, R., & Pantev, C. (2004). Musical training enhances automatic encoding of melodic contour and interval structure. Journal of Cognitive Neuroscience , 16 (6), 1010–1021. https://doi.org/10.1162/0898929041502706 Garrido, M. I., Friston, K. J., Kiebel, S. J., Stephan, K. E., Baldeweg, T., & Kilner, J. M. (2008). The functional anatomy of the MMN: A DCM study of the roving paradigm. NeuroImage , 42 (2), 936–944. https://doi.org/10.1016/j.neuroimage.2008.05.018 Garrido, M. I., Kilner, J. M., Kiebel, S. J., & Friston, K. J. (2009). Dynamic causal modeling of the response to frequency deviants. Journal of Neurophysiology , 101 (5), 2620–2631. https://doi.org/10.1152/jn.90291.2008 Garrido, M. I., Kilner, J. M., Stephan, K. E., & Friston, K. J. (2009). The mismatch negativity: A review of underlying mechanisms. Clinical Neurophysiology , 120 (3), 453–463. https://doi.org/10.1016/j.clinph.2008.11.029 Gellersen, H. M., McMaster, J., Abdurahman, A., & Simons, J. S. (2024). Demands on perceptual and mnemonic fidelity are a key determinant of age-related cognitive decline throughout the lifespan. Journal of Experimental Psychology: General , 153 (1), 200–223. https://doi.org/10.1037/xge0001476 Gellersen, H. M., Trelle, A. N., Henson, R. N., & Simons, J. S. (2021). Executive function and high ambiguity perceptual discrimination contribute to individual differences in mnemonic discrimination in older adults. Cognition , 209 , 104556. https://doi.org/10.1016/j.cognition.2020.104556 Groussard, M., Viader, F., Landeau, B., Desgranges, B., Eustache, F., & Platel, H. (2014). The effects of musical practice on structural plasticity: The dynamics of grey matter changes. Brain and Cognition , 90 , 174–180. https://doi.org/10.1016/j.bandc.2014.06.013 Guo, S., Feng, L., Ding, R., Long, S., Yang, H., Gong, X., Lu, J., & Yao, D. (2023). Functional gradients in prefrontal regions and somatomotor networks reflect the effect of music training experience on cognitive aging. Cerebral Cortex (New York, N.Y.: 1991) , 33 (12), 7506–7517. https://doi.org/10.1093/cercor/bhad056 Habermeyer, B., Herdener, M., Esposito, F., Hilti, C. C., Klarhöfer, M., di Salle, F., Wetzel, S., Scheffler, K., Cattapan-Ludewig, K., & Seifritz, E. (2009). Neural correlates of pre-attentive processing of pattern deviance in professional musicians. Human Brain Mapping , 30 (11), 3736–3747. https://doi.org/10.1002/hbm.20802 Hanna-Pladdy, B., & MacKay, A. (2011). The relation between instrumental musical activity and cognitive aging. Neuropsychology , 25 , 378–386. https://doi.org/10.1037/a0021895 Haumann, N. T., Vuust, P., Bertelsen, F., & Garza-Villarreal, E. A. (2018). Influence of musical enculturation on brain responses to metric deviants. Frontiers in Neuroscience , 12 . https://doi.org/10.3389/fnins.2018.00218 Hautus, M. J. (1995). Corrections for extreme proportions and their biasing effects on estimated values ofd′. Behavior Research Methods, Instruments, & Computers , 27 (1), 46–51. https://doi.org/10.3758/BF03203619 Heilbron, M., & Chait, M. (2018). Great expectations: Is there evidence for predictive coding in auditory cortex? Neuroscience , 389 , 54–73. https://doi.org/10.1016/j.neuroscience.2017.07.061 Herholz, S. C., Lappe, C., & Pantev, C. (2009). Looking for a pattern: An MEG study on the abstract mismatch negativity in musicians and nonmusicians. BMC Neuroscience , 10 (1), 42. https://doi.org/10.1186/1471-2202-10-42 Herman, D., Baker, S., Chow, R., Cazes, J., Alain, C., & Rosenbaum, R. S. (2023). Mismatch negativity as a marker of auditory pattern separation. Cerebral Cortex , 33 (18), 10181–10193. https://doi.org/10.1093/cercor/bhad274 Horváth, J., & Winkler, I. (2004). How the human auditory system treats repetition amongst change. Neuroscience Letters , 368 (2), 157–161. https://doi.org/10.1016/j.neulet.2004.07.004 Huffman, D. J., & Stark, C. E. L. (2017). Age-related impairment on a forced-choice version of the Mnemonic Similarity Task. Behavioral Neuroscience , 131 (1), 55–67. https://doi.org/10.1037/bne0000180 Ille, N., Berg, P., & Scherg, M. (2002). Artifact correction of the ongoing EEG using spatial filters based on artifact and brain signal topographies. Journal of Clinical Neurophysiology , 19 (2), 113–124. Korkki, S. M., Richter, F. R., Jeyarathnarajah, P., & Simons, J. S. (2020). Healthy ageing reduces the precision of episodic memory retrieval. Psychology and Aging , 35 (1), 124–142. https://doi.org/10.1037/pag0000432 Kroenke, K., Strine, T. W., Spitzer, R. L., Williams, J. B. W., Berry, J. T., & Mokdad, A. H. (2009). The PHQ-8 as a measure of current depression in the general population. Journal of Affective Disorders , 114 (1), 163–173. https://doi.org/10.1016/j.jad.2008.06.026 Langnes, E., Vidal-Piñeiro, D., Sneve, M. H., Amlien, I. K., Walhovd, K. B., & Fjell, A. M. (2019). Development and decline of the hippocampal long-axis specialization and differentiation during encoding and retrieval of episodic memories. Cerebral Cortex , 29 (8), 3398–3414. https://doi.org/10.1093/cercor/bhy209 Lim, H.-Y., & Lee, I. (2024). Subpopulations of neurons in the perirhinal cortex enable both modality-specific and modality-invariant recognition of objects. PLOS Biology , 22 (6), e3002713. https://doi.org/10.1371/journal.pbio.3002713 Lindín, M., Correa, K., Zurrón, M., & Díaz, F. (2013). Mismatch negativity (MMN) amplitude as a biomarker of sensory memory deficit in amnestic mild cognitive impairment. Frontiers in Aging Neuroscience , 5 . https://doi.org/10.3389/fnagi.2013.00079 Macdonald, M., & Campbell, K. (2011). Effects of a violation of an expected increase or decrease in intensity on detection of change within an auditory pattern. Brain and Cognition , 77 (3), 438–445. https://doi.org/10.1016/j.bandc.2011.08.014 Mack, M., Marie, D., Worschech, F., Krüger, T. H. C., Sinke, C., Altenmüller, E., James, C. E., & Kliegel, M. (2025). Effects of a 1-year piano intervention on cognitive flexibility in older adults. Psychology and Aging , 40 (2), 218–235. https://doi.org/10.1037/pag0000871 MacLean, J., Stirn, J., Sisson, A., & Bidelman, G. M. (2024). Short- and long-term neuroplasticity interact during the perceptual learning of concurrent speech. Cerebral Cortex (New York, N.Y.: 1991) , 34 (2), bhad543. https://doi.org/10.1093/cercor/bhad543 Makowski, D. (2018). The psycho package: An efficient and publishing-oriented workflow for psychological science. Journal of Open Source Software , 3 (22), 470. https://doi.org/10.21105/joss.00470 Mankel, K., & Bidelman, G. M. (2018). Inherent auditory skills rather than formal music training shape the neural encoding of speech. Proceedings of the National Academy of Sciences , 115 (51), 13129–13134. https://doi.org/10.1073/pnas.1811793115 Maris, E., & Oostenveld, R. (2007). Nonparametric statistical testing of EEG- and MEG-data. Journal of Neuroscience Methods , 164 (1), 177–190. https://doi.org/10.1016/j.jneumeth.2007.03.024 Mowszowski, L., Hermens, D. F., Diamond, K., Norrie, L., Hickie, I. B., Lewis, S. J. G., & Naismith, S. L. (2012). Reduced mismatch negativity in mild cognitive impairment: Associations with neuropsychological performance. Journal of Alzheimer’s Disease , 30 (1), 209–219. https://doi.org/10.3233/JAD-2012-111868 Näätänen, R., & Alho, K. (1997). Mismatch negativity – The measure for central sound representation accuracy. Audiology and Neurotology , 2 (5), 341–353. https://doi.org/10.1159/000259255 Näätänen, R., Paavilainen, P., Rinne, T., & Alho, K. (2007). The mismatch negativity (MMN) in basic research of central auditory processing: A review. Clinical Neurophysiology , 118 (12), 2544–2590. https://doi.org/10.1016/j.clinph.2007.04.026 Nasreddine, Z. S., Phillips, N. A., Bédirian, V., Charbonneau, S., Whitehead, V., Collin, I., Cummings, J. L., & Chertkow, H. (2005). The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society , 53 (4), 695–699. https://doi.org/10.1111/j.1532-5415.2005.53221.x O’Brien, J. L., Nikjeh, D. A., & Lister, J. J. (2015). Interaction of musicianship and aging: A comparison of cortical auditory evoked potentials. Behavioural Neurology , 2015 , 1–12. https://doi.org/10.1155/2015/545917 Opitz, B., Rinne, T., Mecklinger, A., von Cramon, D. Y., & Schröger, E. (2002). Differential contribution of frontal and temporal cortices to auditory change detection: fMRI and ERP results. NeuroImage , 15 (1), 167–174. https://doi.org/10.1006/nimg.2001.0970 Paavilainen, P. (2013). The mismatch-negativity (MMN) component of the auditory event-related potential to violations of abstract regularities: A review. International Journal of Psychophysiology , 88 (2), 109–123. https://doi.org/10.1016/j.ijpsycho.2013.03.015 Pantev, C., Oostenveld, R., Engelien, A., Ross, B., Roberts, L. E., & Hoke, M. (1998). Increased auditory cortical representation in musicians. Nature , 392 (6678), 811–814. https://doi.org/10.1038/33918 Papadaniil, C. D., Kosmidou, V. E., Tsolaki, A., Tsolaki, M., Kompatsiaris, I. (Yiannis), & Hadjileontiadis, L. J. (2016). Cognitive MMN and P300 in mild cognitive impairment and Alzheimer’s disease: A high density EEG-3D vector field tomography approach. Brain Research , 1648 , 425–433. https://doi.org/10.1016/j.brainres.2016.07.043 Parbery-Clark, A., Skoe, E., & Kraus, N. (2009). Musical experience limits the degradative effects of background noise on the neural processing of sound. Journal of Neuroscience , 29 (45), 14100–14107. https://doi.org/10.1523/JNEUROSCI.3256-09.2009 Parbery-Clark, A., Skoe, E., Lam, C., & Kraus, N. (2009). Musician enhancement for speech-in-noise. Ear and Hearing , 30 (6), 653. https://doi.org/10.1097/AUD.0b013e3181b412e9 Parbery-Clark, A., Strait, D. L., Anderson, S., Hittner, E., & Kraus, N. (2011). Musical experience and the aging auditory system: Implications for cognitive abilities and hearing speech in noise. PloS One , 6 (5), e18082. https://doi.org/10.1371/journal.pone.0018082 Pentikäinen, E., Kimppa, L., Makkonen, T., Putkonen, M., Pitkäniemi, A., Salakka, I., Paavilainen, P., Tervaniemi, M., & Särkämö, T. (2022). Benefits of choir singing on complex auditory encoding in the aging brain: An ERP study. Annals of the New York Academy of Sciences , 1514 (1), 82–92. https://doi.org/10.1111/nyas.14789 Phillips, H. N., Blenkmann, A., Hughes, L. E., Bekinschtein, T. A., & Rowe, J. B. (2015). Hierarchical organization of frontotemporal networks for the prediction of stimuli across multiple dimensions. Journal of Neuroscience , 35 (25), 9255–9264. https://doi.org/10.1523/JNEUROSCI.5095-14.2015 Phillips, H. N., Blenkmann, A., Hughes, L. E., Kochen, S., Bekinschtein, T. A., Cam-CAN, & Rowe, J. B. (2016). Convergent evidence for hierarchical prediction networks from human electrocorticography and magnetoencephalography. Cortex , 82 , 192–205. https://doi.org/10.1016/j.cortex.2016.05.001 Picton, T. W. (2010). Human Auditory Evoked Potentials . Plural Publishing. Picton, T. W., Bentin, S., Berg, P., Donchin, E., Hillyard, S. A., Johnson, R., Miller, G. A., Ritter, W., Ruchkin, D. S., Rugg, M. D., & Taylor, M. J. (2000). Guidelines for using human event-related potentials to study cognition: Recording standards and publication criteria. Psychophysiology , 37 (2), 127–152. Pishdadian, S., Hoang, N.V., Baker, S., Moscovitch, M., Rosenbaum, R.S. (2020). Not only memory: Investigating the sensitivity and specificity of the Mnemonic Similarity Task in older adults. Neuropsychologia, 149 , 107670. doi: 10.1016/j.neuropsychologia.2020.107670 Proverbio, A. M., & Bellini, E. (2018). How the degree of instrumental practice in music increases perceptual sensitivity. Brain Research , 1691 , 15–25. https://doi.org/10.1016/j.brainres.2018.04.020 Pulvermüller, F., Huss, M., Kherif, F., Moscoso del Prado Martin, F., Hauk, O., & Shtyrov, Y. (2006). Motor cortex maps articulatory features of speech sounds. Proceedings of the National Academy of Sciences , 103 (20), 7865–7870. https://doi.org/10.1073/pnas.0509989103 Rao, R. P. N., & Ballard, D. H. (1999). Predictive coding in the visual cortex: A functional interpretation of some extra-classical receptive-field effects. Nature Neuroscience , 2 (1), 79–87. https://doi.org/10.1038/4580 Reitan, R. M., & Wolfson, D. (1985). The Halstead-Reitan Neuropsychological Test Battery: Theory and clinical interpretation. Theory and Clinical Interpretation , 4 . https://cir.nii.ac.jp/crid/1370004237582202374 Reuter, E.-M., Voelcker-Rehage, C., Vieluf, S., Winneke, A. H., & Godde, B. (2013). A parietal-to-frontal shift in the P300 is associated with compensation of tactile discrimination deficits in late middle-aged adults. Psychophysiology , 50 (6), 583–593. https://doi.org/10.1111/psyp.12037 Roberts, K. L., & Allen, H. A. (2016). Perception and cognition in the ageing brain: A brief review of the short- and long-term links between perceptual and cognitive decline. Frontiers in Aging Neuroscience , 8 . https://www.frontiersin.org/articles/10.3389/fnagi.2016.00039 Rodrigues, A. C., Loureiro, M. A., & Caramelli, P. (2013). Long-term musical training may improve different forms of visual attention ability. Brain and Cognition , 82 (3), 229–235. https://doi.org/10.1016/j.bandc.2013.04.009 Rolls, E. (2013). The mechanisms for pattern completion and pattern separation in the hippocampus. Frontiers in Systems Neuroscience , 7 . https://www.frontiersin.org/articles/10.3389/fnsys.2013.00074 Rolls, E. (2016). Pattern separation, completion, and categorisation in the hippocampus and neocortex. Neurobiology of Learning and Memory , 129 , 4–28. https://doi.org/10.1016/j.nlm.2015.07.008 Salthouse, T. A. (2010). Selective review of cognitive aging. Journal of the International Neuropsychological Society , 16 (5), 754–760. https://doi.org/10.1017/S1355617710000706 Schröger, E., Näätänen, R., & Paavilainen, P. (1992). Event-related potentials reveal how non-attended complex sound patterns are represented by the human brain. Neuroscience Letters , 146 (2), 183–186. https://doi.org/10.1016/0304-3940(92)90073-G Seinfeld, S., Figueroa, H., Ortiz-Gil, J., & Sanchez-Vives, M. V. (2013). Effects of music learning and piano practice on cognitive function, mood and quality of life in older adults. Frontiers in Psychology , 4 . https://doi.org/10.3389/fpsyg.2013.00810 Spitzer, R. L., Kroenke, K., Williams, J. B. W., & Löwe, B. (2006). A brief measure for assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine , 166 (10), 1092–1097. https://doi.org/10.1001/archinte.166.10.1092 Stark, S. M., & Stark, C. E. L. (2017). Age-related deficits in the mnemonic similarity task for objects and scenes. Behavioural Brain Research , 333 , 109–117. https://doi.org/10.1016/j.bbr.2017.06.049 Stark, S. M., Stevenson, R., Wu, C., Rutledge, S., & Stark, C. E. L. (2015). Stability of age-related deficits in the mnemonic similarity task across task variations. Behavioral Neuroscience , 129 (3), 257–268. https://doi.org/10.1037/bne0000055 Stark, S. M., Yassa, M. A., Lacy, J. W., & Stark, C. E. L. (2013). A task to assess behavioral pattern separation (BPS) in humans: Data from healthy aging and mild cognitive impairment. Neuropsychologia , 51 (12), 2442–2449. https://doi.org/10.1016/j.neuropsychologia.2012.12.014 Strong, J. V., & Mast, B. T. (2019). The cognitive functioning of older adult instrumental musicians and non-musicians. Aging, Neuropsychology, and Cognition . https://www.tandfonline.com/doi/full/10.1080/13825585.2018.1448356 Sussman, E., Winkler, I., & Wang, W. (2003). MMN and attention: Competition for deviance detection. Psychophysiology , 40 (3), 430–435. https://doi.org/10.1111/1469-8986.00045 Suter, E. E., Weiss, C., & Disterhoft, J. F. (2019). Differential responsivity of neurons in perirhinal cortex, lateral entorhinal cortex, and dentate gyrus during time-bridging learning. Hippocampus , 29 (6), 511–526. https://doi.org/10.1002/hipo.23041 Teichert, T., Jedema, H., Zhen, Z., & Gurnsey, K. (2025). Complementary functional profiles of mismatch responses mediated by adaptation and deviance detection point to two distinct auditory short-term memory systems. Journal of Neurophysiology , 134 (1), 314–336. https://doi.org/10.1152/jn.00515.2024 Tervaniemi, M., Castaneda, A., Knoll, M., & Uther, M. (2006). Sound processing in amateur musicians and nonmusicians: Event-related potential and behavioral indices. NeuroReport , 17 (11), 1225–1228. https://doi.org/10.1097/01.wnr.0000230510.55596.8b Thomann, A. E., Berres, M., Goettel, N., Steiner, L. A., & Monsch, A. U. (2020). Enhanced diagnostic accuracy for neurocognitive disorders: A revised cut-off approach for the Montreal Cognitive Assessment. Alzheimer’s Research & Therapy , 12 (1), 39. https://doi.org/10.1186/s13195-020-00603-8 Tiitinen, H., May, P., Reinikainen, K., & Näätänen, R. (1994). Attentive novelty detection in humans is governed by pre-attentive sensory memory. Nature , 372 (6501), 90–92. https://doi.org/10.1038/372090a0 Toner, C. K., Pirogovsky, E., Kirwan, C. B., & Gilbert, P. E. (2009). Visual object pattern separation deficits in nondemented older adults. Learning & Memory , 16 (5), 338–342. https://doi.org/10.1101/lm.1315109 Tsolaki, A. C., Kosmidou, V., Kompatsiaris, I. (Yiannis), Papadaniil, C., Hadjileontiadis, L., Adam, A., & Tsolaki, M. (2017). Brain source localization of MMN and P300 ERPs in mild cognitive impairment and Alzheimer’s disease: A high-density EEG approach. Neurobiology of Aging , 55 , 190–201. https://doi.org/10.1016/j.neurobiolaging.2017.03.025 Vuust, P., Brattico, E., Seppänen, M., Näätänen, R., & Tervaniemi, M. (2012). The sound of music: Differentiating musicians using a fast, musical multi-feature mismatch negativity paradigm. Neuropsychologia , 50 (7), 1432–1443. https://doi.org/10.1016/j.neuropsychologia.2012.02.028 Vuust, P., Heggli, O. A., Friston, K. J., & Kringelbach, M. L. (2022). Music in the brain. Nature Reviews Neuroscience , 23 (5), 287–305. https://doi.org/10.1038/s41583-022-00578-5 Vuust, P., Ostergaard, L., Pallesen, K. J., Bailey, C., & Roepstorff, A. (2009). Predictive coding of music – Brain responses to rhythmic incongruity. Cortex , 45 (1), 80–92. https://doi.org/10.1016/j.cortex.2008.05.014 Vuust, P., Pallesen, K. J., Bailey, C., van Zuijen, T. L., Gjedde, A., Roepstorff, A., & Østergaard, L. (2005). To musicians, the message is in the meter: Pre-attentive neuronal responses to incongruent rhythm are left-lateralized in musicians. NeuroImage , 24 (2), 560–564. https://doi.org/10.1016/j.neuroimage.2004.08.039 Vuust, P., & Witek, M. A. G. (2014). Rhythmic complexity and predictive coding: A novel approach to modeling rhythm and meter perception in music. Frontiers in Psychology , 5 . https://doi.org/10.3389/fpsyg.2014.01111 Wacongne, C., Changeux, J.-P., & Dehaene, S. (2012). A Neuronal Model of Predictive Coding Accounting for the Mismatch Negativity. Journal of Neuroscience , 32 (11), 3665–3678. https://doi.org/10.1523/JNEUROSCI.5003-11.2012 Wallentin, M., Nielsen, A. H., Friis-Olivarius, M., Vuust, C., & Vuust, P. (2010). The Musical Ear Test, a new reliable test for measuring musical competence. Learning and Individual Differences , 20 (3), 188–196. https://doi.org/10.1016/j.lindif.2010.02.004 Wang, X.-D., Wang, M., & Chen, L. (2013). Hemispheric lateralization for early auditory processing of lexical tones: Dependence on pitch level and pitch contour. Neuropsychologia , 51 (11), 2238–2244. https://doi.org/10.1016/j.neuropsychologia.2013.07.015 Whiteford, K. L., Baltzell, L. S., Chiu, M., Cooper, J. K., Faucher, S., Goh, P. Y., Hagedorn, A., Irsik, V. C., Irvine, A., Lim, S.-J., Mesik, J., Mesquita, B., Oakes, B., Rajappa, N., Roverud, E., Schrlau, A. E., Van Hedger, S. C., Bharadwaj, H. M., Johnsrude, I. S., … Oxenham, A. J. (2025). Large-scale multi-site study shows no association between musical training and early auditory neural sound encoding. Nature Communications , 16 (1), 7152. https://doi.org/10.1038/s41467-025-62155-5 Wing, E. A., Geib, B. R., Wang, W.-C., Monge, Z., Davis, S. W., & Cabeza, R. (2020). Cortical Overlap and Cortical-Hippocampal Interactions Predict Subsequent True and False Memory. The Journal of Neuroscience , 40 (9), 1920–1930. https://doi.org/10.1523/JNEUROSCI.1766-19.2020 Yamashita, M., Ohsawa, C., Suzuki, M., Guo, X., Sadakata, M., Otsuka, Y., Asano, K., Abe, N., & Sekiyama, K. (2022). Neural advantages of older musicians involve the cerebellum: Implications for healthy aging through lifelong musical instrument training. Frontiers in Human Neuroscience , 15 . https://www.frontiersin.org/articles/10.3389/fnhum.2021.784026 Yassa, M. A., Lacy, J. W., Stark, S. M., Albert, M. S., Gallagher, M., & Stark, C. E. L. (2011). Pattern separation deficits associated with increased hippocampal CA3 and dentate gyrus activity in nondemented older adults. Hippocampus , 21 (9), 968–979. https://doi.org/10.1002/hipo.20808 Yoo, J., & Bidelman, G. M. (2019). Linguistic, perceptual, and cognitive factors underlying musicians’ benefits in noise-degraded speech perception. Hearing Research , 377 , 189–195. https://doi.org/10.1016/j.heares.2019.03.021 Zendel, B. R., & Alain, C. (2012). Musicians experience less age-related decline in central auditory processing. Psychology and Aging , 27 (2), 410–417. https://doi.org/10.1037/a0024816 Zendel, B. R., & Alain, C. (2013). The influence of lifelong musicianship on neurophysiological measures of concurrent sound segregation. Journal of Cognitive Neuroscience , 25 (4), 503–516. https://doi.org/10.1162/jocn_a_00329 Zendel, B. R., & Alain, C. (2014). Enhanced attention-dependent activity in the auditory cortex of older musicians. Neurobiology of Aging , 35 (1), 55–63. https://doi.org/10.1016/j.neurobiolaging.2013.06.022 Zendel, B. R., West, G. L., Belleville, S., & Peretz, I. (2019). Musical training improves the ability to understand speech-in-noise in older adults. Neurobiology of Aging , 81 , 102–115. https://doi.org/10.1016/j.neurobiolaging.2019.05.015 Zhang, L., Fu, X., Luo, D., Xing, L., & Du, Y. (2021). Musical experience offsets age-related decline in understanding speech-in-noise: Type of training does not matter, working memory is the key. Ear and Hearing , 42 (2), 258–270. Zhang, L., Ross, B., Du, Y., & Alain, C. (2025). Long-term musical training can protect against age-related upregulation of neural activity in speech-in-noise perception. PLOS Biology , 23 (7), e3003247. https://doi.org/10.1371/journal.pbio.3003247 Zhang, L., Wang, X., Alain, C., & Du, Y. (2023). Successful aging of musicians: Preservation of sensorimotor regions aids audiovisual speech-in-noise perception. Science Advances , 9 (17), eadg7056. https://doi.org/10.1126/sciadv.adg7056 Tables Table 1 Participant Demographic and Neuropsychological Data Raw Scaled Raw Scaled Demographics Age (years) 72.08 (6.58) – 71.16 (6.08) – .313 Education (years) 17.73 (2.51) – 16.60 (2.47) – .819 Sex (F:M) 16:10 – 13:12 – .426 TICS-m 37.19 (2.65) – 36.56 (3.70) – .344 MoCA 27.46 (1.86) – 27.20 (1.53) – .360 biPTA Threshold (dB HL) 18.82 (8.08) – 19.61 (7.28) – Musical Abilities MET Melody 39.54 (5.38) – 34.92 (4.62) – 18.41 a MET Rhythm 39.88 (4.50) – 36.80 (5.16) – 2.43 MET Total 79.42 (8.40) – 71.72 (8.78) – 15.11 a Estimates of IQ WASI Vocabulary 67.08 (5.62) 13.19 (1.83) 66.12 (8.59) 12.60 (2.36) .424 WASI Matrix Reasoning 27.15 (2.87) 15.12 (1.63) 26.60 (2.42) 14.72 (1.46) .394 Memory CVLT-3 Learning 44.19 (10.15) 10.69 (2.65) 46.80 (11.63) 11.36 (3.15) .353 CVLT-3 Short Delay FR 9.81 (3.49) 11.38 (3.36) 9.60 (3.67) 11.04 (2.26) .296 CVLT-3 Long Delay FR 10.42 (3.40) 11.12 (3.05) 10.16 (3.26) 10.80 (2.96) .297 CVLT-3 Recognition Discriminability 2.94 (0.81) 11.27 (3.12) 2.76 (0.67) 10.56 (2.42) .393 WAIS-III Digit Symbol IL FR 7.38 (1.17) 9.96 (1.28) 7.48 (1.08) 10.20 (1.08) .295 WAIS-III Digit Symbol IL PR 11.85 (4.07) 10.27 (1.51) 13.20 (4.36) 10.60 (0.87) .355 Language D-KEFS VF Letter Fluency 47.58 (11.42) 13.96 (2.88) 51.08 (13.32) 14.60 (3.59) .344 D-KEFS VF Category Fluency 44.96 (8.58) 14.42 (3.06) 43.48 (6.47) 13.92 (2.48) .288 D-KEFS VF Category Switching 14.12 (3.54) 12.62 (4.09) 14.52 (3.11) 12.88 (3.67) .286 Executive Functioning, Attention, Processing Speed WAIS-III Digit Symbol Coding 73.23 (13.17) 14.46 (2.16) 66.68 (12.42) 13.16 (2.25) 1.68 WAIS-IV Digit Span Forward 10.73 (2.20) 11.38 (2.97) 10.56 (2.33) 11.16 (2.95) .289 WAIS-IV Digit Span Backward 9.15 (2.07) 11.62 (2.53) 9.64 (1.93) 12.20 (2.38) .377 WAIS-IV Digit Span Sequencing 9.12 (1.68) 12.35 (2.38) 9.48 (2.26) 12.68 (2.82) .306 Trail-Making Test A 23.42 (7.90) 14.73 (2.63) 25.36 (9.38) 14.08 (3.17) .347 Trail-Making Test B 55.08 (22.61) 14.23 (3.02) 63.40 (25.49) 13.48 (2.80) .397 Mood GAD-7 1.23 (1.21) – 1.08 (1.73) – .385 PHQ-8 1.77 (2.12) – 1.44 (1.61) – .324 Note. TICS-m = modified Telephone Interview of Cognitive Status; MoCA = Montreal Cognitive Assessment; biPTA = averaged bilateral pure-tone audiometric threshold from 250 to 2000 Hz; dB HL = decibels Hearing Level; MET = Musical Ear Test; WASI = Wechsler Abbreviated Scale of Intelligence; WAIS-III, WAIS-IV = Wechsler Adult Intelligence Scale, 3 rd edition, 4 th edition; CVLT-3 = California Verbal Learning Test, 3 rd edition; IL = Incidental Learning; FR = Free Recall; PR = Paired Recall; D-KEFS = Delis-Kaplan Executive Functioning System; VF = Verbal Fluency; GAD-7 = Generalized Anxiety Disorder 7-item Scale; PHQ-8 = Patient Health Questionnaire 8-item Scale. B 10 = Bayes factor for independent-samples t- tests performed on scaled scores when available; a = Indicates strong evidence in favour of the alternative hypothesis. Table 2 Summary of Cluster-Based Statistics: Standard vs. Deviant Waveforms of the AMMT Oddball Phase [1]¿p#1 Older Musicians 1 FP1, AF7, AF3, F1, F3, F5, F7, FC5, FC3, FC1, C1, C3, CP1, CPz, FPz, FP2, AF8, AF4, AFz, Fz, F2, F4, F6, FC6, FC4, FC2, FCz, Cz, C2, C4, CP2 FC1 8.92 213-393 <.001 2 T7, TP7, P3, P5, P7, P9, PO7, PO3, O1, Iz, Oz, POz, FT8, C6, T8, TP8, CP6, P4, P6, P8, P10, PO8, PO4, O2, PO9, PO10, TP9, TP10, FT10 PO7 -7.39 225-424 <.001 3 AF7, AF3, F1, F3, F5, FC5, FC3, FC1, C1, C5, FPz, FP2, AF8, AF4, AFz, Fz, F2, F4, F6, F8, FT8, FC6, FC4, FC2, FCz, Cz, C2, C4 F2 8.52 551-713 <.001 4 T7, TP7, CP5, P3, P5, P7, P9, PO7, PO3, O1, Iz, Oz, POz, TP8, P6, P8, P10, PO8, O2, PO9, PO10, TP9, TP10 P5 -6.01 549-697 <.001 5 F1, F3, FC3, FC1, C1, C3, AF8, AF4, AFz, Fz, F2, F4, F6, F8, FC6, FC4, FC2, FCz, Cz, C2, C4, C6 FC4 -6.14 162-193 <.001 6 T7, TP7, P7, P9, PO7, O1, Iz, Oz, POz, P8, P10, PO8, PO4, O2, PO9, PO10, TP9, TP10 P10 5.90 166-193 <.001 7 F1, F3, FC3, FC1, C1, C3, AF8, AF4, AFz, Fz, F2, F4, F6, FC4, FC2, FCz F1 -6.24 430-469 .003 Older Non-Musicians 1 FP1, AF7, AF3, F1, F3, F5, FC5, FC3, FC1, C1, C3, FPz, FP2, AF8, AF4, AFz, Fz, F2, F4, F6, F8, FC6, FC4, FC2, FCz, Cz, C2, C4 F2 8.33 543-689 <.001 2 C5, T7, TP7, CP5, CP3, P1, P3, P5, P7, P9, PO7, PO3, O1, Iz, Oz, POz, Pz, C4, C6, T8, TP8, CP6, P6, P8, P10, PO8, PO4, O2, PO9, PO10, TP10 P8 -6.66 516-648 <.001 3 FP1, T7, TP7, P5, P7, P9, PO7, PO3, O1, Iz, Oz, FPz, FP2, AF8, AF4, AFz, F2, F4, F6, F8, FT8, FC4, FC2, C2, C4, C6, T8, TP8, P6, P8, P10, PO8, O2, PO9, PO10, TP10, FT10 T7 -5.15 324-469 <.001 4 C1, TP7, CP1, P1, P3, P5, P7, P9, PO7, PO3, O1, Iz, Oz, POz, Pz, CPz, FCz, Cz, TP8, P2, P4, P8, P10, PO8, PO4, O2, PO9, PO10, TP9, TP10 PO10 7.30 160-244 <.001 5 FP1, AF7, AF3, F1, F3, F5, FC5, FC3, FC1, C1, C3, CP1, CPz, FPz, FP2, AF8, AF4, AFz, Fz, F2, F4, F6, F8, FT8, FC6, FC4, FC2, FCz, Cz, C2, C4, C6 F4 -11.16 160-209 <.001 6 FP1, AF3, F1, F3, F5, FC5, FC3, FC1, C1, CP1, CPz, FPz, FP2, AF8, AF4, AFz, Fz, F2, F4, F6, FC6, FC4, FC2, FCz, Cz, C2 F1 5.66 320-396 <.001 Difference Waveforms 1 FP1, AF3, F1, F3, FC3, FC1, FPz, FP2, AF4, AFz, Fz, F2, F4, FC2, FCz AFz -4.50 301-334 0.044 Note: Musicians: Clusters 1 and 2 identified the presence of the MMN and its polarity reversal; clusters 3 and 4 identified the presence of the LDN and its polarity reversal, respectively; clusters 5 and 6 identified pre-MMN frequency-specific amplitude differences between conditions in the latency of the Auditory P2 (see Figure 1); cluster 7 identified a short frontocentral positivity that may represent the oddball P300. Non-musicians: Clusters 1 and 2 identified the presence of the LDN and its polarity reversal, respectively; clusters 3 and 4 identified the presence of the LDN and its polarity reversal, respectively; clusters 4 and 5 identified pre-MMN frequency-specific amplitude differences between conditions in the latency of the Auditory P2 and its polarity reversal (see Figure 1); clusters 3 and 6 identified the mid-to-late portion of the MMN. Difference Waveforms: Cluster 1 identified the time window in which MMN amplitude was statistically significant between groups. Spurious spatiotemporal clusters are not reported. Table 3 GLMM Parameters of AMMT Test Phase Hit Rates [1]¿p#1 Intercept 1.21 0.21 5.68 < .001 *** Group 0.19 0.30 0.61 .540 Condition (Foil vs. Target) 0.02 0.56 0.04 .966 Condition (Lure vs. Target) -2.08 0.36 -5.82 < .001 *** Trial -0.14 0.06 -2.32 .021 * Age 0.07 0.09 0.76 .447 Group × Condition (Foil) 3.73 0.98 3.79 < .001 *** Group × Condition (Lure) 0.23 0.51 0.45 .653 Group × Trial -0.14 0.09 -1.48 .138 Note. Non-musician was set as the reference for Group, and Target was set as the reference for Condition. Trial (1 to 60) and age variables were z-scored. Table 4 LMM Results of AMMT Test Phase Cumulative Hit Rates per Stimulus Type Intercept (Standard, Non-Musician) -0.229 0.204 2928 -1.13 0.260 Group (Musician vs. Non-Musician) -0.077 0.288 2928 -0.27 0.788 Probe Position 0.780 0.039 2928 19.86 <.001 Stimulus: Foil vs. Standard 0.413 0.288 2928 1.44 0.151 Stimulus: Lure1 vs. Standard 0.517 0.288 2928 1.80 0.073 Stimulus: Lure2 vs. Standard 0.576 0.288 2928 2.00 0.046 Stimulus: Lure3 vs. Standard 0.347 0.288 2928 1.20 0.229 Stimulus: Deviant vs. Standard 0.141 0.288 2928 0.49 0.624 Group x Probe Position 0.067 0.056 420.5 1.21 0.227 Group x Stimulus: Foil -0.051 0.407 2928 -0.12 0.901 Group x Stimulus: Lure1 0.315 0.407 2928 0.77 0.440 Group x Stimulus: Lure2 0.248 0.407 2928 0.61 0.543 Group x Stimulus: Lure3 0.443 0.407 2928 1.09 0.277 Group x Stimulus: Deviant -0.075 0.407 2928 -0.18 0.855 Probe Position x Stimulus: Foil -0.130 0.046 2928 -2.81 0.005 Probe Position x Stimulus: Lure1 -0.505 0.046 2928 -10.88 < .001 Probe Position x Stimulus: Lure2 -0.435 0.046 2928 -9.37 < .001 Probe Position x Stimulus: Lure3 -0.489 0.046 2928 -10.54 < .001 Probe Position x Stimulus: Deviant -0.023 0.046 2928 -0.49 0.624 Group x Probe Position x Stimulus: Foil 0.235 0.066 2928 3.59 < .001 Group x Probe Position x Stimulus: Lure1 0.002 0.066 2928 0.04 0.971 Group x Probe Position x Stimulus: Lure2 -0.035 0.066 2928 -0.53 0.595 Group x Probe Position x Stimulus: Lure3 0.034 0.066 2928 0.51 0.608 Group x Probe Position x Stimulus: Deviant -0.079 0.066 2928 -1.20 0.230 Note. Non-musician was set as the reference for Group, and Standard was set as the reference for Stimulus. Figure Captions Figure 1. a) Grand-averaged standard and deviant waveforms of the oddball phase of the AMMT for musicians ( n = 26) and non-musicians ( n = 25). P1 and N1 components are identified. Waveforms at the frontocentral midline (FCz) and right lateral parietal-occipital electrode sites (PO8) show deviance-related difference; b) Difference waveforms and scalp topographies for the MMN (averaged between 200-400 ms) and LDN (averaged between 550 and 700 ms) for each group. Shaded regions represent time windows for the MMN and LDN shown over the FCz site; waveforms at the PO8 site show the polarity reversal of both components. Figure 2. A) Grand-averaged source reconstructions for each group of the standard-deviant difference waveform, averaged between a conservative 250 to 350 ms time window. B) t -value contrast between groups showing left-lateralized enhanced source activity in older musicians. C) Spatial distribution of the cluster identified from permutation testing as statistically significantly between groups. Figure 3. Grand-averaged event-related potentials of the MST Test phase by group; waveforms are shown from a representative right frontal electrode (F4). a) Target–Foil contrast with scalp topographies showcasing an FN400 modulation averaged over 400 to 600 ms; b) False Alarm Lure–Foil contrast with scalp topographies averaged over 450 to 650 ms. Figure 4. a) Raincloud plots of hit rates on the Auditory MMN Memory Task by group. b) Raincloud plots of d -primes on the Auditory MMN Memory Task by group for target-lure and c) for target-foil discriminability. Figure 5. Hit rates of the AMMT Perceptual Discrimination Task by group, showing greater hit rates for discriminating between similar and dissimilar tone sequences in older musicians than non-musicians. Figure 6. Raincloud plots of a) lure discrimination index and b) old-new recognition performance on the MST by group. Figure 7. Schematic of parallel mediation models testing potential mediators of MMN amplitude and hit rates on the Perceptual Discrimination Task for a) group differences on target-lure discriminability through discriminating similar pairs, and b) on target-foil discriminability through hit rates for dissimilar pairs. Standardized path coefficients are reported. Supplementary Materials Supplementary Table 1 Summary of Cluster-Based Statistics: MST Test Phase Waveforms Old-New Contrast (Old | Targets vs. New | Foils) Older Musicians 1 Fp1, AF7, AF3, F1, F3, FC1, Fpz, Fp2, AF8, AF4, AFz, Fz, F2, F4, F6, F8, FT8, FC6, FC4, FC2, FCz, C2, T8, F10, LO1, LO2, IO1, IO2 AF4 7.12 355-582 <.001 2 C5, T7, CP5, CP3, CP1, P1, P3, P5, P7, P9, PO7, PO3, O1, Iz, Oz, POz, Pz, CP4, CP2, P2, P4, PO4, O2, PO9, TP9 O1 -6.95 402-545 .001 Older Non-Musicians 1 F1, FC1, Fp2, AF8, AF4, AFz, Fz, F2, F4, F6, F8, FT8, FC6, FC4, FC2, FCz, C2, C6, T8, CP6, CP4 FC4 5.30 432-641 <.001 2 PO7, PO3, Iz, Oz, POz, P4, P6, P8, P10, PO8, PO4, O2, PO10, TP10 PO8 -4.78 465-529 .008 False Lure Recognition Contrast (Old | Lures vs. New | Foils) Older Musicians 1 C3, T7, TP7, CP5, CP3, CP1, P1, P3, P5, P7, P9, PO7, PO3, O1, Iz, Oz, POz, Pz, CPz, C2, C4, C6, T8, TP8, CP6, CP4, CP2, P2, P4, P6, P8, P10, PO8, PO4, O2, PO9, PO10, TP9, TP10 O2 -4.77 396-846 <.001 2 F1, F3, FC1, C1, FPz, FP2, AF8, AF4, AFz, Fz, F2, F4, F6, F8, FT8, FC6, FC4, FC2, FCz, C2, C6, T8, TP8, P8, FT10, F10, LO2 F4 5.35 412-639 <.001 3 FP1, AF7, F1, F3, F5, FT7, FC5, FC1, FPz, FP2, AF8, AFz, F6, F8, FC2, FCz, F9, F10, LO1, IO1 FC5 5.94 635-842 .004 4 FP1, AF3, F1, F5, FC5, FC3, FC1, FPz, FP2, AF8, AF4, AFz, F2, F4, F6, F8, FT8, FC6, FC4, FC2, FCz, C2, C4, F10, LO2, IO2 FC6 4.00 170-283 .008 5 TP7, P1, P3, P5, P7, PO7, PO3, O1, Oz, POz, Pz, PO4 O1 -4.43 332-416 .037 Older Non-Musicians 1 F1, AF8, AF4, AFz, Fz, F2, F4, F6, F8, FT8, FC6, FC4, FC2, FCz, C2, C4, C6, T8, CP6 Fz 5.39 504-658 <.001 2 AF4, AFz, Fz, F2, F4, F6, F8, FT8, FC6, FC4, FC2, FCz, C2, C4, C6, T8 FC4 4.45 426-498 .006 3 P7, P9, PO7, PO3, O1, Oz, POz, P4, PO8, PO4, O2, PO9, TP9 Oz -4.50 428-547 .009 4 P7, P9, PO7, PO3, O1, Oz, POz, PO8, PO4, O2, PO9 PO7 -4.38 551-598 .023 Note: Old-New Contrast: Within each group, a pair of clusters identified the frontocentral FN400 and its polarity reversal over parietal-occipital electrodes. False Lure Recognition Contrast: The FN400 was identified within each group. Within the musician group, the FN400 polarity reversal was also identified, as was early visual-evoked responses between conditions due to perceptual differences. Within non-musicians, early and late portions of the FN400 and its polarity reversal were identified. No group difference was identified for the difference waveforms. For ease of interpretation, spurious spatiotemporal clusters (including those exceeding 1000 ms after stimulus onset and overlapping the response time window) are not reported. Supplementary Material File (figure3_mst_fn400 parta.tif) Download 61.79 MB Information & Authors Information Version history V1 Version 1 30 September 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords aging episodic memory mismatch negativity music pattern separation Authors Affiliations Jennifer Bugos 0000-0001-8061-3752 [email protected] University of South Florida Tampa Campus View all articles by this author Ricky Chow 0000-0002-7977-656X York University Rotman Research Institute at Baycrest View all articles by this author Shimin Mo University of Toronto Rotman Research Institute at Baycrest View all articles by this author R. Shayna Rosenbaum York University Rotman Research Institute at Baycrest View all articles by this author Claude Alain University of Toronto Rotman Research Institute at Baycrest View all articles by this author Metrics & Citations Metrics Article Usage 277 views 220 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Jennifer Bugos, Ricky Chow, Shimin Mo, et al. Enhanced Complex Mismatch Negativity and Mnemonic Representations in Older Adult Musicians. Authorea . 30 September 2025. DOI: https://doi.org/10.22541/au.175926445.57433481/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.175926445.57433481/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe910d3d83706eb',t:'MTc3OTI1NjI0Ng=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();