Abstract
Parkinson's disease affects multiple aspects of movement, impacting significantly on everyday tasks. Music is used in interventions for people with Parkinson's, either to pace movements, or as an integral element of activities such as dance. This study explored self-reported vividness of two forms of imagery – motor imagery evoked by music and auditory (including musical) imagery – in people with Parkinson's, and whether and how they use these types of imagery in everyday life. Participants (N = 199) completed: (i) vividness ratings of visual and kinesthetic music-evoked motor imagery, (ii) vividness ratings of auditory imagery, and (iii) ratings and open questions about their everyday use of these types of imagery. While most participants reported experiencing music-evoked motor imagery (with more vivid visual than kinesthetic motor imagery), <20% reported actively using music to support motor imagery in daily activities. In contrast, participants reported a broad range of contexts and uses for musical imagery (imagined music), from supporting movement (e.g., walking or exercise) to emotion regulation, and concentration. Correlational analyses associated vividness of music-evoked motor imagery with an urge to dance and musical training, while the use of musical imagery was associated with singing ability. A minority of participants reported not experiencing either motor or musical imagery, suggesting that interventions based on imagery may not be suitable for all. Nonetheless, even participants with more severe motor symptoms reported experiencing and using both types of imagery, indicating promise for their strategic use at different stages of Parkinson's. Therefore, musical and motor imagery have the potential to support rehabilitation strategies for Parkinson's, either separately or in combination.
Keywords
Introduction
Parkinson's disease is a neurodegenerative condition characterized by movement difficulties, particularly in initiating, pacing, and maintaining the speed and size of movements. Mobility and daily activities are impacted by impairments in walking (gait), balance, and manual dexterity. People with Parkinson's (PwP) also often experience non-motor symptoms such as cognitive impairment, depression, anxiety, apathy, and sleep disturbance. While dopamine replacement medication is currently the standard treatment for Parkinson's, it is not always effective, and there is a growing emphasis on non-pharmacological interventions and self-management of symptoms. Many approaches involve cueing to support movement through visual, auditory, or tactile stimuli, with the rationale that these cues can help to bypass defective basal ganglia circuitry, thereby compensating for difficulties in self-initiation and regulation of movement (Nieuwboer, 2008). Music provides a type of cue that has been incorporated into a range of interventions for PwP. One approach is to use music to facilitate timing, particularly for gait (e.g., Benoit et al., 2014; Thaut et al., 1996), using the beat of the music to entrain actions (Matthews et al., 2020). Alternatively, music may be integral to an activity, such as in dance or singing programs (e.g., Barnish & Barran, 2020; Irons et al., 2021).
The present mixed-methods study used an online survey to explore the relationship between music and two forms of imagery in PwP: motor imagery and auditory imagery. These two forms of imagery can be illustrated by the example of a person listening to marching music to facilitate their walking. First, they might use the marching music to help imagine themselves walking – that is, mentally simulating the movement (motor imagery) to support their physical walking. Second, they might imagine hearing marching music (musical imagery) to help them initiate walking or maintain their walking speed. Multiple studies of music in everyday life demonstrate that people generate music internally, either consciously or non-consciously (which has been termed “involuntary musical imagery” and is commonly referred to as having an “earworm”). As further described below, both motor and musical imagery involve activity in motor areas of the brain and have the potential to facilitate movement therapeutically. However, for imagery-based strategies to be used effectively, it is important to understand whether PwP are still able to generate and experience imagery, particularly those with more severe symptoms. Additionally, vivid motor and musical imagery may be based on prior experience of music or dance (and in the case of motor imagery, also other activities such as sports; Bailes, 2007; Isaac & Marks, 1994), so the role of such experience should also be considered. The effects of Parkinson's and the influence of experience on imagery are both highly relevant to the application of imagery within interventions, since it is possible that interventions based on music may only be suitable for those able to generate imagery or who are already trained or experienced in music or dance. Finally, we aimed to learn from PwP’s own experiences of using imagery as an everyday strategy, to inform future research and intervention development.
Motor imagery is the mental simulation of an action without the production of an overt action (Jeannerod, 1994) and includes both visual and kinesthetic (sensation-based) aspects of the movement. Motor imagery involves activation of brain areas that overlap with those recruited during the execution of movements (Hardwick et al., 2018), which means that it can be effective in the training and rehabilitation of motor skills (Mulder, 2007). Several studies in which PwP rated the vividness of their motor imagery have shown that PwP experience similar vividness levels to those without Parkinson's (e.g., Bek et al., 2022b; Heremans et al., 2011). Qualitative findings have indicated that PwP are able to use motor imagery to support everyday tasks such as getting into bed or opening bottles (Bek et al., 2016, 2022a). Motor imagery has also shown promise as an adjunct to physiotherapy for PwP (Radder et al., 2017) and is proposed to contribute to the benefits of dance in PwP (Bek et al., 2020, 2021a). However, some evidence suggests that PwP may have difficulty in initiating motor imagery or may engage in different strategies for generating motor imagery (e.g., Humphries et al., 2016). Therefore, external stimuli that can elicit motor imagery may be particularly helpful for PwP. Indeed, external visual cues and non-musical auditory cues (e.g., a metronome) have been found to increase the vividness of motor imagery in PwP (Heremans et al., 2012). Moreover, the movements of PwP are facilitated by combined observation (visual input) and imagery of an action (Bek et al., 2019, 2021b). These findings suggest the possibility that music might also provide an effective cue for motor imagery in PwP.
Hearing music commonly evokes motor imagery. Music might be associated with imagery of a specific action (e.g., marching or a particular dance sequence) or might evoke different qualities of movement by imagining oneself moving like a particular type of animal, substance, or object (Bek et al., 2022b). The former (a specific music-action link) is likely to result from prior experience of moving with or producing music, or learned associations (e.g., between hearing marching music and watching people marching; for a review of priming of action by music see Maes et al., 2014). The latter involves simulation of bodily action associated with the qualities and characteristics of the music, which may be shaped by more general imagery of motion evoked by hearing music (e.g., speed and direction; Eitan & Granot, 2006). For example, music that evokes a general upward motion may promote imagery of a lifting or flying movement. Although it has been hypothesized that PwP use music-evoked motor imagery (e.g., Bek et al., 2020), the vividness of motor imagery evoked by music and whether PwP use this as a strategy has not been studied directly. Therefore, our first aim was to explore how vividly PwP experience motor imagery when listening to music, and whether and how they use music to evoke motor imagery in their daily lives. We also explored the relationship between this music-evoked motor imagery and experience in music and dance, since motor imagery is influenced by previously learned movements (e.g., Isaac & Marks, 1994). It has been found in the general population that ratings of groove (the pleasurable urge to move to a rhythm) are increased in individuals who participate more in dance or have greater musical training (Matthews et al., 2019). We hypothesized that the vividness of music-evoked motor imagery would be related to the amount of training in music and dance, as well as the urge to dance when hearing music.
Our second aim was to explore auditory imagery, including musical imagery, in PwP. Auditory imagery is defined as “the introspective persistence of an auditory experience, including one constructed from components drawn from long-term memory, in the absence of direct sensory instigation of that experience” (p. 302, Hubbard, 2010). Musical imagery (Herholz et al., 2012; Zatorre & Halpern, 2005) is a category of auditory imagery that may be voluntary or involuntary (such as with “earworms”, Halpern & Bartlett, 2011). In the general population, musical imagery has been associated with neural activation in many similar areas of the brain compared with external (i.e., heard) music, as well as some different areas including the supplementary motor area (Herholz et al., 2012). Frequency of involuntary musical imagery has been observed to be influenced by musical training (Floridou et al., 2022), which is likely to be linked to recently performing or listening to music (Bailes, 2007).
If PwP are able to use musical imagery, this broadens the scope of using music to support movement by providing a personal “internalized jukebox” of memorized musical cues (Rose et al., 2019). Indeed, there is some behavioral evidence that musical imagery can support movement in PwP. In tasks where people are asked to move in time after a regular beat is removed, there is evidence that PwP use “internalized music” (i.e., remembered and/or imagined music) to initiate or maintain movement (Harrison et al., 2019; Satoh & Kuzuara, 2008). This phenomenon has been described as “carry-over effects” in rhythmic auditory stimulation studies (Thaut et al., 1996) or as “covert synchronization” (Repp, 2005) in timing studies. Additionally, mental singing (i.e., singing in one's mind) has been associated with a larger decrease in gait variability compared with singing out loud for PwP (Harrison et al., 2019).
Given that the pre-supplementary motor area is relatively preserved in early Parkinson's (Tessitore et al., 2014), it has been speculated that musical imagery training for PwP may provide opportunities for engaging compensatory pathways (Schaefer, 2014b). However, the use of internal cueing strategies based on musical imagery has yet to be studied within interventions for PwP. In this context, we investigated how vividly PwP are able to imagine sounds (including music) and whether and how they are already using musical imagery in their daily lives. In regard to factors that might affect musical imagery, we hypothesized that vividness of auditory imagery might decrease with disease duration due to increasing cognitive and/or motor impairment and given that generating auditory imagery draws on recall of music, and overlaps with motor processes (Godøy, 2022; Herholz et al., 2012). We also predicted that participants with more vivid musical imagery and greater singing ability would report greater usage of musical imagery in their daily lives, as they would find it easier and more effective to use this as a strategy.
Materials and Methods
Participants
People with Parkinson's (N = 199) were recruited via Parkinson's UK and existing networks (e.g., mailing lists for those interested in research). Participants provided informed consent electronically at the start of the survey, agreeing to data collection, analysis, and sharing in anonymized form. Participants were aged 20–88 years (M = 65.2; SD = 8.46) and were fluent English speakers. The sample was well-balanced for gender, with 102 female and 97 male participants. Further demographic details are provided in Rose et al. (in press).
Participants reported that they had been diagnosed with Parkinson's disease by a medical professional, with time since diagnosis ranging from 0.5 to 26 years (M = 5.94; SD = 4.63). The Parkinson's Disease Activities of Daily Living Scale (Hobson et al., 2001) was used as a proxy for disease stage: 32 reported no impact of Parkinson's on their daily activities, 128 reported mild impact, 30 moderate impact, 8 high impact, and 1 extreme impact. Quality of life was measured with the Parkinson's Disease Questionnaire (PDQ-8; Jenkinson et al., 1997), and the total score ranged from 0–90.6 (M = 22.8; SD = 18.5). Note that a higher score (out of 100) indicates a lower quality of life. The mean value is between that obtained for PwP at Hoehn and Yahr stages I and II (Jenkinson et al., 1997), reflecting that most participants had mild to moderate symptoms.
All 199 participants completed the section on music-evoked motor imagery and 192 participants completed the auditory imagery questions, but data for the dance section was only available for 145 participants due to a technical error (see Supplementary Materials for demographics of these sub-samples).
Measures
The present study analyzed data from questions about imagery completed by participants as part of a larger survey on the use of music in Parkinson's (Rose et al., in press). The survey was delivered via the Qualtrics online survey platform (Qualtrics, Provo, UT), and the full version is available on the Open Science Framework (osf.io/usgmp/). In the interests of brevity, we describe only the measures and questions relevant to the present study here.
Participants completed demographic questions, the Goldsmiths Musical Sophistication Index (Gold-MSI; Müllensiefen et al., 2014), and other questions about their preferred genres of music and why and how they listen to music. The Gold-MSI is designed to measure musicality in the general population and consists of a general measure of musical sophistication, as well as subscales on (i) musical training, which includes ratings of hours of practice on a musical instrument, and (ii) singing ability, which consists of agreement on a seven-point scale with statements such as “I am able to hit the right notes when I sing along with a recording.”
To introduce music-evoked motor imagery, participants were given the following information: “Most people are able to imagine a movement without actually moving, for example you might be able to close your eyes and imagine what it would look like or feel like to reach out and touch a table in front of you or to stand up out of a chair. Sometimes when people hear music, they imagine themselves moving even while they are still (or before they start moving). We are interested in whether this is something that you experience when listening to music. These next questions are related to this motor imagery or imagined movement.” To measure the vividness of music-evoked motor imagery, participants were then asked: “When you hear music, how vividly can you visualize movements?” (visual motor imagery) and “When you hear music, how intensely can you imagine the sensations of the movement?” (kinesthetic motor imagery). Participants rated their experience on a five-point scale from “no image” to “as clear as a real movement” (visual motor imagery), or “no sensation” to “as intense as a real movement” (kinesthetic motor imagery), adapted from an established measure of motor imagery vividness (Kinesthetic and Visual Imagery Questionnaire KVIQ; Malouin et al., 2007). To measure their use of music-evoked motor imagery, participants then rated their agreement with the statement “I use music to help me imagine movements” on a five-point scale. If they agreed with this statement, they were given an open text box to describe the types of movements they imagine and the music used. They were also asked to rate how vividly they imagine the movements visually and kinesthetically on the same scales as above.
To assess frequency and use of musical imagery, participants were asked to rate “How often do you imagine and/or remember music inside your mind?” on a five-point scale (Always, Most of the time, About half the time, Sometimes, Never) and to describe particular genres or types of music that help them to imagine or remember in an open text box. They then rated 14 different scenarios in which they might use imagined music on a 5-point frequency scale (as above) (Table 3), which were included based on consultations with PwP (Rose et al., 2022b), and an open text box was provided in case participants wished to elaborate.
To measure vividness of auditory imagery in general, participants then completed the Bucknell Auditory Imagery Scale–Vividness (BAIS; Halpern, 2015). This involved rating how vividly they could imagine sounds in different scenarios (e.g., “Consider the beginning of the song ‘Happy Birthday’. The sound of a trumpet beginning the piece.”), on a seven-point scale from “no image” to “as vivid as the actual sound.”
At the end of the survey, participants also completed the Goldsmiths Dance Sophistication Index (Gold-DSI; Rose et al., 2022a), including two subscales that are analyzed here: urge to dance and dance training. The urge to dance subscale involves rating agreement (on a seven-point scale) with five statements, including “When I hear a great track, it just makes me want to dance” and “When I imagine music in my mind, my body wants to move.” The dance training subscale consists of three questions including “I have taken regular dance classes at least once a week for …” (number of years).
Data Processing and Analysis
To constrain the exploration of relationships between different factors, we pre-registered our hypotheses and planned analyses (osf.io/usgmp/; the present study corresponds to research questions 4 and 5). However, we also report some additional relevant analyses that were subsequently identified as important to address our research questions, as indicated below. We also deviated from the planned analyses in the following ways: (i) rather than using the frequency scales for the use of the different imagery types (which may have been interpreted inconsistently between participants), participants were counted as “non-users” (never used) or “users” (all other frequencies); (ii) the auditory imagery scale was split into musical (1, 5, 7, 8, 11, 12, 14) and non-musical (2, 3, 4, 6, 9, 10, 13) items; (iii) non-parametric Spearman correlations were used instead of multiple regression where scores were not normally distributed; these were corrected for multiple comparisons (p < .01). Data analysis was conducted in R (R Core Team, 2013).
Results
Music-evoked Motor Imagery
Music-evoked motor imagery refers to movements that are imagined in response to hearing music. The mean vividness score was 2.74 (SD = 1.37) for visual motor imagery and 2.56 (SD = 1.13) for kinesthetic motor imagery (Figure 1). Participants rated visual motor imagery as significantly more vivid than kinesthetic motor imagery (Z = 3,411.5, p = .005). Twenty-one percent of participants reported experiencing no music-evoked motor imagery, either visually or kinesthetically, while 18% rated both visual and kinesthetic motor imagery as highly vivid (rating 4 or 5 out of 5; Figure 1). As shown in Table 1, high vividness of music-evoked motor imagery (mean of visual and kinesthetic ratings) was not restricted to those with mild Parkinson's motor symptoms.

The number of people with Parkinson's reporting each possible rating of vividness (1 = no image/sensation, 5 = as vivid/intense as real movement) for kinesthetic and visual music-evoked motor imagery.
Vividness and use of music-evoked motor imagery in people with Parkinson's experiencing different levels of impact of motor symptoms on their daily activities.
Overall music-evoked motor imagery vividness (average of visual and kinesthetic) was not normally distributed, so Spearman correlations were used to investigate relationships with other measures (Figure 2). Music-evoked motor imagery vividness correlated positively with the Gold-MSI musical training subscale (rs(194) = .237, p < .001) and Gold-DSI urge to dance subscale (rs(185) = .231, p = .001), but there was no significant relationship with the Gold-DSI dance training subscale (rs(189) = .079, p = .277). Additional exploratory analysis indicated that motor imagery vividness also correlated positively with overall Gold-MSI musical sophistication (rs(140) = .311, p < .001).

The relationship between the vividness of music-evoked motor imagery and dance training (upper left); urge to dance (upper right); music sophistication (lower left); and musical training (lower right).
When asked whether they used music to generate motor imagery, 16% of participants agreed or strongly agreed that they use music to help them imagine movements. Again, this was not restricted to those with milder symptoms (Table 1); in fact, the percentage using music to generate motor imagery was highest in those with moderate symptoms. The participants who used music to generate motor imagery rated their experience of motor imagery as vivid for both visual (mean = 3.44, SD = 1.11) and kinesthetic (mean = 3.25, SD = .95) modalities.
Participants who reported using versus not using music-evoked motor imagery were compared on the music and dance measures (Table 2). Neither Gold-MSI musical training nor Gold-DSI dance training differed significantly between the groups, but the Gold-DSI urge to dance score was significantly higher for participants who used music-evoked motor imagery.
Scores on Gold-MSI and Gold-DSI subscales compared between participants who reported using or not using music-evoked motor imagery in their daily lives.
Qualitative analysis of the comments made by participants reporting the use of music-evoked motor imagery (Figure 3) indicated that participants described imagery of a range of different actions from running and walking (e.g., “I sing nursery rhymes to myself when I am struggling with moving”), and even impossible actions such as flying, to music-related actions such as conducting and dancing. In relation to the music used, the importance of specific genres and songs was highlighted (e.g., “If I am anxiously awaiting results at a hospital, I imagine marching (Radetzsky March), because that helps me perk up and the wait seems shorter”). Figure 3 shows the primary themes (“Range of movements,” “Dance,” and “Music”) and subthemes, and the links between each of these. Multiple types of movements were reported by participants. Dance was prominent in a number of ways; for example, participants described making dance movements (e.g., “dancing in time to the music”) as types of movements, as well as imagining music associated with dancing (i.e., imagining a specific genre of dance music, such as “waltz music”).

Themes from qualitative analysis of how participants used music-evoked motor imagery. The primary themes are depicted in the three shaded boxes and subthemes in the unshaded boxes, and the links between each of these are illustrated.
Overall Auditory Imagery (Including Musical Imagery)
Across all imagined sounds in the BAIS, the mean vividness rating was 3.73 (SD = 1.43), with a range of 1–6.57 (maximum possible rating = 7). The mean was closest to the rating of “fairly vivid” (4) for all sounds, although a small proportion (5.1%) of participants rated “no imagery” in all questions. The ratings are slightly lower than those reported by Halpern (2015) for 74 college students (mean = 5.1; SD = .9, range 2.9–6.9). When different types of auditory imagery were calculated separately, mean vividness ratings were 3.87 (SD = 1.56) for musical imagery and 3.59 (SD = 1.43) for non-musical imagery, both corresponding to “fairly vivid.”
Overall vividness of auditory imagery did not correlate significantly with disease duration (rs(188) = −.083, p = .254) or cognitive impairment as indexed by the PDQ-8 (rs(188) = −.042, p = .565; exploratory analysis). However, additional exploratory analysis revealed a significant correlation between overall vividness of auditory imagery and vividness of music-evoked motor imagery (rs(190) = .338, p < .0001; Figure 4).

The relationship between the vividness of auditory imagery and left-to-right (a) vividness of music-evoked motor imagery, (b) cognition (from PDQ-8), (c) years since diagnosis.
Use and Experience of Musical Imagery
When rating how often PwP imagined music, 10% reported “never,” 46% “sometimes,” 16% “about half of the time,” 22% “most of the time,” and 5% “always.” Table 3 shows the percentage of participants who reported using musical imagery in each of the 14 scenarios. The mean number of scenarios in which participants reported using musical imagery was 8.79 (SD = 3.95) with a range of 0 to 14. Only 3.6% of participants reported not using or experiencing musical imagery in any scenarios, while 10% of participants reported using musical imagery in all 14 scenarios.
The number of scenarios in which participants used musical imagery correlated significantly with both the vividness of musical imagery (rs(188) = .342, p < .0001) and the Gold-MSI singing ability subscale (rs(138) = .275, p = .001), as shown in Figure 5.
The range of uses and scenarios of musical imagery was echoed in the qualitative analysis of the open comments (Figure 6). This also emphasized the importance of memories and specific songs or genres of the music that is imagined, for example “I sing music in my head to link me to other times” and “I think the Beatles [sic] music takes me back to when I was a boy.” Other examples of open comments relating to the most popular scenarios (Table 3) included “Folk songs tend to stick in my head and trigger me to sing along,” “Organ music – Bach, Buxtehude. They can all help with relaxation,” and “I imagine music to energize me and help me exercise.” The qualitative analysis of musical imagery use began with the identification of primary themes (those that occurred most frequently: exercise/sport, walking, daily routine, singing/humming, memory, emotion, Parkinson's, and music), and then identified sub-themes that occurred multiple times in relation to the primary themes (Figure 6). Some sub-themes related to more than one primary theme; for example, one participant associated a memory of music with a time before they were diagnosed with Parkinson's: “Going to a Mike Oldfield concert at the Albert Hall. Took me to a place where I had no problems especially no Parkinson's.” This response contributed to formation of the “reminiscence” sub-theme, which is part of both the “memory” and “Parkinson's” primary themes. These kinds of associations informed how themes and sub-themes were connected in Figure 6.

The relationship between the number of scenarios in which participants reported using musical imagery and vividness of musical imagery (left panel), singing ability (right panel).

Themes from qualitative analysis of how participants used musical imagery. The primary themes are depicted in solid-color boxes and the sub-themes in white boxes, joined to the main themes with arrows. The themes may be grouped into higher level domains: motor actions, everyday life, cognition, emotion, and music.
Percentage of participants who endorsed using musical imagery in each scenario, presented in descending order.
Discussion
This study used an online survey to explore self-reported vividness of music-evoked motor imagery (imagined movement while listening to music) and auditory imagery (imagined music and other sounds) in PwP, as well as whether and how participants used these types of imagery in their daily lives.
Motor Imagery and Music
Music-evoked motor imagery was reported as being at least moderately vivid by the majority of PwP. Visual motor imagery was rated as more vivid than kinesthetic motor imagery, revealing a similar pattern to ratings of general (i.e., not music-evoked) motor imagery as measured by the KVIQ in PwP (Bek et al., 2022b; Readman et al., 2023) and in the general population (e.g., Malouin et al., 2007). This finding also provides some face validity for our novel rating of music-evoked motor imagery, suggesting that participants were able to rate their experiences of this type of imagery in a similar way to general motor imagery. Nevertheless, the questions about music-evoked motor imagery were quite broad, and future work should consider measuring ratings of imagery for specific actions. Further research could also attempt to disentangle the frequency of use of motor imagery generated spontaneously by music (e.g., walking imagined from marching music) versus using music as a context for endogenously generating motor imagery (e.g., gentle classical music to imagine graceful movement). Moreover, while the present study investigated vividness, the ability to control or manipulate imagery (Kraeutner et al., 2020) should also be considered as this is likely to be critical to its therapeutic application.
We hypothesized that the vividness of music-evoked motor imagery would be related to prior experience in music and dance, and found that vividness was positively associated with a greater urge to dance. The urge to dance captures the sense of compulsion to move in response to music, so it is perhaps to be expected that this would be associated with imagined movements evoked by listening to music. In contrast, formal dance training did not correlate with music-evoked motor imagery vividness. It should be noted that the level of dance training was relatively low in our sample of participants (Rose et al., in press), which might explain the non-significant correlation. More importantly, the absence of a strong relationship indicates that formal dance training is not a prerequisite to this type of imagery. Vividness of music-evoked motor imagery was also associated with musical training and musical sophistication. This fits with higher ratings of groove – the pleasurable urge to move to a rhythm – being found for musicians in the general population (Matthews et al., 2019).
As noted in the introduction, greater dance participation is also linked to higher ratings of groove in the general population (Matthews et al., 2019). Although the directionality of the relationship between dance and groove is unclear, the possibility of increasing the urge to dance in PwP through participation in music or dance should be investigated longitudinally. Nevertheless, dancing is an activity in which the use of motor imagery can be promoted and facilitated among PwP (Bek et al., 2021a, 2022a, Fontanesi & DeSouza, 2021). Imagery is intrinsic to dance (Bläsing et al., 2012), and dance training has been associated with increased use of imagery (Nordin & Cumming, 2008). Therapeutic effects of dance on motor and non-motor symptoms have been found in PwP (Bek et al., 2020; Carapellotti et al., 2020), and recent findings indicated that the use of both motor and analogy/metaphor imagery, as well as singing, were associated with increased self-reported benefits of home-based dance participation in PwP (Bek et al., 2021a). Therefore, either within dance or in other contexts, listening to music could be used as an external auditory stimulus to support the generation and practice of motor imagery, in a similar manner to the use of videos as an external visual stimulus in action observation therapies (Bek et al., 2021b).
Less than one fifth of PwP reported using music to help them imagine movements in their daily lives, indicating that it may be useful as a strategy and that there is considerable scope to increase the number of people who use this strategy. This fits with recent findings from a pilot study of a combined action observation and motor imagery intervention, which indicated the need for educational materials to support PwP's understanding and use of motor imagery (Bek et al., 2021b). Although caution is needed due to the imbalance of group sizes, PwP who reported using music to support motor imagery in the present study reported a stronger urge to dance than those who did not use music in this way. The qualitative data indicated a range of uses of music to evoke motor imagery, including imagery of some specific music-related movements (dance, playing an instrument) but also other actions (running, flying). The role of familiarity in music-evoked motor imagery has been highlighted in previous work. For example, in a case study of an experienced break-dancer, activation of the primary auditory cortex was decreased, and activation of the supplementary motor area was increased, when imagining dance while listening to familiar compared with unfamiliar music (Olshansky et al., 2015). Moreover, in a longitudinal neuroimaging study of expert ballet dancers imagining dancing to music, activity in brain regions associated with auditory and motor processing decreased over time as the music became more familiar and more strongly associated with the learned dance choreography, suggesting increased neural efficiency (Bar & DeSouza, 2016). An important question for future research is whether it is feasible for PwP to develop new associations between music and specific actions through training, which they might be able to capitalize on in their daily lives.
Importantly, vivid music-evoked motor imagery was not restricted to those with milder symptoms, which suggests that it could be a useful therapeutic tool across disease stages. Indeed, the proportion of participants using imagery as a strategy was highest among those with moderate symptoms, which could relate to a greater need for movement-related strategies as the condition progresses.
Auditory and Musical Imagery
PwP reported experiencing moderately vivid auditory imagery, both musical and non-musical. As noted in the results, the vividness ratings on the BAIS were slightly lower than those reported by Halpern (2015) for college students. This could reflect a Parkinson's-related decline in auditory imagery vividness, or it might reflect the fact that no participants in Halpern's study experienced a complete absence of auditory imagery. Self-reported vividness of auditory imagery did not correlate with disease duration in the present study. This suggests that even PwP with more significant motor symptoms are capable of generating auditory imagery, indicating the potential to use this therapeutically.
Participants reported a range of uses for musical imagery beyond supporting walking or other types of movement. This finding indicates that, similar to heard music (e.g., Karageorghis et al., 2020), the use of musical imagery should be considered much more broadly than for improving movement in PwP, such as addressing non-motor symptoms including mood and motivation. The range of uses of musical imagery was similar to that reported for listening to music by the same participants (Rose et al., in press) as well as older adults without a neurological condition (Laukka, 2007), although in the present study the use of musical imagery in some contexts was rated more highly than music listening. For example, a greater proportion of participants reported using musical imagery while walking (52%) than reported listening to music while walking (35%; Rose et al., in press). An important element of musical imagery is its flexibility, as it does not rely on the availability of external resources to listen to music. This is also encouraging for the use of mental singing, which has been suggested to support walking in PwP (Harrison et al., 2019; Satoh & Kuzuara, 2008). Musical imagery is also particularly suitable for use during discrete actions or events, such as gait initiation, due to the need for temporal accuracy (Maslivec et al., 2020).
It was striking that 89% of PwP reported the experience of having music “stuck” in their head (a phenomenon known as an “earworm”). Although musical imagery is usually both common and positive in the general population (Beaty et al., 2013), negative involuntary musical imagery may result from everyday exposure, such as hearing advertisements (Williamson et al., 2012). Earworms have also been reported to interfere with sleep quality (Scullin et al., 2021). Therefore, the possible negative consequences of earworms in PwP should be explored further. In particular, the development of strategies that make therapeutic use of musical imagery needs to be approached carefully to avoid irritations associated with earworms and should take an individualized approach. Musical and/or motor imagery should be linked with specific tasks or clinical goals (e.g., initiating movement, improving mood) (McCullough Campbell & Margulis, 2015) to support positive connections and provide a sense of agency (Killingly et al., 2021).
Participants with more vivid musical imagery and higher scores on the singing ability subscale of the Gold-MSI reported using musical imagery in a greater number of scenarios. This could suggest either that more frequent use of musical imagery might increase vividness and singing ability or, conversely, that those who already have better musical abilities tend to use imagery more widely as a strategy. This link between imagery and singing fits with findings in the broader literature of relationships between musical imagery vividness and ability to imitate pitch (Halpern, 2015). Individual differences in musical imagery should be considered in future studies of mental singing in PwP (Harrison et al., 2019; Satoh & Kuzuara, 2008) and measured before and after training. The importance of the songs or type of music imagined during musical imagery was highlighted by the qualitative analysis in the present study, and future work could explore the characteristics of groove and rhythmic complexity of the music imagined by PwP (see Senn et al., 2019). It will also be important to explore whether rhythmic imagery is a critical ingredient for the benefits of musical imagery for action. Indeed, there is overlap between the brain areas involved in musical and motor imagery and those recruited during temporal processing (Schaefer, 2014a).
It is also worth noting that both types of music-related imagery investigated in the present study (music-evoked motor imagery and musical imagery) could be used concurrently in daily life. Imagined music can be used to mentally simulate movements (motor imagery) to support the physical execution of those movements, as indicated by examples described by participants in their open comments. Moreover, the vividness of the two types of imagery was significantly correlated, as is also found in the general population (Floridou et al., 2022). This association may relate to our multisensory perceptual experiences and the co-occurrence of imagery in more than one sensory modality (Floridou et al., 2022). It has also been pointed out that generating motor imagery of music-related actions (e.g., drumming) can lead to volitional musical imagery (Godøy, 2022).
Concluding Remarks
The majority of participants with Parkinson's in this study reported moderately vivid musical imagery and music-evoked motor imagery, both of which were similar across levels of disease severity. Therefore, there is considerable promise for interventions based on these forms of music-related imagery. It should be noted, however, that the participants in this study are likely to be highly motivated with a strong interest in music or dance, so further investigation is needed in a broader sample of people with Parkinson's. It will also be important to develop and incorporate objective measures of these imagery strategies in future studies. Despite the majority reporting the ability to generate music-evoked motor imagery, a smaller proportion of participants (<20%) reported actively using this type of imagery, indicating that further support or education may be needed to facilitate the everyday application of imagery. Additionally, a proportion of participants reported experiencing neither musical nor motor imagery. This means that interventions based on these are unlikely to be universally suitable, but it is possible that imagery could be trained in some of these individuals. In contrast, many participants reported that they were already using musical imagery in a broad range of contexts, highlighting the potential application of musical imagery to both motor and non-motor symptoms. This study provides intriguing insights into and preliminary support for the use of music and motor imagery as part of a toolbox of non-pharmacological rehabilitation strategies for people with Parkinson's.
Supplemental Material
sj-docx-1-mns-10.1177_20592043231197919 - Supplemental material for Vividness and Use of Imagery Related to Music and Movement in People with Parkinson's: A Mixed-methods Survey Study
Supplemental material, sj-docx-1-mns-10.1177_20592043231197919 for Vividness and Use of Imagery Related to Music and Movement in People with Parkinson's: A Mixed-methods Survey Study by Ellen Poliakoff, Judith Bek, Michelle Phillips, William R. Young and Dawn C. Rose in Music & Science
Footnotes
Acknowledgements
We thank Abigail Davies and Susanna Ward, who were involved in various aspects of this project, and the people with Parkinson’s who provided feedback on early versions of the survey. We thank the people with Parkinson’s who participated in the survey and Parkinson’s UK for assisting with recruitment. We are also grateful to Leo Willis, Connie Willis, and Eve Edmonds, who assisted with data collection and cleaning.
Action Editor
Mats Küssner, Humboldt-Universität zu Berlin, Institut für Musikwissenschaft und Medienwissenschaft.
Peer Review
Valentin Begel, McGill University, Department of Psychology Rebecca Schaefer, Leiden University, Institute for Psychology, Health, Medical and Neuropsychology Unit.
Author Contributions
Conception and design of the study (EP, MP, WY, DCR), ethics application and data collection (DCR), quantitative analysis (EP), qualitative analysis (MP), lead on writing (EP), interpretation and contribution to writing (JB, MP, WY, DCR).
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (grant number 10001C_204290/1).
Ethical Approval
This study received ethical approval from the Ethics Committee HSLU Ethics Committee, Lucerne University of Applied Sciences and Arts, Switzerland, in accordance with the Geneva Convention of Human Rights.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
