Chronometric and pupil-size measurements illuminate the relationship between motor execution and motor imagery in expert pianists

Abstract

Recent years have witnessed an upsurge of research interest in motor imagery (MI; sometimes known as mental practice) or the mental simulation of actions without any concomitant bodily movement. While numerous experimental studies have demonstrated the efficacy of MI in improving skilled performance in fields such as music, sport and medical surgery, few to date have investigated the extent to which MI and motor execution share similar cognitive mechanisms. Therefore, to address this gap, the present studies explored the relationship between the executed and imagined movements of expert pianists. Study 1 explored the effects of movement complexity and force on the time required for nine pianists to actually perform and imagine performing a musical composition. Results revealed that although the durations of participants’ imagined performances were longer than those of executed ones, stage-duration variations during execution were mirrored in the stage-duration variations during MI. In Study 2, seven pianists’ pupil-size measurements (obtained using Tobii eye-tracking glasses) were used to explore changes in cognitive effort between executed and imagined piano performance. Results showed that pupil-size measurements during executed and imagined piano playing were similar. The significance of the findings is discussed and some potential new directions for research are identified.

Keywords

chronometry expert musicians functional equivalence motor imagery pupillometry

The term motor imagery (MI) – also called mental practice (MP; Driskell, Copper, & Moran, 1994) – refers to the conscious mental simulation of an action without concomitant bodily movement (Debarnot, Sperduti, Di Rienzo, & Guillot, 2014). This conscious simulation is thought to be constructed in working memory (Munzert, Lorey, & Zentgraf, 2009).

In recent years, the relationship between motor execution and MI has been explored within sport psychology (e.g. see Guillot & Collet, 2008) and cognitive neuroscience (e.g. see Bach, Allami, Tucker, & Ellis, 2014). Whereas researchers in sport psychology have typically evaluated the efficacy of MI/MP training in improving skilled performance, cognitive neuroscientists have considered the theoretical substrates of MI. In this latter regard Jeannerod (1994, 2001) postulated that motor execution and MI are functionally equivalent in the sense that they share key representations of movement at neural and cognitive levels, with similar processes guiding both movement types. Tellingly, some studies have demonstrated similar neural substrate for motor execution and imagery (for review, see Hétu et al., 2013).

Historically, research on functional equivalence has used the mental chronometry paradigm (Moran, Guillot, MacIntyre, & Collet, 2012). This paradigm assumes that if executed and imagined actions rely on similar motor representations, then their temporal organisation should also be similar. Hence, there should be a close correspondence between the time required to mentally perform a given action and that required for its actual execution. Such temporal congruence between movement execution and imagery has been demonstrated in a variety of contexts (for reviews, see Guillot & Collet, 2005; Guillot, Hoyek, Louis, & Collet, 2012). Nevertheless, a number of intervening variables affect this relationship. Among these variables are level of expertise (Reed, 2002), task complexity (Calmels, Holmes, Lopez, & Naman, 2006; Decety, Jeannerod, & Prablanc, 1989), added mass (Munzert, Blischke, & Kruger, 2015), duration of task (Grealy & Shearer, 2008) and MI instructions (Guillot & Collet, 2005). For example, in a study of springboard divers, Reed (2002) found that temporal congruence was a function of the diver’s level of expertise. She proposed that temporal congruence between executed and imagined movements may be related to the level of cognitive effort required for the task. Classically, Fitts and Posner (1967) suggested that as a person’s skill-level increases, the planning and execution of a movement becomes more automated and involves lower levels of conscious cognition. Thus, studies have demonstrated that cortical neural activity is reduced and refined in individuals who have engaged in intense practice of a movement (for review, see Debarnot et al., 2014).

Another interesting finding in Reed’s (2002) study was that movement imagery durations increased relative to the durations for actual execution for all divers (expert or novice) as the complexity of the dive increased. Despite Reed’s (2002) seminal study, no research has yet explored the effect of complexity on the temporal congruence between executed and imagined movements among experts. However, Calmels and Fournier (2001) examined the executed and imagined movement times for experienced gymnasts’ complex floor routines. Contrary to Reed’s findings of longer imagery times for difficult movements, Calmels and Fournier (2001) discovered that imagery times were shorter than physical execution times for the entire routine.

In a subsequent study, Calmels et al. (2006) examined the time required by elite junior gymnasts to actually execute and to imagine performing an entire gymnastic vault. They reported that there was no significant difference between the time it took to imagine the entire vault and that to actually execute it. Analysis of the vault stages, however, showed that three out of the four generated temporal incongruence. Interestingly, the imagery durations for vault stages involving the most complex movements were longer than those for the actual performance. This finding suggests that complexity may be a crucial factor in modifying the relationship between movement execution and movement imagery (Guillot & Collet, 2005; Guillot et al., 2012). Calmels et al.’s (2006) finding of temporal congruence between execution and imagery of the entire vault may be influenced by methodological considerations. Specifically, temporal congruence may be the result of a balance between the more lengthy easier stages (which had shorter imagery times than execution times) and the less lengthy difficult stages (which had longer imagery times). Calmels et al. (2006) also stated that longer imagery times for the two more complex stages may have been due to either a different level of processing occurring during MI or to the fact that the more complex stages required greater force to execute. Increased force may have been perceived as an increase in duration (see Decety et al., 1989).

Unfortunately, regarding the second of these possible explanations, there have been inconsistencies in the few studies exploring the effects of force on temporal congruence. Thus, whereas Decety et al. (1989) reported that heavier loads, or movements requiring more force, generally result in longer imagery than execution times, Gentili, Cahouet, Ballay, and Papaxanthis (2004) discovered that an added load resulted in longer executed and imagined movement durations than when performed without a load. In an attempt to explain such contradictions, Slifkin (2008) suggested that inexperience leads people to form incomplete mental representations which may generate longer imagery times because of the increased information-processing necessary to assemble the various components of a movement representation.

Clearly, the preceding review highlights an important unresolved issue – the question of whether or not the complexity or force of a movement affects the temporal relationship between its actual and imagined execution. To address this issue, the present research sought to explore the correspondence between the executed and imagined movements of expert musicians, a novel population in this field. Study 1 investigated the effects of movement complexity and force on expert pianists’ durations to actually perform and to imagine performing a bespoke musical composition. Study 2, using the same task, examined for the first time expert pianists’ pupil size (an index of cognitive effort; see Burge et al., 2013) as they performed the composition. This use of pupillometry as an objective marker of cognitive effort (Bijleveld, Custers, & Aarts, 2009) enabled us to explore the processing resources invested in executed and imagined movement performances. Generally, increases in pupil size occur when tasks require more processing resources (Kahneman, 1973).¹ Study 2 used eye-tracking technology to capture pupil size measurements. Eye-trackers enable precise measurement of indices such as the number and length of eye fixations (see McCormick, Causer, & Holmes, 2012; Heremans, Helsen, & Feys, 2008; Gueugneau, Crognier & Papaxanthis, 2008). Eye-tracking technology captures images of the eye in high-resolution, facilitating precise measurement of pupil size (Klinger, Kumar, & Hanrahan, 2008). The present research is the first to examine pupil size during imagined movement using eye-tracking technology. We expected that changes in pupil size would illuminate the cognitive processes occurring during imagined musical performance. We also expected that if similar processes and mechanisms underlie executed and imagined movements, then this would be reflected in the pupil-size data.

Study 1

The aim of Study 1 was to investigate temporal aspects of the relationship between the executed and imagined movements of expert pianists when performing a specially devised musical composition. Arising from Calmels et al.’s (2006) study, three research questions were addressed. First, how close is the temporal correspondence between real and imagined movements when pianists play the entire composition? We predicted that the time it takes expert pianists to perform an entire musical composition would be similar for execution and imagination. Second, does the complexity of a movement exert similar effects on real and imagined duration? Finally, what effects do instructions to use force (fortissimo) have on real and imagined duration when playing a musical composition? We predicted that movement complexity and force would alter the temporal relationship between the two movement types and lead to longer imagined movement durations than those for executed movements.

The research was reviewed and approved by the University College Dublin Ethics Committee.

Method

Design

Two repeated measures 2 x 2 factorial designs were employed. In the first, the two independent variables were “mode” (i.e. executed and imagined piano playing) and “complexity” (i.e. complex and less complex piano playing). In the second, they were mode (as before) and “force” (instructions to play either loudly – fortissimo, or to play softly – pianissimo, which is accomplished on the piano by adjusting playing force, allocating more force or less force, respectively). Pianists completed eight conditions in all: actual performance of complex, less complex, forceful and less forceful stages of a musical composition and imagined performance of the same using MI.

Participants

Ten expert pianists (7 men; 3 women) (M = 38.6 years, SD = 13.71 years, range = 20–61 years) recruited from various Schools and Academies of Music in Ireland participated in the study. Nine of these participants performed piano at an international level and one at national level. As expertise is a function of the number of years of experience (Ericsson, 2006), all participants had a minimum of 10 years piano playing experience (Ericsson, Krampe, & Tesch-Römer, 1993; M = 26.1 years, SD = 12.09 years, range = 10–40 years). Participants were given no information about the experimental hypotheses. All had either normal or corrected to normal vision and all provided informed consent.

Materials

A bespoke musical composition (see Supplementary Material online), comprising extracts from Exercises 42a (bars 1–8), 12 (bars 1–8), 45 (bars 5–12) and 28 (bars 6–13) from Brahms’ 51 Exercises for piano (first published 1893, Schirmer, Ed., Vol. 1600) was used. There were four distinct stages: Stages 1 and 3 represented “less complex” conditions, Stages 2 and 4 represented “complex” conditions. Stages 1 and 2 represented “less force” conditions and Stages 3 and 4 represented “force” conditions. Each stage was matched for length (8 bars). “Complexity” was operationally defined by the texture of the music and the hand-movement patterns involved. “Force” was operationally defined by the level of dynamic required, represented by fortissimo (more force, very loud) and pianissimo (less force, very soft). A speed of 76 crotchet beats per minute was set. At this speed, the time to perform the entire piece of music was 75.8 seconds while to perform a single stage was 18.95 seconds. Written instructions on performance requirements accompanied the musical score (e.g. to be played as a technical piece rather than one involving subjective interpretation/emotion). The music was printed in black on two white A4 pages and during the experimental session mounted on a W29cm x H42cm white board to prevent distraction from the surrounding environment. Participants played on an upright or grand piano. A Cedrus Response Pad (Model RB-530), connected to a laptop computer, was pressed by the participant’s foot to record, using SuperLab Software, the duration of each stage of the composition.

The Vividness of Movement Imagery Questionnaire-2 (VMIQ-2; Roberts, Callow, Hardy, Markland, & Bringer, 2008) was administered to participants before the study began in order to determine their MI abilities.

A post-experimental questionnaire consisting of six Likert-type scale questions and two open-ended questions was used for manipulation checks and for examination of participants’ imagery process.

Procedure

For familiarisation purposes, participants were provided with the musical score and performance instructions approximately four weeks prior to the experimental session. At the beginning of this session, each pianist completed the VMIQ-2. MI scores range from 12 (perfectly vivid feeling of movement) to 60 (no image of movement at all). As a cut-off, scores of less than 30 were required (indicating reasonably clear and vivid imagery overall) for participation and were achieved by all participants (M = 19.9, SD = 6.06). Prior to the executed piano performance, participants warmed up without time limits. This warm-up facilitated their adjustment to the foot response pad. Participants were then reminded of the performance instructions. Then, they were asked to perform the piece of music once as a practice. Participants were seated on a piano stool in front of the piano. The response pad was placed in front of the right piano sustaining pedal to maintain ecological validity. Participants were instructed to include a foot tap (response pad button press) at the onset of every stage of the piece of music and also at the end of the piece. They were further instructed to look straight ahead, focusing on the music notation. If necessary, they could look down briefly at the piano before re-focusing on the notation as quickly as possible. Participants were asked to maintain the set metronome speed. SuperLab recorded all responses and a metronome tempo of 76 crotchet beats per minute was given for 5 seconds immediately prior to the performance. Further practice trials were permitted if requested or necessary (e.g. if extra foot taps mistakenly occurred).

Following the practice trial, participants proceeded with the experimental executed performance using the same procedure. Next, participants proceeded with the imagined performance. Again they followed the same procedure outlined above. They also maintained the exact seating and posture position as in the executed performance with the only exception being that their hands were placed palms down on corresponding thighs rather than on the keyboard. Participants were instructed to feel themselves playing the piano (using MI) rather than simply visualising or hearing themselves playing. Participants’ hands/fingers were observed and visually recorded throughout to ensure they did not move. On completion of the experiment, participants undertook an imagery process questionnaire with manipulation checks.

Results

The mean duration required by participants either to play or to imagine playing an entire musical composition on the piano – with times also recorded for each of its four component stages – was analysed. As there were two complex, two less complex, two forceful and two less forceful stages, the mean movement duration across each pair of stages was calculated prior to main analyses. The means and standard deviations are provided in Table 1.

Table 1.

Mean numbers of seconds for whole, complex, less complex, forceful and less forceful executed and imagined movements, Study 1.

	Executed		Imagined
Stage	M (SD)	95% CI	M (SD)	95% CI
Whole	71.185 (3.051)	[68.840, 73.530]	81.099 (7.350)	[75.449, 86.749]
Complex	17.583 (0.826)	[16.948, 18.218]	20.044 (1.934)	[18.557, 21.531]
Less Complex	18.010 (0.848)	[17.359, 18.663]	20.505 (1.777)	[19.139, 21.871]
Force	17.316 (0.896)	[16.627, 18.005]	20.052 (1.742)	[18.713, 21.391]
Less Force	18.276 (1.084)	[17.443, 19.109]	20.497 (2.110)	[18.875, 22.119]

Note. CI = confidence interval.

Parametric statistics were used. Participant number 10 was an outlier in some of the conditions and therefore their data was omitted from all analyses – thereby reducing the sample size to 9 participants.

In order to minimise experiment-generated error, the alpha value for post-hoc paired samples t-tests was set at p < .008, using Bonferroni correction (an alpha of .05 was divided by 6 – the number of comparisons made).

Data for whole movements was analysed using a one-tailed paired sample t-test. Results revealed that on average, participants executed the whole movement (M = 71.185 s, SD = 3.051) significantly quicker than they played it in their imagination (M = 81.099 s, SD = 7.350), t(8) = -3.783, p < .05.

In order to investigate the effects of complexity and force on timing, two-way repeated measures analyses of variance (ANOVA) were completed. Mean durations for complexity were examined with a 2 (mode: executed and imagined) x 2 (complexity: complex and less complex) ANOVA. This ANOVA revealed a main effect of mode, F(1,8) = 14.313, p = .005, and also a main effect of complexity, F(1,8) = 7.481, p = .026, on the number of milliseconds for movement. However, there was no significant interaction, F(1,8) = .018, p = .897. A series of post-hoc paired samples t-tests revealed that on average, participants’ executed piano playing of the complex movements (M = 17.583 s, SD = 0.826) was significantly quicker than was their imagined playing of them (M = 20.044 s, SD = 1.934;), t(8) = -3.596, p < .008. On average, participants also executed less complex movements (M = 18.011 s, SD = 0.848) significantly more quickly than they did when imagining playing them (M = 20.505 s, SD = 1.777), t(8) = -3.838, p < .008. No significant difference was found between the mean duration required for execution of complex movements (M = 17.583, SD = 0.826) and that required for execution of less complex movements (M = 18.010, SD = 0.848), t(8) = -1.853, p > .008. Similarly, no differences were evident between the times required for imagined playing of complex (M = 20.044, SD = 1.934) and less complex (M = 20.505, SD = 1.777) movements, t(8) = -2.558, p > .008.

Mean durations for force were examined with a 2 (mode: executed and imagined) x 2 (force: with force and with less force) ANOVA. This analysis revealed a main effect of mode on the duration of movement, F(1,8) = 14.313, p = .005. There was no main effect of force, F(1, 8) = 3.907, p > .05, and no significant interaction, F(1, 8) = 1.447, p > .05. A series of post-hoc paired samples t-tests was conducted to identify where the significant differences lay. Force conditions were included in this as previous analyses of descriptive statistics showed a considerable difference in the mean durations for the force conditions across mode. The t-tests revealed that on average, participants executed forceful movements (M = 17.316, SD = 0.896) significantly more quickly than they imagined them (M = 20.052, SD = 1.741), t(8) = -4.322, p < .008. No significant difference was found between the mean duration required for executed (M = 18.276, SD = 1.083) and imagined (M = 20.497, SD = 2.110) movements with less force, t(8) = -2.997, p > .008. The results showed that the durations for expert pianists to imagine performing the entire musical composition and the complex and forceful stages were longer than those when actually performing them.

Discussion

Overall, the findings of Study 1 revealed that the temporal relationship between executed and imagined performance of the entire composition (complex stages, and forceful stages) yielded longer durations for expert pianists’ imagined performance of a musical composition than for their executed performance. Longer imagery times than execution times for participants’ performance of the entire composition are consistent with Guillot, Collet, and Dittmar’s (2004) finding of longer imagery than execution times for attention-demanding sport skills. However, they do not support Calmels et al.’s (2006) findings of temporal congruence between gymnasts’ durations to execute an entire vault and duration to imagine performing it. Regarding the complexity of the movement, we had predicted that relative to less complex stages, the performance of complex stages of a musical composition would take longer to imagine playing than to actually execute. However, although longer imagery times were observed for complex stages than simpler ones, there was no significant difference between complex and less complex movement durations, whether executed or imagined. Therefore, in Study 1, the level of complexity of the movement did not alter the temporal relationship between executed and imagined movements. Nevertheless, longer imagery times than execution times were observed at all levels of complexity. Our finding of longer imagery times for complex stages is consistent with those from other chronometric studies (Calmels et al., 2006; Decety et al. 1989; Guillot et al., 2004; Reed, 2002). However, the finding of longer imagery times for less complex stages is inconsistent with previous research on less complex movements, for which temporal congruence between executed and imagined movements has been demonstrated (Decety & Michel, 1989; Maruff & Velakoulis, 2000). Regarding the chronometric effects of instructions to use force, the findings revealed that times for imagined piano performance were longer than those for executed performance. These results are consistent with those of Munzert et al. (2015), Decety et al. (1989) and Cerritelli et al. (2000). It should be noted that the pianists had equal levels of experience with the force and less force conditions. Therefore, in relation to our results, Slifkin’s (2008) and Munzert et al.’s (2015) suggestion that incongruent imagined and executed movement durations may be due to a lack of experience (hence, inaccurate mental representations), seems improbable.

Overall, there are two intriguing trends in our results. First, we found that the variation in the average durations of participants’ imagined performances was much greater than that in their executed performances. This difference may reflect, in part, the heterogeneity of participants’ imagery abilities (Guillot et al., 2008). Second, variations in the durations between executed stages were mirrored in the durations between imagined stages. Thus, times for less complex executed and imagined movements were 2.4% and 2.3% longer than times for complex executed and imagined movements, respectively. This close mirroring suggests that – in line with the functional equivalence hypothesis – both movement types share at least some movement representations and underlying cognitive processes during MI. In this latter regard, Study 2 sought to explore the cognitive effort that underlies the playing of actual and imagined piano movements.

Study 2

The results of Study 1 showed that the durations of participants’ imagined performances of all stages of a musical composition were longer than those of their actually executed performances. However, imagery times closely mirrored those during execution, which suggests that similar cognitive mechanisms may operate during both movement types. It has long been known that expert performers have the ability to execute movements automatically – with little or no conscious regulation of them (Fitts & Posner, 1967). Conversely, as MI requires conscious processing (Jeannerod, 2001; Munzert et al., 2009), it is plausible that it generates longer imagery times than executed actions. The main aim of Study 2 (which participants undertook one hour after the completion of Study 1), was to use pupil-size measurements (obtained using an eye-tracking system) to illuminate the role of attentional effort in the relationship between executed and imagined movement. Two research questions were addressed. First, how close is the correspondence between pianists’ pupil-size measurements during real and imagined piano playing? Second, does the complexity of the movement or instructions to use force exert similar effects on pianists’ pupil sizes during real and imagined piano playing? As this study used a novel method, no a priori hypotheses regarding the effects complexity or force on this relationship were tested. However, given the emerging research literature on cognitive pupillometry (e.g. see Beatty & Lucero-Wagoner, 2000; Burge et al., 2013) we expected that pupil-size measures would offer insight into the attentional processes occurring during movement execution and MI.

Method

Participants

The same 10 expert pianists (7 men; 3 women) (M = 38.6 years, SD = 13.71 years, range = 20–61 years) who participated in Study 1 also participated in Study 2. Three participants were removed from analyses as they produced less than 20% of useable eye-data samples, likely due to drooping eyelids or astigmatism. The remaining 7 participants (6 men; 1 woman) (M = 36.43 years, SD = 15.29 years, range = 20–61 years) were given no information about the experimental hypotheses and provided informed consent.

Materials

The same experimental materials that were used in Study 1 were also used in Study 2. Additionally, Tobii eye-tracking glasses, with a sampling rate of 30 Hz (Tobii Technology, Stockholm, Sweden) were used to capture pupil dilation.

Procedure

The procedure was the same as that used in Study 1, except that prior to the practice trial Tobii eye-tracking glasses were fitted and calibrated to nine points. Participants wore the glasses in the same position for the entire experimental session. Consistent room luminance was maintained throughout.

Results

The mean percentages of the average values of the right pupil size measured during calibration were analysed. These data were obtained when participants played and imagined playing an entire musical composition on the piano – with measurements also acquired for each of its four component stages. The means and standard deviations are provided in Table 2.

Table 2.

Mean pupil size for whole, complex, less complex, forceful and less forceful executed and imagined movements, expressed as percentages (%) of the average values of the pupil size measured during calibration, Study 2.

	Executed		Imagined
Stage	M (SD)	95% CI	M (SD)	95% CI
Whole	95.57 (35.39)	[62.84, 128.30]	83.79 (25.06)	[60.61, 106.97]
Complex	95.83 (35.90)	[62.63, 129.03]	84.34 (25.88)	[60.41, 108.28]
Less Complex	95.30 (34.96)	[62.97, 127.63]	83.22 (24.27)	[60.77, 105.67]
Force	96.78 (35.98)	[63.50, 130.06]	83.87 (25.39)	[60.39, 107.35]
Less Force	94.35 (34.82)	[62.15, 126.55]	83.70 (24.77)	[60.79, 106.61]

Note. CI = confidence interval.

Data with z-scores greater than or equal to 3 and less than or equal to -3 were excluded from analyses thereby eliminating extreme outliers (note that about 99% of a distribution falls between these z-score values). Missing values for pupil size (e.g. eye blinks) were recorded as -999.

Chronometric results (i.e. the mean numbers of milliseconds for executed and imagined movements) were consistent with those of Study 1 (see Table 3 for means and standard deviations).

Table 3.

Mean numbers of seconds for whole, complex, less complex, forceful and less forceful executed and imagined movements, Study 2.

	Executed		Imagined
Stage	M (SD)	95% CI	M (SD)	95% CI
Whole	71.095 (3.677)	[67.694, 74.496]	84.201 (9.739)	[75.194, 93.208]
Complex	17.276 (0.606)	[16.716, 17.837]	20.261 (2.230)	[18.199, 22.323]
Less Complex	18.271 (1.641)	[16.753, 19.789]	21.839 (3.152)	[18.924, 24.754]
Force	17.269 (1.199)	[16.160, 18.378]	20.710 (2.133)	[18.737, 22.683]
Less Force	18.278 (1.145)	[17.219, 19.337]	21.391 (3.031)	[18.588, 24.194]

Note. CI = confidence interval.

Analysis of mean pupil-size measurements revealed that those during executed performance of the entire composition (M = 95.57, SD = 35.39) and the complex (M = 95.83, SD = 35.90), less complex (M = 95.30, SD = 34.96), forceful (M = 96.78, SD = 35.98) and less forceful (M = 94.35, SD = 34.82) stages were greater than those during imagined performance of the entire composition (M = 83.79, SD = 25.06) and during the imagined complex (M = 84.34, SD = 25.88), less complex (M = 83.22, SD = 24.27), forceful (M = 83.87, SD = 25.39), and less forceful (M = 83.70, SD = 24.77) stages, respectively (see Figure 1 and Table 2).

Figure 1.

Pupil-size measurements for pianists’ actual and imagined performances of a piece of music. Stage 1 (less complex) of the piece = 1–12 seconds, Stage 2 (complex) = 13–24 seconds, Stage 3 (less complex) = 25–36 seconds and Stage 4 (complex) = 37–48 seconds.

Data for whole movements were analysed using a two-tailed paired sample t-test. There was no significant difference in pupil size between participants’ executed performance of the entire composition (M = 95.57, SD = 35.39) and their imagined performance of the same (M = 83.79, SD = 25.06), t(6) = 2.405, p = .053, d = 0.384.

To investigate the effect of complexity and force on expert pianists’ pupil size during executed and imagined movements, the data for complexity and force were analysed using two-way repeated measures analysis of variance (ANOVA). Mean pupil-size measurements for complexity were examined with a 2 (mode: executed and imagined) x 2 (complexity: complex and less complex) ANOVA. This revealed no significant main effect of mode, F(1,6) = 5.784, p = .053, or stage, F(1,6) = .674, p = .443, and no significant interaction, F(1,6) = .363, p = .569. Mean pupil-size measurements for force were examined with a 2 (mode: executed and imagined) x 2 (force: with force and with less force) ANOVA. This revealed no significant main effect of mode, F(1,6) = 5.781, p = .053, or force, F(1,6) = 4.865, p = .070, and no significant interaction, F(1,6) = 5.862, p = .052. Overall, the results showed that expert pianists’ mean pupil-size measurements during their executed and imagined performances of a composition with complex, less complex, forceful and less forceful stages were similar.

Discussion

The main aim of Study 2 was to use expert pianists’ pupil-size measurements to understand the cognitive effort occurring during imagined performance of a musical composition in relation to that occurring during actual performance of the music. Results revealed that participants’ pupil-size measurements during executed and imagined performance of the composition were similar. Further, neither the complexity nor force of the movement affected pupil size during executed or imagined movements significantly. Similar pupil-size measurements during executed and imagined performances suggests that both movement types share similar mental processes, as pupil dilation is a robust index of attentional effort (Beatty & Lucero-Wagoner, 2000). Although MI requires greater conscious processing than executed movement (Jeannerod, 2001; Munzert et al., 2009), this additional cognitive effort was not reflected in the pupil-size findings during imagery. Possible reasons for this will be discussed in relation to the findings from Study 1 in the general discussion.

General discussion

The present research used a novel approach to investigate the temporal relationship between the executed and imagined movements of expert pianists when performing a musical composition (Study 1) and the cognitive effort underlying these movement types (Study 2). In Study 1, we predicted that the time it takes expert pianists to perform the entire composition would be similar for executed and imagined actions, and that complexity and force would alter this temporal relationship and lead to longer imagined movement durations than those for executed movements. The findings of Study 1 showed that the durations for participants to imagine performing the entire composition and the complex, less complex, forceful and less forceful stages, were longer than the durations to execute them. Although not in line with our prediction, longer imagery times than execution times for performance of the entire composition are consistent with previous research using complex movements (Guillot et al., 2004). Further, findings of longer imagery times for complex stages are also consistent with previous studies (Calmels et al., 2006; Decety et al., 1989; Guillot et al., 2004; Reed, 2002). However, previous research using less complex movements has generally found temporal congruence between execution and imagery times (for reviews, see Guillot & Collet, 2005; Guillot et al., 2012) and therefore findings of longer imagery durations in this instance do not provide support for such studies. Longer imagery times than those for execution of forceful movements are consistent with previous research (Cerritelli et al., 2000; Decety et al., 1989; Munzert et al., 2015).

One possible explanation for the overall longer durations for expert pianists to imagine playing the composition than to actually play it may be the requirement of different levels of processing during the two movement types. MI requires greater conscious processing than movement execution (Jeannerod, 2001; Munzert et al., 2009). Further, it is likely that greater levels of inhibition of a motor command are necessary during imagery (Guillot et al., 2007). Therefore, it was suggested that extra cognitive demands may be present during MI and account for the longer imagery times in this instance.

In Study 2, we expected that pupil-size data would offer insight into the attentional processes occurring during movement execution and MI. The findings revealed that pupil dilation during executed and imagined performance of the musical composition was similar, and that the complexity or force of the movement did not alter this. At first glance, these findings lend support to the argument that executed and imagined movements share similar mechanisms. However, when these findings are combined with those from Study 1, it seems likely that the relationship between executed and imagined movements is more complex than we had expected. In both studies, chronometric measurements for executed and imagined movements revealed that durations to imagine performing the composition were longer than those to actually execute it. Furthermore, both studies showed a close mirroring of executed stage-duration variations in imagined stage-duration variations, for example, both showed longer durations for less complex stages than complex stages. On this basis, it seems reasonable to conclude that similar cognitive processes underlie both movement types. However, although differences between pupil-size measurements during execution and imagery did not reach a statistical significance, measurements for the latter were on average 12% smaller than those during execution (see Figure 1) and therefore are of practical importance. Accordingly, an obvious question arises. Why were average pupil-size measurements during MI in Study 2 smaller if durations were longer? One possible explanation for this apparent paradox concerns the processing demands of the task.

A study by Granholm, Asarnow, Sarkin, and Dykes (1996) which manipulated the load of a digit span recall task (from a low load of 5 digits to an excessive load of 13 digits) found that pupil size increased with greater processing demands. However, this increase was only observed when processing demands stayed within resource limits. When participants’ processing resources were overloaded (i.e. recall of greater than 9 digits) their pupil size began to decrease. By extrapolation, it is possible that in the present study the level and amount of processing required during MI overloaded participants’ cognitive processing resources. Clearly, this hypothesis needs to be tested systematically in future research.

Unfortunately, both of the present studies have methodological limitations that could have affected the findings. Specifically, four problems may be identified. First, owing to the level of expertise required for participation, the sample size (N = 10) was small – although not unusual for research on expert populations (e.g. Hiroki & Shiro, 2012; Schorer, Jaitner, Wollny, Fath, & Baker, 2012). In Study 2, it is entirely possible that potential differences between pupil-size measurements during execution and imagery could not be detected as the small sample size limited statistical discriminative power. Replication would be valuable with a larger sample size. Second, Tobii glasses (with a sampling rate of 30 Hz, i.e. capturing an image of the eye every 33 ms) facilitated the use of a naturalistic task to gather eye-tracking data. Unfortunately, eye trackers vary in their ability to capture pupil-size data and typically range from sampling rates of 60 Hz (capturing an image every 16.6 ms) to 2000 Hz (capturing an image every 0.5 ms; Brisson et al., 2013). Due to the low sampling rate of the glasses used in the present study, we could only examine the relative size of the pupil and not true pupil dilation. Third, although participants stated (in post-experiment questionnaires) that they had found stages two and four of the music complex, they also stated that one and three were not easy. Given these assessments, it is questionable whether there was sufficient difference between the complex and less complex stages. In preparing the stimuli for these studies, ecological validity was one of our key priorities. By manipulating the level of complexity and force through textural/dynamic changes within a single musical composition, such ecological validity was enhanced and a systematic investigation of both variables was facilitated. However, future studies might consider using a number of individual stimuli (rated by expert pianists for level of complexity/force) and testing complexity according to either texture or harmonic language and force according to dynamics or articulation. Finally, although the method of using mean pupil-size measurements as an index of cognitive processing seems robust (Beatty & Lucero-Wagoner, 2000), it is not a fine-grained method of analysis. Accordingly, subtle changes in pupil size during executed or imagined performances may have been masked. A more precise method of analysing the data, such as wavelet analysis (see Lew, Dyre, Soule, Ragsdale, & Werner, 2010), may have been more appropriate as a tool for investigating exactly when pupil measurements during the imagined performance were greater or smaller than those during the executed performance. A key priority for future researchers is to use pupillometry and other eye-tracking metrics (such as fixation length and number) to explore MI in relation to movement execution for different types of complex movements.

The present studies represent an important contribution to MI literature as there has been no published research as yet investigating the temporal functional equivalence between, and the pupil-size measurements of, the executed and imagined complex movements of expert musicians.

Conclusion

The findings of the present studies demonstrate that the relationship between executed and imagined movements is not straightforward. The results of Study 1 revealed a temporal incongruence between the actual performance and imagined performance of a musical composition. On one level, this appears to defy the idea that motor execution and MI rely on similar processes. However, when performance durations were examined more comprehensively there was a close temporal correspondence between the executed stage-duration variations and the imagined stage-duration variations. Consequently, we suggested that longer imagery times may have been due to additional cognitive demands during MI. The results of Study 2 revealed that pianists’ pupil-size measurements during executed and imagined performances of the composition were broadly similar. Overall, the results can be interpreted as supporting functional equivalence – the principle that motor execution and MI share similar mental processes and representations. Although further research is necessary to explore the precise cognitive mechanisms underlying MI, the findings of the present studies support the efficacy of using MI as a tool for developing expertise in skilled music performance.

Footnotes

Acknowledgements

The authors would like to thank the pianists who participated in the studies, Colin Burke (Psychology Laboratory IT Specialist) and the reviewers who offered some excellent advice.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Bach

Allami

B. K.

Tucker

Ellis

(2014). Planning-related motor processes underlie mental practice and imitation learning. Journal of Experimental Psychology: General, 143(3), 1277–1294. doi:10.1037/a0035604

Beatty

Lucero-Wagoner

(2000). The pupillary system. In Cacioppo

J. T.

Tassinary

L. G.

Berntson

G. G.

(Eds.), Handbook of psychophysiology (pp. 142–162). New York, NY: Cambridge University Press.

Bijleveld

Custers

Aarts

(2009). The unconscious eye opener: Pupil dilation reveals strategic recruitment of resources upon presentation of subliminal reward cues. Psychological Science, 20(11), 1313–1315. doi:10.1111/j.1467–9280.2009.02443.x

Brisson

Mainville

Mailloux

Beaulieu

Serres

Sirois

(2013). Pupil diameter measurement errors as a function of gaze direction in corneal reflection eyetrackers. Behavioural Research, 45, 1322–1331. doi:10.3758/s13428–013–0327–0

Burge

W. K.

Ross

L. A.

Amthor

F. R.

Mitchell

W. G.

Zotov

Visscher

K. M.

(2013). Processing speed training increases the efficiency of attentional resource allocation in young adults. Frontiers in Human Neuroscience, 7, 1–7. doi:10.3389/fnhum.201300684

Calmels

Founier

J. F.

(2001). Duration of physical and mental execution of gymnastic routines. Sport Psychologist, 15, 142–150.

Calmels

Holmes

Lopez

Naman

(2006). Chronometric comparison of actual and imaged complex movement patterns. Journal of Motor Behaviour, 38(5), 339–348. doi:10.3200/JMBR.38.5.339–348

Cerritelli

Maruff

Wilson

Currie

(2000). The effect of an external load on the force and timing components of mentally represented actions. Behavioural Brian Research, 108, 91–96. doi:10.1016/S0166-4328(99)00138-2

Debarnot

Sperduti

Di Rienzo

Guillot

(2014). Expert bodies, expert minds: How physical and mental training shape the brain. Frontiers in Human Neuroscience, 8, 1–13. doi:10.3389/fnhum.2014.00280

10.

Decety

Jeannerod

Prablanc

(1989). The timing of mentally represented actions. Behavioural Brain Research, 34, 35–42. doi:10.1016/S0166–4328(89)80088–9

11.

Decety

Michel

(1989). Comparative analysis of actual and mental movement times in two graphic tasks. Brain and Cognition, 11, 87–97. doi:10.1016/0278–2626(89)90007–9

12.

Driskell

J. E.

Copper

Moran

(1994). Does mental practice enhance performance? Journal of Applied Psychology, 79(4), 481–492.

13.

Ericsson

K. A.

(2006). The influence of experience and deliberate practice on the development of superior expert performance. In Ericsson

K. A.

Charness

Feltovich

P. J.

Hoffman

R. R.

(Eds.), The Cambridge handbook of expertise and expert performance (pp. 685–705). Cambridge, UK: Cambridge University Press.

14.

Ericsson

K. A.

Krampe

R. T.

Tesch-Römer

(1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100(3), 363–406.

15.

Fitts

P. M.

Posner

M. I.

(1967). Human performance. Belmont, CA: Brooks/Cole.

16.

Gentili

Cahouet

Ballay

Papaxanthis

(2004). Inertial properties of the arm are accurately predicted during motor imagery. Behavioural Brain Research, 155, 231–239. doi:10.1016/j.bbr.2004.04.027

17.

Granholm

Asarnow

R. F.

Sarkin

A. J.

Dykes

K. L.

(1996). Pupillary responses index cognitive resource limitations. Psychophysiology, 33(4), 457–461. doi:10.1111/j.1469–8986.1996.tb01071.x

18.

Grealy

M. A.

Shearer

G. F.

(2008). Timing processes in motor imagery. European Journal of Cognitive Psychology, 20(5), 867–892. doi:10.1080/09541440701618782

19.

Gueugneau

Crognier

Papaxanthis

(2008). The influence of eye movements on the temporal features of executed and imagined arm movements. Brain Research, 1187, 95–102. doi:10.1016/j.brainres.2007.10.042

20.

Guillot

Collet

(2005). Duration of mentally simulated movement: a review. Journal of Motor Behaviour, 37, 10–20.

21.

Guillot

Collet

(2008). Contribution of the motor imagery integrative model in sport: A review and theoretical investigation of motor imagery use. International Review of Sport & Exercise Psychology, 1(1), 31–44. doi:10.1080/17509840701823139

22.

Guillot

Collet

Dittmar

(2004). Relationship between visual and kinaesthetic imagery, field dependence-independence and complex motor skills. Journal of Psycophysiology, 18(4), 190–198. doi:10.1027/0269–8803.18.4.190

23.

Guillot

Collet

Nguyen

V. A.

Malouin

Richards

Doyon

(2008). Functional neuroanatomical networks associated with expertise in motor imagery. NeuroImage, 41, 1471–1483. doi:10.1016/j.neuroimage.2008.03.042

24.

Guillot

Hoyek

Louis

Collet

(2012). Understanding the timing of motor imagery: Recent findings and future directions. International Review of Sport & Exercise Psychology, 5(1), 3–22. doi:10.1080/1750984X.2011.623787

25.

Guillot

Lebon

Rouffet

Champely

Doyon

Collet

(2007). Muscular responses during motor imagery as a function of muscle contraction types. International Journal of Psychophysiology, 66, 18–27. doi:10.1016/j.ijpsycho.2007.05.009

26.

Heremans

Helsen

W. F.

Feys

(2008). The eyes as a mirror of our thoughts: Quantification of motor imagery of goal-directed movements through eye movement registration. Behavioural Brain Research, 187, 351–360. doi:10.1016/j.bbr.2007.09.028

27.

Hétu

Grégoire

Saimpont

Coll

M-P.

Eugènec

Michonc

P-E.

Jackson

P. L.

(2013). The neural network of motor imagery: An ALE meta-analysis. Neuroscience and Biobehavioural Reviews, 37, 930–949. doi:10.1016/j.neubiorev.2013.03.017

28.

Hiroki

Shiro

(2012). Experts in fast-ball sports reduce anticipation timing cost by developing inhibitory control. Brain and Cognition, 80(1), 23–32. doi:10.1016/j.bandc.2012.04.004

29.

Jeannerod

(1994). The representing brain: Neural correlates of motor intention and imagery. Behavioural and Brain Sciences, 17, 187–245. doi:10.1017/S0140525X00034026

30.

Jeannerod

(2001). Neural simulation of action: A unifying mechanism for motor cognition. NeuroImage, 14, S103–S109. doi:10.1006/nimg.2001.0832

31.

Kahneman

(1973). Attention and effort. Englewood Cliffs, NJ: Prentice-Hall.

32.

Klingner

Kumar

Hanrahan

(2008, March). Measuring the task-evoked pupillary response with a remote eye tracker. In Proceedings of the 2008 symposium on eye tracking research and applications. Savannah, Georgia.

33.

Lew

Dyre

B. P.

Soule

Ragsdale

S. A.

Werner

(2010). Assessing mental workload from skin conductance and pupillometry using wavelets and genetic programming. Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 54(3), 254–258. doi:10.1177/154193121005400315

34.

Maruff

Velakoulis

(2000). The voluntary control of motor imagery: Imagined movements in individuals with feigned impairment and conversion disorder. Neuropsychologia, 38, 1251–1260.

35.

McCormick

S. A.

Causer

Holmes

P. S.

(2012). Eye gaze metrics reflect a shared motor representation for action observation and movement imagery. Brain and Cognition, 80, 83–88. doi:10.1016/j.bandc.2012.04.010

36.

Moran

Guillot

MacIntyre

Collet

(2012). Re-imagining motor imagery: Building bridges between cognitive neuroscience and sport psychology. British Journal of Psychology, 103, 224–247. doi: 10.1111/j.2044–8295.2011.02068.x

37.

Munzert

Blischke

Kruger

(2015). Motor imagery of locomotion with an additional load: Actual load experience does not affect differences between physical and mental durations. Experimental Brian Research, 233(3), 809–816. doi:10.1007/s00221–014–4156-x

38.

Munzert

Lorey

Zentgraf

(2009). Cognitive motor processes: The role of motor imagery in the study of motor representations. Brain Research Reviews, 60(2), 306–326. doi:10.1016/j.brainresrev.2008.12.024

39.

Reed

C. L.

(2002). Chronometric comparisons of imagery to action: Visualising versus physically performing springboard dives. Memory and Cognition, 30(8), 1169–1178.

40.

Roberts

Callow

Hardy

Markland

Bringer

(2008). Movement imagery ability: Development and assessment of a revised version of the Vividness of Movement Imagery Questionnaire. Journal of Sport & Exercise Psychology, 30, 200–221.

41.

Schorer

Jaitner

Wollny

Fath

Baker

(2012). Influence of varying focus of attention conditions on dart throwing performance in experts and novices. Experimental Brain Research, 217(2), 287–297. doi:10/1007/s00221–011–2992–5

42.

Slifkin

A. B.

(2008). High loads induce differences between actual and imagined movement duration. Experimental Brain Research, 185(2), 297–307. doi:10.1007/s00221–007–1154–2

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.

0.00 MB

1.45 MB