Brain activity correlates differentially with increasing temporal complexity of rhythms during initialisation, synchronisation, and continuation phases of paced finger tapping
Introduction
The human motor system has the potential to orchestrate an almost infinite number of different movement sequences. These may encompass a wide range of complexities, including the number of limbs used, number of trajectories, sequence length, and relative timing of movement. Neuroimaging work has shown that the involvement of some brain regions in movement varies with some of these aspects of complexity (reviewed in Harrington et al., 2000). One important aspect which has been little investigated in terms of these variations is the temporal structure of the sequence. Rhythmic finger tapping (Wing, 2002) provides a convenient task in which to study this by varying the number of intervals in a sequence while keeping all other parameters (number of movements, mean movement frequency, number of external stimuli) the same. Previous imaging studies of externally paced rhythmic tapping, at a fixed complexity level, have shown a fairly consistent pattern of activity in the motor system. Thus, contralateral sensorimotor cortex and ipsilateral cerebellum are normally involved, with additional activation of areas such as the basal ganglia, thalamus, and sensory cortices, depending upon task specifics such as the nature of the pacing stimuli (Jancke, Loose, Lutz, Specht, & Shah, 2000a; Lutz, Specht, Shah, & Jancke, 2000; Rao et al., 1997, Rubia et al., 2000).
After repeated practice, even very complex movement sequences can be executed without overt attention and are therefore sometimes referred to as ‘automatic’ (Passingham, 1996). Because brain activity observed during overlearned movements differs from that observed during less fully learned movements (Jenkins, Brooks, Nixon, Frackowiak, & Passingham, 1994; Penhune & Doyon, 2002), it is important to avoid learning-related confounds when investigating movement related brain activity. Several studies of movement complexity have achieved this by using overlearned movement sequences (Boecker et al., 1998; Catalan, Honda, Weeks, Cohen, & Hallett, 1998; Harrington et al., 2000, Haslinger et al., 2002, Sadato et al., 1995). These authors varied different aspects of complexity and observed slightly different activity patterns for each. Dorsal premotor cortex (dPMC) and cerebellum were the most commonly reported areas, with activity in the former correlating with most types of complexity (Catalan et al., 1998, Harrington et al., 2000, Haslinger et al., 2002; Sadato, Campbell, Ibanez, Deiber, & Hallett, 1996), and in the latter specifically with the number of digits used (Harrington et al., 2000, Haslinger et al., 2002).
Our goal here was to determine how brain activity during an overlearned movement sequence varies with the temporal complexity of the sequence. We defined temporal complexity as the number of different intervals included in a measure of fixed overall duration and fixed number of elements. We used a variant (Vorberg & Hambuch, 1978) of the synchronise/continue task (Wing & Kristofferson, 1973) with auditory cues (Rao et al., 1997) in which subjects first synchronised finger tapping responses with an auditory rhythm and then continued to tap the rhythm in the absence of cues. For comparison with previous work we also included a simple contrast between activation in synchronisation and continuation phases. The synchronisation phase was preceded by a brief initiate phase, in which the subjects selected the rhythm and initiated tapping. Recent work has shown that movement selection and initiation elicits activity in areas not directly involved in movement performance (Picard & Strick, 2001; Rowe & Passingham, 2001). Thus, our analysis examined whether activity differed in the initiate and synchronise phases.
Error detection and correction processes in rhythm tracking have been shown to vary with sequence temporal complexity (Large, Fink, & Kelso, 2002). Moreover, the demand placed on movement selection mechanisms might also be expected to vary with temporal complexity. We therefore used a parametric approach to evaluate the effect of a number of rhythms varying in temporal complexity. We expected that brain activity associated with these processes should correlate with temporal complexity during selection and initiation. Once a sequence has been selected and initiated, however, it is unclear whether execution in the absence of pacing stimuli requires complexity dependent processing. Thus, it has been suggested that timing of a hierarchical rhythmic sequence, that would otherwise require several levels of embedded timekeeping, might be simplified by a process of linearization (e.g. during initiation) that allows use of a single timekeeper in execution (Vorberg & Wing, 1996). Based upon these studies (Large et al., 2002; Picard & Strick, 2001; Rowe & Passingham, 2001; Vorberg & Wing, 1996) we hypothesised that different brain regions would be activated during synchronise and continue phases of the task, and that activation correlating with difficulty might be limited to initiate and synchronise phases of the task.
Section snippets
Subjects
Ten right handed subjects, who gave informed consent, participated. The mean age was 27 years and five subjects were female. The experiment was approved by the Central Oxfordshire Research Ethics Committee.
Task
The task involved the production of temporal rhythms by tapping with the right index finger on a force sensor (response detection threshold set at 1.5 N). Each 42 s trial consisted of three phases: in the 6 s initiate phase a sequence of auditory cues (100 Hz tones of 50 ms duration, at an
Behaviour
All results are for group data averaged across the two slightly different duration ratios for each rhythm. Fig. 2 shows the mean intervals (and target intervals) in each rhythm produced by each subject during the fMRI session. The mean accuracy, averaged across intervals produced, was 89.8% during synchronise, and 89.2% during continue. A two (synchronise/continue) × four (isochronous, two-, four- and six-element rhythm) repeated measures ANOVA, performed on the mean accuracy measures showed a
Discussion
The purpose of this study was to describe the brain activity involved in production of movement sequences of varied temporal complexity, and to determine how that activity correlates with temporal complexity. We took as our measure of complexity the number of different intervals in each six-element measure. Our design allowed examination of three task phases: initiate (listening to a presented sequence and attempting to tap in time with it when ready, presumably by selecting a prelearned
Acknowledgements
This work was funded by MRC grant G9901257; P.A.L., A.M.W., P.A.P. and P.P. were supported by the same. R.C.M. is supported by the Wellcome Trust. Additional funding was provided by the MRC fMRIB unit in Oxford. We thank fMRIB staff for generous technical support and advice. We are also grateful to Nick Roach for further technical assistance.
References (40)
- et al.
The role of lateral premotor-cerebellar-parietal circuits in motor sequence control: A parametric fMRI study
Brain Research and Cognitive Brain Research
(2002) - et al.
Cortical activations during paced finger-tapping applying visual and auditory pacing stimuli
Brain Research and Cognitive Brain Research
(2000) - et al.
Cortical activations in primary and secondary motor areas for complex bimanual movements in professional pianists
Brain Research and Cognitive Brain Research
(2000) - et al.
A global optimisation method for robust affine registration of brain images
Medical Image Analysis
(2001) Location and function of the human frontal eye-field: A selective review
Neuropsychologia
(1996)- et al.
Imaging the premotor areas
Current Opinion in Neurobiology
(2001) - et al.
Working memory for location and time: Activity in prefrontal area 46 relates to selection rather than maintenance in memory
Neuroimage
(2001) - et al.
Functional frontalisation with age: Mapping neurodevelopmental trajectories with fMRI
Neuroscience and Biobehavioral Reviews
(2000) - et al.
The functional organization of the lateral frontal cortex: Conjecture or conjuncture in the electrophysiology literature
Trends in Cognitive Sciences
(1998) - et al.
Conscious and subconscious sensorimotor synchronization—prefrontal cortex and the influence of awareness
Neuroimage
(2002)