- Email: [email protected]
Effects of task complexity on rhythmic reproduction performance in adults
Human Movement Science 32 (2013) 203–213
Contents lists available at SciVerse ScienceDirect
Human Movement Science journal homepage: www.elsevier.com/locate/humov
Effects of task complexity on rhythmic reproduction performance in adults Flora Iannarilli, Giuseppe Vannozzi, Marco Iosa, Caterina Pesce, Laura Capranica ⇑ DiSMUS, University of Rome Foro Italico, Italy
a r t i c l e
i n f o
Article history: Available online 27 February 2013 PsycINFO classification: 3357 Keywords: Rhythmic ability Movement task Rhythm Evaluation Movement complexity
a b s t r a c t The aim of the present study was to investigate the effect of task complexity on the capability to reproduce rhythmic patterns. Sedentary musically illiterate individuals (age: 34.8 ± 4.2 yrs; M ± SD) were administered a rhythmic test including three rhythmic patterns to be reproduced by means of finger-tapping, foot-tapping and walking. For the quantification of subjects’ ability in the reproduction of rhythmic patterns, qualitative and quantitative parameters were submitted to analysis. A stereophotogrammetric system was used to reconstruct and evaluate individual performances. The findings indicated a good internal stability of the rhythmic reproduction, suggesting that the present experimental design is suitable to discriminate the participants’ rhythmic ability. Qualitative aspects of rhythmic reproduction (i.e., speed of execution and temporal ratios between events) varied as a function of the perceptualmotor requirements of the rhythmic reproduction task, with larger reproduction deviations in the walking task. Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction Rhythm is a fundamental component of the organization and the execution of body movements, easily observable during normal daily activities (e.g., walking, running, clapping hands, drumming fingers), and in dance, music, and sport (e.g., rhythmic gymnastics and synchronized swimming) contexts. Theories in cognitive and musical sciences considered the ‘‘dynamogenous’’ effect of rhythm (Fraisse, 1979) and highlighted the role of the body as a mediator for music awareness (Dalcroze, 2008; Leman, 2007). In fact, music emerges from physical movement (Tood, Cousins, & Lee, 2007; ⇑ Corresponding author. Tel.: +39 06.36 73 32 07; fax: +39 06 36 73 33 30. E-mail address: [email protected] (L. Capranica). 0167-9457/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.humov.2012.12.004
204
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
Trainor, 2007) and rhythm is embodied and expressed through movements (Philliphs-Silver & Trainor, 2007), which require the organization of muscular activity to execute a specific chronological order of motor actions (Meinel & Schnabel, 1998; Sakai, Hikosaka, & Nakamura, 2004). Actually, studies on human coordination refer to rhythm as a crucial aspect for movements such as walking (Hausdorff, Peng, Ladin, Wei, & Goldberger, 1995) and tapping (Chen, Ding, & Kelso, 1997, 2001), which fluctuates in relation to higher nervous system centres and/or lower motor neuron control (Golubitsky, Stewart, Buono, & Collins, 1999; Ijspeert, 2008; Ivanenko, Poppele, & Lacquaniti, 2006). The capability to translate an acoustic perception into a corresponding motor behavior is considered a fundamental characteristic of rhythmic ability (Fraisse, Pichot, & Clairouin, 1949), which is defined as the ability to perform a succession of regulated, recurring gross motor events requiring both spatial and temporal accuracy (Zachopoulou & Mantis, 2001). In considering that performances mainly depend upon the rhythmic structure (Gilden, Thornton, & Mallon, 1995), to discriminate the rhythmic ability of different populations, the use of a large range of pitch intervals and several complex rhythms is recommended (Persichini & Capranica, 2004; Trehub & Hannon, 2009). In particular, the ability to accurately process and reproduce rhythmic patterns depends on the interval ratios between events, with spontaneous reproduction of 1:1 and 2:1 rhythmic structures (Drake, 1993; Essens & Povel, 1985) and preferences for small integer ratios (1:2, 1:3, or 1:4) as compared to larger (1:5) or non-integer (in the sense of non-simple) (1:2.5 or 1:3.5) ratios (Drake, 1993; Essens, 1986; Sakai et al., 1999). In general, during learning or rehearsal, non-integer ratios are often distorted and shifts toward simpler ratios are observed (Drake, 1993; Essens, 1986; Essens & Povel, 1985; Sakai et al., 1999; Summers, 1975; Trehub & Hannon, 2009). Furthermore, binary subdivisions are easier to be discriminated and reproduced with respect to ternary subdivisions (Jones, 1987; Povel, 1981; Trehub & Hannon, 2009). To evaluate the individual ability to correctly reproduce a rhythmic pattern, both its quantitative (i.e., number of events) and qualitative (i.e., speed of execution and temporal ratios among events) aspects must be considered (Persichini & Capranica, 2004). In fact, individuals might repeatedly deviate from the intervals between events around a constant value, thereby creating a considerable error in the speed of execution (i.e., overall faster or slower performances), yet producing the prominent rhythmic structure. On the other hand, individuals could correctly reproduce the established total time of the pattern even though the proportionality between intervals of rhythmic structures is not respected (Persichini & Capranica, 2004). Unfortunately, thorough comparison of evidence concerning rhythmic ability is hard because of large evaluation differences such as the use of heterogeneous stimuli (i.e., auditory, visual or audio-visual), procedures (i.e., different forms of coordination, in-phase or anti-phase movements), effectors (i.e., hands, feet, or voice), and rhythmic variables (i.e., number of events, total duration of the pattern, intervals between events) (Drake, 1993; Hennig et al., 2011; Joiner & Shelhamer, 2009; Meeuwsen, Flohr, & Fink, 1998; Persichini & Capranica, 2004; Repp, 2005; Smoll & Palmatier, 1977). A further relevant aspect in rhythmic ability evaluation is related to the movement task requirements for rhythmic pattern reproduction. To generate wide possibilities of motor programs in relation to rhythmic movements, a modular control of limb kinematics based on various feedback and feed-forward parameters has been hypothesized (Ivanenko et al., 2006). Although it has been hypothesized that a general timing process assists the individual in performing movements of different limbs and body parts (i.e., finger, forearm and foot tapping) (Keele & Hawkins, 1982; Keele, Ivry, & Pokorny, 1987), some authors (Getchell, Forrester, & Whitall, 2001; Robertson et al., 1999) argued that specific timing processes are needed when tasks are not dynamically equivalent, such as tapping, clapping, galloping, etc. In daily life, individuals are confronted with tasks of different perceptual-motor complexity, entailing various patterns of inter-limb coordination under varying time constraints. This calls for research involving different rhythmic structures to be reproduced by means of motor tasks requiring various body segments (i.e., hand, foot, and whole body). A theoretical framework appropriate to jointly consider and operationalize the complexity of rhythmic patterns and movement task requirements is that proposed by Wood (1986) in the general study of tasks. He proposed to combine the frameworks of (1) ‘‘task as behavior requirements’’ and (2) ‘‘task qua task’’, providing a feasible operationalization of tasks as (1) behavioral responses involving motor activities and (2) patterns of stimuli impinging on the individual, requiring processing of information cues. The relationship between behavioral responses and information cues is one dimension of task complexity. The timing requirements to per-
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
205
form movement tasks represent a key aspect of this relationship. Also, task complexity is in inverse relation to what is referred to as component redundancy, which is the degree of overlap among the demands imposed by different responses to be executed or information cues to be processed. According to this theoretical framework, in the present study, task complexity should therefore be determined by the interval ratios between events and their degree of redundancy within a rhythm, as well as by the type of movement task requirements for rhythmic reproduction. Thus, the present study aimed at investigating (1) whether the reproduction of rhythmic patterns is influenced by movement task complexity (i.e., finger-tapping, foot-tapping, and walking) and (2) whether the effect of movement task complexity on rhythmic reproduction performance also depends on the complexity of the rhythmic pattern to be reproduced. It was hypothesized that (1) the correct reproduction of the quantitative aspects of the patterns (i.e., numbers of events in each rhythm) does not depend on movement task complexity. This may be expected because of component redundancy. In fact, once the individuals have correctly identified the number of events, they have to reproduce them by means of multiple executions of the same movement. Also, it was hypothesized that (2) increasing movement task and rhythm pattern complexity worsens the qualitative aspects of rhythmic reproduction, increasing the ratios between events and total duration of performance (i.e., reducing the internal consistency). This was expected because the qualitative aspects of rhythmic reproduction, more strictly reflecting the time constraints of the perceptual-motor task, should have a higher discriminative power than the mere quantity of events reproduced by movement. 2. Methods 2.1. Experimental approach to the problem The Local Review Board approved the design of the study, which included three movement tasks (i.e., tapping, stepping and walking). According to the literature, tapping has been used to evaluate how the central nervous system manages the perception and reproduction of rhythmic patterns (Repp, 2005) and has been considered the control condition (Getchell et al., 2001; Nagasaki, 1990); stepping has been used to manipulate the transduction delay between tapping and its central representation, considering the increased neural distance between effectors and brain (Aschersleben, 2002; Repp, 2005); and walking has been used to investigate complex coordination (Hausdorff, Yogev, Springer, Simon, & Giladi, 2005; Nagasaki, Itoh, Hashizume, & Furuna, 1996; Styns, van Noorden, Moelants, & Leman, 2007) of an automatic, rhythmic and regular motor activity (Beauchet, Allali, Berrut, Dubost, & Assal, 2008) in relation to rhythmic constraints. Because a standardized metrical stimulus event can only exist in synthesized patterns (Jones & Yee, 1993), computer software (Cubase SX3, Steinberg, Hamburg, Germany) was used to synthesize a set of audio stimuli with specific time signature, speed, timbre and loudness, and a commercial media player was used to reproduce each audio stimulus during the experimental session. The individual’s ability to reproduce rhythmic structures was evaluated by means of a validated rhythmic test (Persichini & Capranica, 2004), which includes three rhythmic patterns (Fig. 1). In particular, the first rhythmic pattern (r1), lasting 3000 ms, consisted of 6 events distributed in two repeated segmentations of two one-eight notes and one-quarter-notes each and was a 4/4 signature played with a 60 beats/time tempo. The second rhythmic pattern (r2), lasting 2252 ms and characterized by different segmentation from multi-level time hierarchies, consisted of 9 events and was a 4/4 played with a 60 beat/min tempo. The third rhythm (r3), lasting 3375 ms and characterized by an irregular structure, consisted of 8 events and was a 12/8 time signature played with 80 beat/min. Three rhythmic parameters were considered necessary for the quantification of the individual’s ability in the reproduction of rhythmic patterns (Persichini & Capranica, 2004): (1) DNE, calculated as follows: DNE = NEm NEs where NEm is the number of events reproduced by the participant and NEs is the number of the events included in the standard rhythmic pattern. The correct participant’s reproduction of number of events will correspond at 100% of DNE equal to zero.
206
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
Fig. 1. The music notation and inter-event time intervals (D) of the three proposed rhythmic patterns.
(2) DT, calculated as follows: DT = Tm Ts where Tm is the duration of participant’s performance and Ts is the total duration of the standard rhythmic pattern. For DT parameter, being zero the perfect reproduction with respect to the standard rhythm pattern, positive values indicated slow reproduction, whereas negative values indicated fast execution; and (3) the ratio among rhythmic events (R) was calculated as follows:
Dtmi Dtsi n Ts 1 X Tm R¼ Dtsi n i¼1 T s
where i is the index of the interval between two consecutive events (it can vary between 1 and n = NEs 1), Dtsi is the temporal duration of the i-th interval in the standard rhythmic pattern, and Dtmi is the corresponding performed temporal interval. When compared to the standard rhythm pattern, high values will indicate worst performances. The DT and R parameters were calculated only if DNE = 0.
2.2. Participants Seventeen adult participants (8 men, 9 women) ranging in age from 28 to 44 years (mean 34.8 ± 4.2 yrs), signed a consent for voluntary participation in the study. To control for potential confounding factors, criteria for inclusion in this study were that participants had neither evident perceptual disorder, nor formal rhythmic training and musical experience.
2.3. Apparatus A 9-camera Vicon MX stereophotogrammetric system (ViconÒ, UK, sampling frequency = 120 samples.s1) was used to reconstruct the 3D position of 6 retro-reflective spherical markers located on the skin of the subjects (Cappozzo, Della Croce, Leardini, & Chiari, 2005). Modern stereophotogrammetric systems like this one are able to reconstruct this value with an accuracy of about 2 mm. In this paper, the kinematic information was processed aiming at extracting temporal values from which the main parameters were estimated using the software MATLAB (The Mathworks Inc, MA, USA). The relevant temporal resolution obtained, strictly related to sampling frequency, was lower than 4.17 ms.
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
207
2.4. Procedures The participants were tested individually during an experimental session lasting about 20 min. Before starting the test session, each subject was equipped with 6 retro-reflective markers (2 located on the second finger nails of the left and right hand, and 4 located on the fifth metatarsal head and heel of their shoes), and familiarized with the following motor tasks: (1) Finger-tapping (T): sitting on a chair in front of a table, the participant was asked to reproduce the rhythmic pattern by tapping events alternating the two second fingertips on the table surface; (2) Foot-tapping (i.e., stepping, S): from a standing position, the participant was asked to reproduce the rhythmic pattern by stepping in place alternating events with the two feet on the ground; (3) Walking (W): from a standing position, the participant was asked to reproduce the rhythmic pattern by walking forward alternating events with the two foot strikes. Participants were allowed to familiarize themselves with the three rhythms. In fact, each rhythmic pattern was presented three times (McAdams & Bigand, 1999) to the listener. At the end of a sequence, the participants were asked to reproduce the rhythmic pattern from memory by means of tapping, or stepping, or walking. According to the literature (Persichini & Capranica, 2004; Summers & Pressing, 1994), this paradigm produces better overall performance with respect to a synchronization task, allowing participants to organize their responses. For the test, both rhythm presentation (r1, r2, r3) and reproduction modalities (T, S, W) were randomized, obtaining nine different experimental conditions (for example: r1T, r3S, r2W, r3T, r1W, r2S, r3W, r2T, r1S). Then, for each experimental condition, three trials (set) were required and recorded with a 1-min rest in between one set of test and its following one. 2.5. Data reduction To estimate rhythmic pattern reproduction correctness, the following time instants were derived from the reconstructed 3D marker positions: (1) time instants at which local minimum of the vertical marker displacement was observed; and (2) time instants at which local maximum of the vertical acceleration was observed in the range of minimum marker height. From these parameters, temporal occurrence of each participant’s event was determined as the time coinciding to the maximum of the vertical acceleration in correspondence of an interval around a minimum marker height (Fig. 2). This procedure was designed to prevent possible false positives caused by acceleration peaks not corresponding to real events. For the T task, each event was estimated taking into account the marker located on the second fingernail, while for S and W tasks both foot markers were taken into account and the first occurrence of a vertical acceleration peak was considered. Best performances were selected and means between trials were also calculated. 2.6. Statistical analysis Prior to the study, a .05 level of confidence was selected. Statistical analyses were carried out using statistics package SPSS (17.0) and data are presented as means and standard deviations. Preliminary analysis ascertained no differences between genders and groups. Thus, data were pooled for further analysis. To analyze performance consistency for each rhythmic pattern and motor task, the correct/wrong performances were counted; a chi-square test was applied to DNE and intraclass correlation coefficients (ICCs) were calculated for DT and R values. ICCs ranging from .80 to 1.00, from .60 to .79, or lower than .60 were considered highly, moderately reliable or questionable, respectively. Furthermore, Pearson’s correlations between individual’s best and average DT and R performances were calculated. Kruskall-Wallis test was applied to DNE with the rhythmic patterns (r1, r2 and r3) and movement tasks (T, S, and W) as factors. Only for performances having DNE = 0, DT and R best performances were submitted to a 3 (Rhythm: r1, r2 and r3) 3 (Task: T, S, and W) ANOVA for repeated measures. When a significant interaction was obtained, post hoc comparisons were performed using Fisher’s least significant difference test and the Bonferroni alpha level correction was applied to eliminate an inflated type I error risk for multiple comparisons. Furthermore, to provide meaningful analysis for compari-
208
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
Fig. 2. Representation of a tapping reproduction of the first rhythmic pattern: (a) marker vertical position (m); (b) marker vertical acceleration (m/s2). Black line: right finger; grey line: left finger. In the higher part of the b subplot, vertical thin lines and circles indicate the theoretical and the reproduced events, respectively, following the alignment of the first event.
sons between small groups, Cohen’s effect size was also calculated for significant effects. An effect size <0.2 was considered trivial, from 0.2 to 0.6 small, from 0.7 to 1.2 moderate, and >1.2 large.
3. Results For DNE, no difference emerged between trials. High ICCs were observed for DT (0.97 ± 0.02) and r (0.74 ± 0.21), indicating a good intra-individual stability. In general, moderate significant correlations between mean and best DT were found (Table 1), whereas the relative picture for R always showed high and significant correlation coefficients (Table 2). For DNE, percentages of correct reproduction of rhythmic patterns are reported in Fig. 3 as a function of rhythm and movement task. Failures were observed more often for r2 with respect to r1 and r3, although no difference emerged as a function of rhythm and/or movement task. For DT, differences emerged for movement task, F(2, 24) = 18.66, p < .0001, ES range: 0.41–0.79, rhythmic patterns, F(2, 24) = 187.27, p < .0001, ES range: 1.94–2.20, and their interaction, F(4, 48) = 14.15, p < .0001. DT values were significantly higher (p < .0001) for r2 with respect to the other two rhythmic patterns. Furthermore, differences were found between all movement tasks, with higher values for walking, lower for tapping and intermediate for stepping. Fig. 4 illustrates the Rhyth-
Table 1 Correlation coefficients (r), 95% confidence intervals (95% CI), and significant values (p) between best and average DT values of for the three rhythmic patterns (i.e., r1, r2, and r3) and movement tasks (i.e., tapping, stepping, and walking). Rhythmic pattern
Movement task
r
r1
Tapping Stepping Walking
.580 .581 .733
95% CI .14–.83 .12–.84 .39–.90
p
r2
Tapping Stepping Walking
.355 .721 .994
.17–.72 .35–.90 .980.99
.177 .002 <.0001
r3
Tapping Stepping Walking
.228 .476 .959
.30–.65 .01–.78 .89–.99
.395 .053 <.0001
.015 .018 .001
209
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
Table 2 Correlation coefficients (r), 95% confidence intervals (95% CI), and significant values (p) between best and average R values for the three rhythmic patterns (i.e., r1, r2, and r3) and movement tasks (i.e., tapping, stepping, and walking). Rhythmic pattern
Movement task
r
95% CI
p
r1
Tapping Stepping Walking
.951 .796 .651
.87–.98 .50–.93 .25–.86
<.0001 <.0001 .005
r2
Tapping Stepping Walking
.971 .942 .945
.92–.99 .84–.98 .83–.98
<.0001 <.0001 <.0001
r3
Tapping Stepping Walking
.875 .995 .935
.67–.96 .98–.99 .83–.98
<.0001 <.0001 <.0001
Fig. 3. Frequency of occurrence (%) of correct execution of the number of events (DNE) for the three rhythmic patterns (i.e., r1, r2, and r3) and movement tasks (i.e., tapping, stepping, and walking).
Fig. 4. Mean ± SD of differences between the duration of subject’s performance and the total duration of the standard rhythm pattern (DT) for the three rhythmic patterns (i.e., r1, r2, and r3) and movement tasks (i.e., tapping, stepping, and walking). ⁄ Significant differences with respect to walking (p < .05) # Significant differences with respect to stepping (p < .01)
210
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
Fig. 5. Mean and SD of the ratio among rhythmic events (R) for the three rhythmic patterns (i.e., r1, r2, and r3) and movement tasks (i.e., tapping, stepping, and walking). ⁄ Significant differences between r2 and the other two rhythms (p < .001). # Significant differences between walking and the other two tasks (p < .0001)
mic Pattern Task interaction. Only for r1, stepping did not show differences with respect to walking due to a floor effect. For r2, a difference (p < .01) also emerged between tapping and stepping. For the R parameter (Figs. 5a and b), only main effects were found (movement task: F(2, 24) = 28.05, p < .0001, ES range: 0.94–1.21; rhythmic pattern: F(2, 24) = 16.011, p < .0001, ES range: 0.40–0.95). Similar to simple effects for DT, post hoc comparisons showed differences for r2 (p < .001) and W (p < .0001), with worst performances as compared to the other two rhythm and task conditions.
4. Discussion The present study aimed at assessing the effects of movement complexity on the ability of healthy adults to reproduce rhythmic patterns. The main findings of this study are: (1) the good internal stability of the rhythmic reproductions, indicating that this test is a valid instrument for the evaluation of rhythmic ability in musically illiterate adults; and (2) the good discriminative power of the qualitative aspects of rhythmic reproduction (i.e., speed of execution and temporal ratios between events), which are sensitive to the complexity of both rhythmic pattern and movement task requirements. Consistency in performance is likely to occur in experienced individuals and a learning effect could be expected in novice individuals depending on rhythmic pattern and movement task complexity. Instead, despite being musically illiterate, the participants in this study showed a good stability between multiple trials performed within a single day, with significant correlations between best and mean performance values substantiating the construct validity of the employed rhythmic test (Persichini & Capranica, 2004). The stability across trials also supports its ecological validity, since the behaviors observed and recorded (i.e., tapping, stepping and walking) required motor performances commonly used in natural settings of daily living, thus allowing generalization of the findings of this study to real world conditions. Independently of movement task complexity, DNE reproductions showed a ceiling effect, especially evident for the first and third rhythmic patterns. Conversely, DT and R parameters allowed discriminating the individual’s rhythmic ability in relation to task complexity. Specifically, they were sensitive to different dimensions of task complexity (Wood, 1986), with R being independently affected by the degree of complexity of the behavioral responses and rhythmic information cues and DT by the relationship between them. The distinctive relationship between the events influenced the individual’s capability to reproduce rhythmic patterns more than their time hierarchy (Drake, 1993; Persichini & Capranica, 2004; Sakai et al., 2004; Trehub & Hannon, 2009). Thus, the role of rhythmic pattern construct becomes crucial for the correct interpretation of temporal information and for the consequent motor reproduction. As expected (Drake, 1993; Persichini & Capranica, 2004), the simple rhythmic framework of the first pattern, characterized by two reiterated subunits (i.e., 1:1:2 ratio) with a sep-
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
211
aration of a long duration in the middle, were more easily reproduced. Despite the fact that the third rhythm presented a ternary subdivision, individuals tended to be more able to perceive and correctly reproduce its time duration and inter-event organization with respect to the binary structure of the second rhythm, which was rendered more challenging by the inclusion of a 3:1 ratio and four short (i.e., 188/188/188/188 ms) inter-event interval relationships (Fig. 5, left panel) (Drake, 1993; Large, 2000; Persichini & Capranica, 2004; Povel, 1981). Also a different degree of component redundancy (Wood, 1986) as concerns the number of different event group (four in the second rhythm and three in the third one) may have contributed to the observed differences in performance. The qualitative parameters of rhythmic ability proved to depend not only on perceptive processes but also on motor coordination (Fraisse, 1979). Regarding the total reproduction duration (reflecting movement velocity), the present data highlight a tendency to slow down in correspondence to increasing complexity of rhythmic patterns, with limited, intermediate, and longest delays being observed for the reproduction of the first, third, and second pattern, respectively. The longest delay observed in walking rhythm reproduction was paralleled by largest ratios between events for this task, suggesting a coupled influence of perceptual and motor determinants particularly in reproduction tasks involving more complex interlimb coordination and balance control. Previous findings (Persichini & Capranica, 2004) reported anticipatory responses during the first and third patterns for tapping. This phenomenon was confirmed in the present study and occurred also for stepping, indicating similar control mechanisms for upper and lower limbs when reproducing rhythmic patterns. These results are in disagreement with the hypothesis that different sensory information and transmission time differently affect hand vs. foot rhythmic reproductions. Instead, they better support a sensory accumulator model (Aschersleben, 2002; Repp, 2005) that includes different rates from different sensory channels, which effectively control different kinematics of foot versus finger tapping. It has been proposed that a broad variety of task requirements (i.e., hand tapping, foot tapping, and vocal performances) share common mechanisms of rhythmic timing involving memory processes, with rhythmic performances not completely accurate but deviating from the ideal pattern (Gilden et al., 1995; Hennig et al., 2011; Henning, Fleischmann, & Geisel, 2012; Roberts, Eykholt, & Thaut, 2000). Temporal fluctuations from the correct timing could depend on a sort of erroneous anticipatory behavior (Joiner & Shelhamer, 2009). In particular, the literature on synchronized tapping tasks reported an anticipation tendency in synchronization tasks (Ascherleben & Prinz, 1995; Fraisse, 1966; Fraisse & Voillaume, 1971; Peters, 1989; Repp, 2005), which could be reduced with increasing rhythmic complexity or subdivision (Repp, 2005). Accordingly, anticipatory responses were not present in the reproduction of the most complex structure of the second pattern, which, in contrast, tended to be delayed and showed the largest differences with respect to the standard and significant differences between the three tasks. The interaction between movement task and rhythmic pattern complexity showed the largest differences in rhythmic reproduction performance among the three movement tasks for the more complex rhythm. Also, it showed that walking task performances differed from the simpler tapping and stepping performances in all three rhythm reproductions. The ability to monitor, detect, and adjust deviations from intended behaviors is a hallmark of skilled performance (Palmer & Drake, 1997). Therefore, it is possible to speculate that a rhythmic pattern with different inter-onset intervals challenges unskilled individuals to actuate appropriate corrective strategies to avoid reproduction errors, especially when combining different kinematics of body segments in relation to space. To substantiate these findings, further investigations on perception and action coordinated in time should include populations of different age (i.e., children and older individuals) and skill level (i.e., athletes and musicians). For instance, rhythmic gymnasts and musicians might use deviations from a given rhythmic pattern to accentuate and interpret the piece. Thus, fluctuations of the R parameter, which is considered the most important indicator of proficiency in reproducing time relations among events (Krampe, Engbert, & Kliegl, 2001; Persichini & Capranica, 2004), could represent a benchmark for further generalizations or differentiations. The novel aspect of the present study was represented by the combined manipulation of the complexity of the perceptual and movement task requirements which represents a relevant dimension of the overall task complexity (Wood, 1986). The present findings highlighted a complex interplay be-
212
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
tween perceptual and motor demands in determining the individual rhythmic reproduction ability. This interplay emerged when considering the performance parameter of total duration, with largest deviations in the reproduction of the second pattern by walking. In contrast, the performance parameter of inter-event ratio did not show the expected interaction between rhythmic and motor task complexity, indicating that individuals adopt similar behaviors when challenged by major constraints. These findings might provide useful practical information about how to individualize methodological strategies in relation to age, sporting discipline, and performance level. Although walking is widely considered as an automatic motor activity requiring a limited amount of cognitive effort, explicit cognitive control is required when different constraints (i.e., obstacles, directional changes or stride time modification) are requested (Beauchet et al., 2008). In fact, studies on motor programs highlight that additional activation components overlap the existing locomotor primitivies (Ivanenko et al., 2006). The results obtained from the qualitative parameters of this test confirmed that rhythmic constraints generate a significant need for cognitive supervision during walking to properly integrate the sensory information and regulate dynamic balance and gait (Hausdorff et al., 2005). In particular, in the presence of distorted information due to extrinsic or intrinsic factors, the spontaneous walking rhythm could prevail over reproduction of an external rhythm, thus leading to a failure in rhythm reproduction. Because the individual capacity to correctly program and execute appropriate corrective strategies may have the potential to reduce reproduction errors, further research is needed to investigate the behaviors of proficient performers (i.e., athletes, dancers, and musicians), who are required to produce movements in a fluent manner according to strict rate or tempo constraints and preserving certain durational relationship between events. References Ascherleben, G., & Prinz, W. (1995). Synchronizing actions with events: The role of sensory information. Perception & Psychophysics, 57, 305–317. Aschersleben, G. (2002). Temporal control of movements in sensorimotor synchronization. Brain and Cognition, 48, 66–79. Beauchet, O., Allali, G., Berrut, G., Dubost, V., & Assal, F. (2008). Gait analysis in demented subjects: Interests and perspectives. Neuropsychiatric Disease and Treatment, 4, 155–160. Cappozzo, A., Della Croce, U., Leardini, A., & Chiari, L. (2005). Human movement analysis using stereophotogrammetry. Part 1: Theoretical background. Gait & Posture, 21, 186–196. Chen, Y., Ding, M., & Kelso, J. A. S. (1997). Long memory processes (1/f a type) in human coordination. Physical Review Letters, 79, 4501–4504. Chen, Y., Ding, M., & Kelso, J. A. S. (2001). Origin of timing errors in human sensorimotor coordination. Journal of Motor Behavior, 33, 3–8. Dalcroze, J. (2008). Il ritmo, la musica e l’educazione. Torino: Educazione Musicale EDT/SIEM. Drake, C. (1993). Reproduction of musical rhythms by children, adult musicians, and adult nonmusicians. Perception & Psychophysics, 53, 25–33. Essens, P. J. (1986). Hierarchical organization of temporal patterns. Perception & Psychophysics, 40, 69–73. Essens, P. J., & Povel, D. J. (1985). Metrical and nonmetrical representations of temporal patterns. Perception & Psychophysics, 37, 1–7. Fraisse, P. (1966). L’anticipation de stimulus rythmiques. Vitesse d’établissement et precision de la synchronisation. L’année psychologique (Vol. 66, pp. 15–36). Fraisse, P. (1979). Psicologia del ritmo. Roma: Armando Editore. Fraisse, P., & Voillaume, C. (1971). Les repéres du sujet dans la synchronisation et dans la pseudosynchronisation. L’année psychologique (Vol. 71, pp. 359–369). Fraisse, P., Pichot, P., & Clairouin, G. (1949). Les aptitudes rythmiques: Etude comparée des oligophrénes et des enfants normaux. Journal de Psychologie Normale et Pathologique, 42, 309–330. Getchell, N., Forrester, L., & Whitall, J. (2001). Individual differences and similarities in the stability, timing consistency, and natural frequency of rhythmic coordinated actions. Research Quarterly for Exercise and Sport, 72, 13–21. Gilden, D. L., Thornton, T., & Mallon, M. W. (1995). 1/f noise in human cognition. Science, 267, 1837–1839. Golubitsky, M., Stewart, I., Buono, P.-L., & Collins, J. J. (1999). Symmetry in locomotor central pattern generators and animal gaits. Nature, 401, 693–695. Hausdorff, J. M., Peng, C.-K., Ladin, Z., Wei, J. Y., & Goldberger, A. L. (1995). Is walking a random walk? Evidence for long-range correlations in stride interval of human gait. Journal of Applied Physiology, 78, 349–358. Hausdorff, J. M., Yogev, G., Springer, S., Simon, E. S., & Giladi, N. (2005). Walking is more like catching than tapping: Gait in the elderly as a complex cognitive task. Experimental Brain Research, 164, 541–548. Hennig, H., Fleischmann, R., Fredebohm, A., Hagmayer, Y., Nagler, J., Witt, A., et al (2011). The nature and perception of fluctuations in human musical rhythms. PLoS One, 6, e26457. Henning, H., Fleischmann, R., & Geisel, T. (2012). Musical rhythms: The science of being slightly off. Physics Today, 65, 64–65. IJspeert, A. J. (2008). Central pattern generators for locomotion control in animals and robots: A review. Neural Networks, 21, 642–653. Ivanenko, Y. P., Poppele, R. E., & Lacquaniti, F. (2006). Motor control programs and walking. The Neuroscientist, 12, 339–348.
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
213
Joiner, W. M., & Shelhamer, M. (2009). A model of time estimation and error feedback in predictive timing behavior. Journal of Computational Neuroscience, 26, 119–138. Jones, M. R. (1987). Dynamic pattern structure in music: Recent theory and research. Perception & Psychophysics, 41, 621–634. Jones, M. R., & Yee, W. (1993). Attending to auditory events: The role of temporal organization. In S. McAdams & E. Bigand (Eds.), Thinking in sound: The cognitive psychology of human audition (pp. 69–112). Oxford: Clarendon Press. Keele, S. W., & Hawkins, H. L. (1982). Explorations of individual differences relevant to high level skill. Journal of Motor Behavior, 14, 2–23. Keele, S. W., Ivry, R. I., & Pokorny, R. A. (1987). Force control and its relation to timing. Journal of Motor Behavior, 19, 96–114. Krampe, R. T., Engbert, R., & Kliegl, R. (2001). Age-specific problems in rhythmic training. Psychology and Aging, 16, 12–30. Large, E. W. (2000). On synchronizing movements to music. Human Movement Science, 19, 527–566. Leman, M. (2007). Embodied music cognition and mediation technology. Cambridge, MA: The MIT-Press. McAdams, S., & Bigand, E. (1999). Introduction to auditory cognition. In S. McAdams & E. Bigand (Eds.), Thinking in sound: The cognitive psychology of human audition (pp. 1–10). Oxford: Clarendon Press. Meeuwsen, H. J., Flohr, J. W., & Fink, R. (1998). Computerized assessment of the skills of synchronization and the imitation and timing of rhythmic patterns. Technological directions in music education: Institute for music research, University of Texas at San Antonio, 93–95. Meinel, K., & Schnabel, G. (1998). Bewegungslehre – Sportmotorik [Theory of movement – motor behavior in sport] (9th ed.). Berlin: Sportverlag Berlin. Nagasaki, H. (1990). Rhythm and variability of timing in periodic tapping. Human Movement Science, 9, 177–194. Nagasaki, H., Itoh, H., Hashizume, K., & Furuna, T. (1996). Walking patterns and finger rhythm of older adults. Perception and Motor Skills, 82, 435–447. Palmer, D., & Drake, C. (1997). Monitoring and planning capacities in the acquisition of music performance skills. Canadian Journal of Experimental Psychology, 51, 369–384. Persichini, C., & Capranica, L. (2004). Reproduction of rhythmic patterns in young individuals: Toward the construction of a computerized rhythmic test. International Journal of Sport Psychology, 35, 246–265. Peters, M. (1989). The relationship between variability of intertap intervals and interval duration. Psychological Research, 51, 38–42. Philliphs-Silver, J., & Trainor, L. J. (2007). Hearing what the body feels: Auditory encoding of rhythmic movement. Cognition, 105, 533–546. Povel, D. J. (1981). Internal representation of simple temporal patterns. Journal of Experimental Psychology: Human Perception and Performance, 7, 3–18. Repp, B. H. (2005). Sensorimotor synchronization: A review of the tapping literature. Psychonomic Bulletin & Review, 12, 969–992. Roberts, S., Eykholt, R., & Thaut, M. H. (2000). Analysis of correlations and search for evidence of deterministic chaos in rhythmic motor control by the human brain. Physical Review E, 62, 2597–2607. Robertson, S. D., Zelaznik, H. N., Lantero, D. A., Bojczyk, K. G., Spencer, R. M., Doffin, J. G., et al (1999). Correlations for timing consistency among tapping and drawing tasks: Evidence against a single timing process for motor control. Journal of Experimental Psychology: Human Perception and Performance, 25, 1316–1330. Sakai, K., Hikosaka, O., Miyauchi, S., Takino, R., Tamada, T., Iwata, N. K., et al (1999). Neural representation of a rhythm depends on its interval ratio. The Journal of Neuroscience, 19, 10074–10081. Sakai, K., Hikosaka, O., & Nakamura, K. (2004). Emergence of rhythm during motor learning. Trends in Cognitive Sciences, 8, 547–553. Smoll, F. L., & Palmatier, R. A. (1977). A second-generation motor rhythm analysis system. Behavior Research Methods & Instrumentation, 9, 508–510. Styns, F., van Noorden, L., Moelants, D., & Leman, M. (2007). Walking on music. Human Movement Science, 26, 769–785. Summers, J. J. (1975). The role of timing in motor program representation. Journal of Motor Behavior, 7, 229–241. Summers, J. J., & Pressing, J. (1994). Coordinating the two hands in polyrhythmic tapping. In S. P. Swinnen, H. Heuer, J. Massion, & P. Casaer (Eds.), Interlimb coordination: Neural, dynamical, and cognitive constraints (pp. 571–591). New York: Academic Press, Inc. Tood, N. P., Cousins, R., & Lee, C. S. (2007). The contribution of anthropomorphic factors to individual differences in the perception of rhythm. Empirical Musicology Review, 2, 1–13. Trainor, L. J. (2007). Do preferred beat have rate and entrainment to beat have a common origin in movement? Empirical Musicology Review, 2, 17–20. Trehub, S. E., & Hannon, E. E. (2009). Conventional rhythms enhance infants’ and adults’ perception of musical patterns. Cortex, 45, 110–118. Wood, R. E. (1986). Task complexity: Definition of the construct. Organizational Behavior and Human Decision Processes, 37, 60–82. Zachopoulou, E., & Mantis, K. (2001). The role of rhythmic ability on the forehand performance in tennis. Physical Education & Sport Pedagogy, 6, 117–126.
Contents lists available at SciVerse ScienceDirect
Human Movement Science journal homepage: www.elsevier.com/locate/humov
Effects of task complexity on rhythmic reproduction performance in adults Flora Iannarilli, Giuseppe Vannozzi, Marco Iosa, Caterina Pesce, Laura Capranica ⇑ DiSMUS, University of Rome Foro Italico, Italy
a r t i c l e
i n f o
Article history: Available online 27 February 2013 PsycINFO classification: 3357 Keywords: Rhythmic ability Movement task Rhythm Evaluation Movement complexity
a b s t r a c t The aim of the present study was to investigate the effect of task complexity on the capability to reproduce rhythmic patterns. Sedentary musically illiterate individuals (age: 34.8 ± 4.2 yrs; M ± SD) were administered a rhythmic test including three rhythmic patterns to be reproduced by means of finger-tapping, foot-tapping and walking. For the quantification of subjects’ ability in the reproduction of rhythmic patterns, qualitative and quantitative parameters were submitted to analysis. A stereophotogrammetric system was used to reconstruct and evaluate individual performances. The findings indicated a good internal stability of the rhythmic reproduction, suggesting that the present experimental design is suitable to discriminate the participants’ rhythmic ability. Qualitative aspects of rhythmic reproduction (i.e., speed of execution and temporal ratios between events) varied as a function of the perceptualmotor requirements of the rhythmic reproduction task, with larger reproduction deviations in the walking task. Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction Rhythm is a fundamental component of the organization and the execution of body movements, easily observable during normal daily activities (e.g., walking, running, clapping hands, drumming fingers), and in dance, music, and sport (e.g., rhythmic gymnastics and synchronized swimming) contexts. Theories in cognitive and musical sciences considered the ‘‘dynamogenous’’ effect of rhythm (Fraisse, 1979) and highlighted the role of the body as a mediator for music awareness (Dalcroze, 2008; Leman, 2007). In fact, music emerges from physical movement (Tood, Cousins, & Lee, 2007; ⇑ Corresponding author. Tel.: +39 06.36 73 32 07; fax: +39 06 36 73 33 30. E-mail address: [email protected] (L. Capranica). 0167-9457/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.humov.2012.12.004
204
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
Trainor, 2007) and rhythm is embodied and expressed through movements (Philliphs-Silver & Trainor, 2007), which require the organization of muscular activity to execute a specific chronological order of motor actions (Meinel & Schnabel, 1998; Sakai, Hikosaka, & Nakamura, 2004). Actually, studies on human coordination refer to rhythm as a crucial aspect for movements such as walking (Hausdorff, Peng, Ladin, Wei, & Goldberger, 1995) and tapping (Chen, Ding, & Kelso, 1997, 2001), which fluctuates in relation to higher nervous system centres and/or lower motor neuron control (Golubitsky, Stewart, Buono, & Collins, 1999; Ijspeert, 2008; Ivanenko, Poppele, & Lacquaniti, 2006). The capability to translate an acoustic perception into a corresponding motor behavior is considered a fundamental characteristic of rhythmic ability (Fraisse, Pichot, & Clairouin, 1949), which is defined as the ability to perform a succession of regulated, recurring gross motor events requiring both spatial and temporal accuracy (Zachopoulou & Mantis, 2001). In considering that performances mainly depend upon the rhythmic structure (Gilden, Thornton, & Mallon, 1995), to discriminate the rhythmic ability of different populations, the use of a large range of pitch intervals and several complex rhythms is recommended (Persichini & Capranica, 2004; Trehub & Hannon, 2009). In particular, the ability to accurately process and reproduce rhythmic patterns depends on the interval ratios between events, with spontaneous reproduction of 1:1 and 2:1 rhythmic structures (Drake, 1993; Essens & Povel, 1985) and preferences for small integer ratios (1:2, 1:3, or 1:4) as compared to larger (1:5) or non-integer (in the sense of non-simple) (1:2.5 or 1:3.5) ratios (Drake, 1993; Essens, 1986; Sakai et al., 1999). In general, during learning or rehearsal, non-integer ratios are often distorted and shifts toward simpler ratios are observed (Drake, 1993; Essens, 1986; Essens & Povel, 1985; Sakai et al., 1999; Summers, 1975; Trehub & Hannon, 2009). Furthermore, binary subdivisions are easier to be discriminated and reproduced with respect to ternary subdivisions (Jones, 1987; Povel, 1981; Trehub & Hannon, 2009). To evaluate the individual ability to correctly reproduce a rhythmic pattern, both its quantitative (i.e., number of events) and qualitative (i.e., speed of execution and temporal ratios among events) aspects must be considered (Persichini & Capranica, 2004). In fact, individuals might repeatedly deviate from the intervals between events around a constant value, thereby creating a considerable error in the speed of execution (i.e., overall faster or slower performances), yet producing the prominent rhythmic structure. On the other hand, individuals could correctly reproduce the established total time of the pattern even though the proportionality between intervals of rhythmic structures is not respected (Persichini & Capranica, 2004). Unfortunately, thorough comparison of evidence concerning rhythmic ability is hard because of large evaluation differences such as the use of heterogeneous stimuli (i.e., auditory, visual or audio-visual), procedures (i.e., different forms of coordination, in-phase or anti-phase movements), effectors (i.e., hands, feet, or voice), and rhythmic variables (i.e., number of events, total duration of the pattern, intervals between events) (Drake, 1993; Hennig et al., 2011; Joiner & Shelhamer, 2009; Meeuwsen, Flohr, & Fink, 1998; Persichini & Capranica, 2004; Repp, 2005; Smoll & Palmatier, 1977). A further relevant aspect in rhythmic ability evaluation is related to the movement task requirements for rhythmic pattern reproduction. To generate wide possibilities of motor programs in relation to rhythmic movements, a modular control of limb kinematics based on various feedback and feed-forward parameters has been hypothesized (Ivanenko et al., 2006). Although it has been hypothesized that a general timing process assists the individual in performing movements of different limbs and body parts (i.e., finger, forearm and foot tapping) (Keele & Hawkins, 1982; Keele, Ivry, & Pokorny, 1987), some authors (Getchell, Forrester, & Whitall, 2001; Robertson et al., 1999) argued that specific timing processes are needed when tasks are not dynamically equivalent, such as tapping, clapping, galloping, etc. In daily life, individuals are confronted with tasks of different perceptual-motor complexity, entailing various patterns of inter-limb coordination under varying time constraints. This calls for research involving different rhythmic structures to be reproduced by means of motor tasks requiring various body segments (i.e., hand, foot, and whole body). A theoretical framework appropriate to jointly consider and operationalize the complexity of rhythmic patterns and movement task requirements is that proposed by Wood (1986) in the general study of tasks. He proposed to combine the frameworks of (1) ‘‘task as behavior requirements’’ and (2) ‘‘task qua task’’, providing a feasible operationalization of tasks as (1) behavioral responses involving motor activities and (2) patterns of stimuli impinging on the individual, requiring processing of information cues. The relationship between behavioral responses and information cues is one dimension of task complexity. The timing requirements to per-
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
205
form movement tasks represent a key aspect of this relationship. Also, task complexity is in inverse relation to what is referred to as component redundancy, which is the degree of overlap among the demands imposed by different responses to be executed or information cues to be processed. According to this theoretical framework, in the present study, task complexity should therefore be determined by the interval ratios between events and their degree of redundancy within a rhythm, as well as by the type of movement task requirements for rhythmic reproduction. Thus, the present study aimed at investigating (1) whether the reproduction of rhythmic patterns is influenced by movement task complexity (i.e., finger-tapping, foot-tapping, and walking) and (2) whether the effect of movement task complexity on rhythmic reproduction performance also depends on the complexity of the rhythmic pattern to be reproduced. It was hypothesized that (1) the correct reproduction of the quantitative aspects of the patterns (i.e., numbers of events in each rhythm) does not depend on movement task complexity. This may be expected because of component redundancy. In fact, once the individuals have correctly identified the number of events, they have to reproduce them by means of multiple executions of the same movement. Also, it was hypothesized that (2) increasing movement task and rhythm pattern complexity worsens the qualitative aspects of rhythmic reproduction, increasing the ratios between events and total duration of performance (i.e., reducing the internal consistency). This was expected because the qualitative aspects of rhythmic reproduction, more strictly reflecting the time constraints of the perceptual-motor task, should have a higher discriminative power than the mere quantity of events reproduced by movement. 2. Methods 2.1. Experimental approach to the problem The Local Review Board approved the design of the study, which included three movement tasks (i.e., tapping, stepping and walking). According to the literature, tapping has been used to evaluate how the central nervous system manages the perception and reproduction of rhythmic patterns (Repp, 2005) and has been considered the control condition (Getchell et al., 2001; Nagasaki, 1990); stepping has been used to manipulate the transduction delay between tapping and its central representation, considering the increased neural distance between effectors and brain (Aschersleben, 2002; Repp, 2005); and walking has been used to investigate complex coordination (Hausdorff, Yogev, Springer, Simon, & Giladi, 2005; Nagasaki, Itoh, Hashizume, & Furuna, 1996; Styns, van Noorden, Moelants, & Leman, 2007) of an automatic, rhythmic and regular motor activity (Beauchet, Allali, Berrut, Dubost, & Assal, 2008) in relation to rhythmic constraints. Because a standardized metrical stimulus event can only exist in synthesized patterns (Jones & Yee, 1993), computer software (Cubase SX3, Steinberg, Hamburg, Germany) was used to synthesize a set of audio stimuli with specific time signature, speed, timbre and loudness, and a commercial media player was used to reproduce each audio stimulus during the experimental session. The individual’s ability to reproduce rhythmic structures was evaluated by means of a validated rhythmic test (Persichini & Capranica, 2004), which includes three rhythmic patterns (Fig. 1). In particular, the first rhythmic pattern (r1), lasting 3000 ms, consisted of 6 events distributed in two repeated segmentations of two one-eight notes and one-quarter-notes each and was a 4/4 signature played with a 60 beats/time tempo. The second rhythmic pattern (r2), lasting 2252 ms and characterized by different segmentation from multi-level time hierarchies, consisted of 9 events and was a 4/4 played with a 60 beat/min tempo. The third rhythm (r3), lasting 3375 ms and characterized by an irregular structure, consisted of 8 events and was a 12/8 time signature played with 80 beat/min. Three rhythmic parameters were considered necessary for the quantification of the individual’s ability in the reproduction of rhythmic patterns (Persichini & Capranica, 2004): (1) DNE, calculated as follows: DNE = NEm NEs where NEm is the number of events reproduced by the participant and NEs is the number of the events included in the standard rhythmic pattern. The correct participant’s reproduction of number of events will correspond at 100% of DNE equal to zero.
206
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
Fig. 1. The music notation and inter-event time intervals (D) of the three proposed rhythmic patterns.
(2) DT, calculated as follows: DT = Tm Ts where Tm is the duration of participant’s performance and Ts is the total duration of the standard rhythmic pattern. For DT parameter, being zero the perfect reproduction with respect to the standard rhythm pattern, positive values indicated slow reproduction, whereas negative values indicated fast execution; and (3) the ratio among rhythmic events (R) was calculated as follows:
Dtmi Dtsi n Ts 1 X Tm R¼ Dtsi n i¼1 T s
where i is the index of the interval between two consecutive events (it can vary between 1 and n = NEs 1), Dtsi is the temporal duration of the i-th interval in the standard rhythmic pattern, and Dtmi is the corresponding performed temporal interval. When compared to the standard rhythm pattern, high values will indicate worst performances. The DT and R parameters were calculated only if DNE = 0.
2.2. Participants Seventeen adult participants (8 men, 9 women) ranging in age from 28 to 44 years (mean 34.8 ± 4.2 yrs), signed a consent for voluntary participation in the study. To control for potential confounding factors, criteria for inclusion in this study were that participants had neither evident perceptual disorder, nor formal rhythmic training and musical experience.
2.3. Apparatus A 9-camera Vicon MX stereophotogrammetric system (ViconÒ, UK, sampling frequency = 120 samples.s1) was used to reconstruct the 3D position of 6 retro-reflective spherical markers located on the skin of the subjects (Cappozzo, Della Croce, Leardini, & Chiari, 2005). Modern stereophotogrammetric systems like this one are able to reconstruct this value with an accuracy of about 2 mm. In this paper, the kinematic information was processed aiming at extracting temporal values from which the main parameters were estimated using the software MATLAB (The Mathworks Inc, MA, USA). The relevant temporal resolution obtained, strictly related to sampling frequency, was lower than 4.17 ms.
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
207
2.4. Procedures The participants were tested individually during an experimental session lasting about 20 min. Before starting the test session, each subject was equipped with 6 retro-reflective markers (2 located on the second finger nails of the left and right hand, and 4 located on the fifth metatarsal head and heel of their shoes), and familiarized with the following motor tasks: (1) Finger-tapping (T): sitting on a chair in front of a table, the participant was asked to reproduce the rhythmic pattern by tapping events alternating the two second fingertips on the table surface; (2) Foot-tapping (i.e., stepping, S): from a standing position, the participant was asked to reproduce the rhythmic pattern by stepping in place alternating events with the two feet on the ground; (3) Walking (W): from a standing position, the participant was asked to reproduce the rhythmic pattern by walking forward alternating events with the two foot strikes. Participants were allowed to familiarize themselves with the three rhythms. In fact, each rhythmic pattern was presented three times (McAdams & Bigand, 1999) to the listener. At the end of a sequence, the participants were asked to reproduce the rhythmic pattern from memory by means of tapping, or stepping, or walking. According to the literature (Persichini & Capranica, 2004; Summers & Pressing, 1994), this paradigm produces better overall performance with respect to a synchronization task, allowing participants to organize their responses. For the test, both rhythm presentation (r1, r2, r3) and reproduction modalities (T, S, W) were randomized, obtaining nine different experimental conditions (for example: r1T, r3S, r2W, r3T, r1W, r2S, r3W, r2T, r1S). Then, for each experimental condition, three trials (set) were required and recorded with a 1-min rest in between one set of test and its following one. 2.5. Data reduction To estimate rhythmic pattern reproduction correctness, the following time instants were derived from the reconstructed 3D marker positions: (1) time instants at which local minimum of the vertical marker displacement was observed; and (2) time instants at which local maximum of the vertical acceleration was observed in the range of minimum marker height. From these parameters, temporal occurrence of each participant’s event was determined as the time coinciding to the maximum of the vertical acceleration in correspondence of an interval around a minimum marker height (Fig. 2). This procedure was designed to prevent possible false positives caused by acceleration peaks not corresponding to real events. For the T task, each event was estimated taking into account the marker located on the second fingernail, while for S and W tasks both foot markers were taken into account and the first occurrence of a vertical acceleration peak was considered. Best performances were selected and means between trials were also calculated. 2.6. Statistical analysis Prior to the study, a .05 level of confidence was selected. Statistical analyses were carried out using statistics package SPSS (17.0) and data are presented as means and standard deviations. Preliminary analysis ascertained no differences between genders and groups. Thus, data were pooled for further analysis. To analyze performance consistency for each rhythmic pattern and motor task, the correct/wrong performances were counted; a chi-square test was applied to DNE and intraclass correlation coefficients (ICCs) were calculated for DT and R values. ICCs ranging from .80 to 1.00, from .60 to .79, or lower than .60 were considered highly, moderately reliable or questionable, respectively. Furthermore, Pearson’s correlations between individual’s best and average DT and R performances were calculated. Kruskall-Wallis test was applied to DNE with the rhythmic patterns (r1, r2 and r3) and movement tasks (T, S, and W) as factors. Only for performances having DNE = 0, DT and R best performances were submitted to a 3 (Rhythm: r1, r2 and r3) 3 (Task: T, S, and W) ANOVA for repeated measures. When a significant interaction was obtained, post hoc comparisons were performed using Fisher’s least significant difference test and the Bonferroni alpha level correction was applied to eliminate an inflated type I error risk for multiple comparisons. Furthermore, to provide meaningful analysis for compari-
208
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
Fig. 2. Representation of a tapping reproduction of the first rhythmic pattern: (a) marker vertical position (m); (b) marker vertical acceleration (m/s2). Black line: right finger; grey line: left finger. In the higher part of the b subplot, vertical thin lines and circles indicate the theoretical and the reproduced events, respectively, following the alignment of the first event.
sons between small groups, Cohen’s effect size was also calculated for significant effects. An effect size <0.2 was considered trivial, from 0.2 to 0.6 small, from 0.7 to 1.2 moderate, and >1.2 large.
3. Results For DNE, no difference emerged between trials. High ICCs were observed for DT (0.97 ± 0.02) and r (0.74 ± 0.21), indicating a good intra-individual stability. In general, moderate significant correlations between mean and best DT were found (Table 1), whereas the relative picture for R always showed high and significant correlation coefficients (Table 2). For DNE, percentages of correct reproduction of rhythmic patterns are reported in Fig. 3 as a function of rhythm and movement task. Failures were observed more often for r2 with respect to r1 and r3, although no difference emerged as a function of rhythm and/or movement task. For DT, differences emerged for movement task, F(2, 24) = 18.66, p < .0001, ES range: 0.41–0.79, rhythmic patterns, F(2, 24) = 187.27, p < .0001, ES range: 1.94–2.20, and their interaction, F(4, 48) = 14.15, p < .0001. DT values were significantly higher (p < .0001) for r2 with respect to the other two rhythmic patterns. Furthermore, differences were found between all movement tasks, with higher values for walking, lower for tapping and intermediate for stepping. Fig. 4 illustrates the Rhyth-
Table 1 Correlation coefficients (r), 95% confidence intervals (95% CI), and significant values (p) between best and average DT values of for the three rhythmic patterns (i.e., r1, r2, and r3) and movement tasks (i.e., tapping, stepping, and walking). Rhythmic pattern
Movement task
r
r1
Tapping Stepping Walking
.580 .581 .733
95% CI .14–.83 .12–.84 .39–.90
p
r2
Tapping Stepping Walking
.355 .721 .994
.17–.72 .35–.90 .980.99
.177 .002 <.0001
r3
Tapping Stepping Walking
.228 .476 .959
.30–.65 .01–.78 .89–.99
.395 .053 <.0001
.015 .018 .001
209
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
Table 2 Correlation coefficients (r), 95% confidence intervals (95% CI), and significant values (p) between best and average R values for the three rhythmic patterns (i.e., r1, r2, and r3) and movement tasks (i.e., tapping, stepping, and walking). Rhythmic pattern
Movement task
r
95% CI
p
r1
Tapping Stepping Walking
.951 .796 .651
.87–.98 .50–.93 .25–.86
<.0001 <.0001 .005
r2
Tapping Stepping Walking
.971 .942 .945
.92–.99 .84–.98 .83–.98
<.0001 <.0001 <.0001
r3
Tapping Stepping Walking
.875 .995 .935
.67–.96 .98–.99 .83–.98
<.0001 <.0001 <.0001
Fig. 3. Frequency of occurrence (%) of correct execution of the number of events (DNE) for the three rhythmic patterns (i.e., r1, r2, and r3) and movement tasks (i.e., tapping, stepping, and walking).
Fig. 4. Mean ± SD of differences between the duration of subject’s performance and the total duration of the standard rhythm pattern (DT) for the three rhythmic patterns (i.e., r1, r2, and r3) and movement tasks (i.e., tapping, stepping, and walking). ⁄ Significant differences with respect to walking (p < .05) # Significant differences with respect to stepping (p < .01)
210
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
Fig. 5. Mean and SD of the ratio among rhythmic events (R) for the three rhythmic patterns (i.e., r1, r2, and r3) and movement tasks (i.e., tapping, stepping, and walking). ⁄ Significant differences between r2 and the other two rhythms (p < .001). # Significant differences between walking and the other two tasks (p < .0001)
mic Pattern Task interaction. Only for r1, stepping did not show differences with respect to walking due to a floor effect. For r2, a difference (p < .01) also emerged between tapping and stepping. For the R parameter (Figs. 5a and b), only main effects were found (movement task: F(2, 24) = 28.05, p < .0001, ES range: 0.94–1.21; rhythmic pattern: F(2, 24) = 16.011, p < .0001, ES range: 0.40–0.95). Similar to simple effects for DT, post hoc comparisons showed differences for r2 (p < .001) and W (p < .0001), with worst performances as compared to the other two rhythm and task conditions.
4. Discussion The present study aimed at assessing the effects of movement complexity on the ability of healthy adults to reproduce rhythmic patterns. The main findings of this study are: (1) the good internal stability of the rhythmic reproductions, indicating that this test is a valid instrument for the evaluation of rhythmic ability in musically illiterate adults; and (2) the good discriminative power of the qualitative aspects of rhythmic reproduction (i.e., speed of execution and temporal ratios between events), which are sensitive to the complexity of both rhythmic pattern and movement task requirements. Consistency in performance is likely to occur in experienced individuals and a learning effect could be expected in novice individuals depending on rhythmic pattern and movement task complexity. Instead, despite being musically illiterate, the participants in this study showed a good stability between multiple trials performed within a single day, with significant correlations between best and mean performance values substantiating the construct validity of the employed rhythmic test (Persichini & Capranica, 2004). The stability across trials also supports its ecological validity, since the behaviors observed and recorded (i.e., tapping, stepping and walking) required motor performances commonly used in natural settings of daily living, thus allowing generalization of the findings of this study to real world conditions. Independently of movement task complexity, DNE reproductions showed a ceiling effect, especially evident for the first and third rhythmic patterns. Conversely, DT and R parameters allowed discriminating the individual’s rhythmic ability in relation to task complexity. Specifically, they were sensitive to different dimensions of task complexity (Wood, 1986), with R being independently affected by the degree of complexity of the behavioral responses and rhythmic information cues and DT by the relationship between them. The distinctive relationship between the events influenced the individual’s capability to reproduce rhythmic patterns more than their time hierarchy (Drake, 1993; Persichini & Capranica, 2004; Sakai et al., 2004; Trehub & Hannon, 2009). Thus, the role of rhythmic pattern construct becomes crucial for the correct interpretation of temporal information and for the consequent motor reproduction. As expected (Drake, 1993; Persichini & Capranica, 2004), the simple rhythmic framework of the first pattern, characterized by two reiterated subunits (i.e., 1:1:2 ratio) with a sep-
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
211
aration of a long duration in the middle, were more easily reproduced. Despite the fact that the third rhythm presented a ternary subdivision, individuals tended to be more able to perceive and correctly reproduce its time duration and inter-event organization with respect to the binary structure of the second rhythm, which was rendered more challenging by the inclusion of a 3:1 ratio and four short (i.e., 188/188/188/188 ms) inter-event interval relationships (Fig. 5, left panel) (Drake, 1993; Large, 2000; Persichini & Capranica, 2004; Povel, 1981). Also a different degree of component redundancy (Wood, 1986) as concerns the number of different event group (four in the second rhythm and three in the third one) may have contributed to the observed differences in performance. The qualitative parameters of rhythmic ability proved to depend not only on perceptive processes but also on motor coordination (Fraisse, 1979). Regarding the total reproduction duration (reflecting movement velocity), the present data highlight a tendency to slow down in correspondence to increasing complexity of rhythmic patterns, with limited, intermediate, and longest delays being observed for the reproduction of the first, third, and second pattern, respectively. The longest delay observed in walking rhythm reproduction was paralleled by largest ratios between events for this task, suggesting a coupled influence of perceptual and motor determinants particularly in reproduction tasks involving more complex interlimb coordination and balance control. Previous findings (Persichini & Capranica, 2004) reported anticipatory responses during the first and third patterns for tapping. This phenomenon was confirmed in the present study and occurred also for stepping, indicating similar control mechanisms for upper and lower limbs when reproducing rhythmic patterns. These results are in disagreement with the hypothesis that different sensory information and transmission time differently affect hand vs. foot rhythmic reproductions. Instead, they better support a sensory accumulator model (Aschersleben, 2002; Repp, 2005) that includes different rates from different sensory channels, which effectively control different kinematics of foot versus finger tapping. It has been proposed that a broad variety of task requirements (i.e., hand tapping, foot tapping, and vocal performances) share common mechanisms of rhythmic timing involving memory processes, with rhythmic performances not completely accurate but deviating from the ideal pattern (Gilden et al., 1995; Hennig et al., 2011; Henning, Fleischmann, & Geisel, 2012; Roberts, Eykholt, & Thaut, 2000). Temporal fluctuations from the correct timing could depend on a sort of erroneous anticipatory behavior (Joiner & Shelhamer, 2009). In particular, the literature on synchronized tapping tasks reported an anticipation tendency in synchronization tasks (Ascherleben & Prinz, 1995; Fraisse, 1966; Fraisse & Voillaume, 1971; Peters, 1989; Repp, 2005), which could be reduced with increasing rhythmic complexity or subdivision (Repp, 2005). Accordingly, anticipatory responses were not present in the reproduction of the most complex structure of the second pattern, which, in contrast, tended to be delayed and showed the largest differences with respect to the standard and significant differences between the three tasks. The interaction between movement task and rhythmic pattern complexity showed the largest differences in rhythmic reproduction performance among the three movement tasks for the more complex rhythm. Also, it showed that walking task performances differed from the simpler tapping and stepping performances in all three rhythm reproductions. The ability to monitor, detect, and adjust deviations from intended behaviors is a hallmark of skilled performance (Palmer & Drake, 1997). Therefore, it is possible to speculate that a rhythmic pattern with different inter-onset intervals challenges unskilled individuals to actuate appropriate corrective strategies to avoid reproduction errors, especially when combining different kinematics of body segments in relation to space. To substantiate these findings, further investigations on perception and action coordinated in time should include populations of different age (i.e., children and older individuals) and skill level (i.e., athletes and musicians). For instance, rhythmic gymnasts and musicians might use deviations from a given rhythmic pattern to accentuate and interpret the piece. Thus, fluctuations of the R parameter, which is considered the most important indicator of proficiency in reproducing time relations among events (Krampe, Engbert, & Kliegl, 2001; Persichini & Capranica, 2004), could represent a benchmark for further generalizations or differentiations. The novel aspect of the present study was represented by the combined manipulation of the complexity of the perceptual and movement task requirements which represents a relevant dimension of the overall task complexity (Wood, 1986). The present findings highlighted a complex interplay be-
212
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
tween perceptual and motor demands in determining the individual rhythmic reproduction ability. This interplay emerged when considering the performance parameter of total duration, with largest deviations in the reproduction of the second pattern by walking. In contrast, the performance parameter of inter-event ratio did not show the expected interaction between rhythmic and motor task complexity, indicating that individuals adopt similar behaviors when challenged by major constraints. These findings might provide useful practical information about how to individualize methodological strategies in relation to age, sporting discipline, and performance level. Although walking is widely considered as an automatic motor activity requiring a limited amount of cognitive effort, explicit cognitive control is required when different constraints (i.e., obstacles, directional changes or stride time modification) are requested (Beauchet et al., 2008). In fact, studies on motor programs highlight that additional activation components overlap the existing locomotor primitivies (Ivanenko et al., 2006). The results obtained from the qualitative parameters of this test confirmed that rhythmic constraints generate a significant need for cognitive supervision during walking to properly integrate the sensory information and regulate dynamic balance and gait (Hausdorff et al., 2005). In particular, in the presence of distorted information due to extrinsic or intrinsic factors, the spontaneous walking rhythm could prevail over reproduction of an external rhythm, thus leading to a failure in rhythm reproduction. Because the individual capacity to correctly program and execute appropriate corrective strategies may have the potential to reduce reproduction errors, further research is needed to investigate the behaviors of proficient performers (i.e., athletes, dancers, and musicians), who are required to produce movements in a fluent manner according to strict rate or tempo constraints and preserving certain durational relationship between events. References Ascherleben, G., & Prinz, W. (1995). Synchronizing actions with events: The role of sensory information. Perception & Psychophysics, 57, 305–317. Aschersleben, G. (2002). Temporal control of movements in sensorimotor synchronization. Brain and Cognition, 48, 66–79. Beauchet, O., Allali, G., Berrut, G., Dubost, V., & Assal, F. (2008). Gait analysis in demented subjects: Interests and perspectives. Neuropsychiatric Disease and Treatment, 4, 155–160. Cappozzo, A., Della Croce, U., Leardini, A., & Chiari, L. (2005). Human movement analysis using stereophotogrammetry. Part 1: Theoretical background. Gait & Posture, 21, 186–196. Chen, Y., Ding, M., & Kelso, J. A. S. (1997). Long memory processes (1/f a type) in human coordination. Physical Review Letters, 79, 4501–4504. Chen, Y., Ding, M., & Kelso, J. A. S. (2001). Origin of timing errors in human sensorimotor coordination. Journal of Motor Behavior, 33, 3–8. Dalcroze, J. (2008). Il ritmo, la musica e l’educazione. Torino: Educazione Musicale EDT/SIEM. Drake, C. (1993). Reproduction of musical rhythms by children, adult musicians, and adult nonmusicians. Perception & Psychophysics, 53, 25–33. Essens, P. J. (1986). Hierarchical organization of temporal patterns. Perception & Psychophysics, 40, 69–73. Essens, P. J., & Povel, D. J. (1985). Metrical and nonmetrical representations of temporal patterns. Perception & Psychophysics, 37, 1–7. Fraisse, P. (1966). L’anticipation de stimulus rythmiques. Vitesse d’établissement et precision de la synchronisation. L’année psychologique (Vol. 66, pp. 15–36). Fraisse, P. (1979). Psicologia del ritmo. Roma: Armando Editore. Fraisse, P., & Voillaume, C. (1971). Les repéres du sujet dans la synchronisation et dans la pseudosynchronisation. L’année psychologique (Vol. 71, pp. 359–369). Fraisse, P., Pichot, P., & Clairouin, G. (1949). Les aptitudes rythmiques: Etude comparée des oligophrénes et des enfants normaux. Journal de Psychologie Normale et Pathologique, 42, 309–330. Getchell, N., Forrester, L., & Whitall, J. (2001). Individual differences and similarities in the stability, timing consistency, and natural frequency of rhythmic coordinated actions. Research Quarterly for Exercise and Sport, 72, 13–21. Gilden, D. L., Thornton, T., & Mallon, M. W. (1995). 1/f noise in human cognition. Science, 267, 1837–1839. Golubitsky, M., Stewart, I., Buono, P.-L., & Collins, J. J. (1999). Symmetry in locomotor central pattern generators and animal gaits. Nature, 401, 693–695. Hausdorff, J. M., Peng, C.-K., Ladin, Z., Wei, J. Y., & Goldberger, A. L. (1995). Is walking a random walk? Evidence for long-range correlations in stride interval of human gait. Journal of Applied Physiology, 78, 349–358. Hausdorff, J. M., Yogev, G., Springer, S., Simon, E. S., & Giladi, N. (2005). Walking is more like catching than tapping: Gait in the elderly as a complex cognitive task. Experimental Brain Research, 164, 541–548. Hennig, H., Fleischmann, R., Fredebohm, A., Hagmayer, Y., Nagler, J., Witt, A., et al (2011). The nature and perception of fluctuations in human musical rhythms. PLoS One, 6, e26457. Henning, H., Fleischmann, R., & Geisel, T. (2012). Musical rhythms: The science of being slightly off. Physics Today, 65, 64–65. IJspeert, A. J. (2008). Central pattern generators for locomotion control in animals and robots: A review. Neural Networks, 21, 642–653. Ivanenko, Y. P., Poppele, R. E., & Lacquaniti, F. (2006). Motor control programs and walking. The Neuroscientist, 12, 339–348.
F. Iannarilli et al. / Human Movement Science 32 (2013) 203–213
213
Joiner, W. M., & Shelhamer, M. (2009). A model of time estimation and error feedback in predictive timing behavior. Journal of Computational Neuroscience, 26, 119–138. Jones, M. R. (1987). Dynamic pattern structure in music: Recent theory and research. Perception & Psychophysics, 41, 621–634. Jones, M. R., & Yee, W. (1993). Attending to auditory events: The role of temporal organization. In S. McAdams & E. Bigand (Eds.), Thinking in sound: The cognitive psychology of human audition (pp. 69–112). Oxford: Clarendon Press. Keele, S. W., & Hawkins, H. L. (1982). Explorations of individual differences relevant to high level skill. Journal of Motor Behavior, 14, 2–23. Keele, S. W., Ivry, R. I., & Pokorny, R. A. (1987). Force control and its relation to timing. Journal of Motor Behavior, 19, 96–114. Krampe, R. T., Engbert, R., & Kliegl, R. (2001). Age-specific problems in rhythmic training. Psychology and Aging, 16, 12–30. Large, E. W. (2000). On synchronizing movements to music. Human Movement Science, 19, 527–566. Leman, M. (2007). Embodied music cognition and mediation technology. Cambridge, MA: The MIT-Press. McAdams, S., & Bigand, E. (1999). Introduction to auditory cognition. In S. McAdams & E. Bigand (Eds.), Thinking in sound: The cognitive psychology of human audition (pp. 1–10). Oxford: Clarendon Press. Meeuwsen, H. J., Flohr, J. W., & Fink, R. (1998). Computerized assessment of the skills of synchronization and the imitation and timing of rhythmic patterns. Technological directions in music education: Institute for music research, University of Texas at San Antonio, 93–95. Meinel, K., & Schnabel, G. (1998). Bewegungslehre – Sportmotorik [Theory of movement – motor behavior in sport] (9th ed.). Berlin: Sportverlag Berlin. Nagasaki, H. (1990). Rhythm and variability of timing in periodic tapping. Human Movement Science, 9, 177–194. Nagasaki, H., Itoh, H., Hashizume, K., & Furuna, T. (1996). Walking patterns and finger rhythm of older adults. Perception and Motor Skills, 82, 435–447. Palmer, D., & Drake, C. (1997). Monitoring and planning capacities in the acquisition of music performance skills. Canadian Journal of Experimental Psychology, 51, 369–384. Persichini, C., & Capranica, L. (2004). Reproduction of rhythmic patterns in young individuals: Toward the construction of a computerized rhythmic test. International Journal of Sport Psychology, 35, 246–265. Peters, M. (1989). The relationship between variability of intertap intervals and interval duration. Psychological Research, 51, 38–42. Philliphs-Silver, J., & Trainor, L. J. (2007). Hearing what the body feels: Auditory encoding of rhythmic movement. Cognition, 105, 533–546. Povel, D. J. (1981). Internal representation of simple temporal patterns. Journal of Experimental Psychology: Human Perception and Performance, 7, 3–18. Repp, B. H. (2005). Sensorimotor synchronization: A review of the tapping literature. Psychonomic Bulletin & Review, 12, 969–992. Roberts, S., Eykholt, R., & Thaut, M. H. (2000). Analysis of correlations and search for evidence of deterministic chaos in rhythmic motor control by the human brain. Physical Review E, 62, 2597–2607. Robertson, S. D., Zelaznik, H. N., Lantero, D. A., Bojczyk, K. G., Spencer, R. M., Doffin, J. G., et al (1999). Correlations for timing consistency among tapping and drawing tasks: Evidence against a single timing process for motor control. Journal of Experimental Psychology: Human Perception and Performance, 25, 1316–1330. Sakai, K., Hikosaka, O., Miyauchi, S., Takino, R., Tamada, T., Iwata, N. K., et al (1999). Neural representation of a rhythm depends on its interval ratio. The Journal of Neuroscience, 19, 10074–10081. Sakai, K., Hikosaka, O., & Nakamura, K. (2004). Emergence of rhythm during motor learning. Trends in Cognitive Sciences, 8, 547–553. Smoll, F. L., & Palmatier, R. A. (1977). A second-generation motor rhythm analysis system. Behavior Research Methods & Instrumentation, 9, 508–510. Styns, F., van Noorden, L., Moelants, D., & Leman, M. (2007). Walking on music. Human Movement Science, 26, 769–785. Summers, J. J. (1975). The role of timing in motor program representation. Journal of Motor Behavior, 7, 229–241. Summers, J. J., & Pressing, J. (1994). Coordinating the two hands in polyrhythmic tapping. In S. P. Swinnen, H. Heuer, J. Massion, & P. Casaer (Eds.), Interlimb coordination: Neural, dynamical, and cognitive constraints (pp. 571–591). New York: Academic Press, Inc. Tood, N. P., Cousins, R., & Lee, C. S. (2007). The contribution of anthropomorphic factors to individual differences in the perception of rhythm. Empirical Musicology Review, 2, 1–13. Trainor, L. J. (2007). Do preferred beat have rate and entrainment to beat have a common origin in movement? Empirical Musicology Review, 2, 17–20. Trehub, S. E., & Hannon, E. E. (2009). Conventional rhythms enhance infants’ and adults’ perception of musical patterns. Cortex, 45, 110–118. Wood, R. E. (1986). Task complexity: Definition of the construct. Organizational Behavior and Human Decision Processes, 37, 60–82. Zachopoulou, E., & Mantis, K. (2001). The role of rhythmic ability on the forehand performance in tennis. Physical Education & Sport Pedagogy, 6, 117–126.