Coding of on-line and pre-planned movement sequences

Abstract

Recent experiments have demonstrated that complex multi-element movement sequences were coded in visual–spatial coordinates even after extensive practice, while relatively simple spatial–temporal movement sequences are coded in motor coordinates after a single practice session. The purpose of the present experiment was to determine if the control process rather than the difficulty of the sequence played a role in determining the pattern of effector transfer. To accomplish this, different concurrent feedback conditions were provided to two groups of participants during practice of the same movement sequence. The results indicated that when concurrent visual feedback was provided during the production of the movement, which was thought to encourage on-line control, the participants performed transfer tests with the contra-lateral limb better when the visual–spatial coordinates were reinstated than when the motor coordinates were reinstated. When concurrent visual feedback was not provided, which was thought to encourage pre-planned control, the opposite was observed. The data are consistent with the hypothesis that the mode of control dictates the coordinate system used to code the movement sequence rather than sequence difficulty or stage of practice as has been proposed.

Introduction

The manner in which movement sequences are represented, processed, and transferred has garnered a great deal of experimental and theoretical attention in the last 20 years. Hikosaka et al. (1999; also see Bapi, Doya, and Harner (2000)) proposed that the learning of a movement sequence involves both a fast developing, effector-independent component represented in visual–spatial coordinates (e.g., spatial locations of end effectors and/or sequential target positions) and a slower developing effector-dependent component that is represented in motor coordinates (e.g., activation patterns of the agonist/antagonist muscles and/or achieved joint angles). Hikosaka et al., 1999, Hikosaka et al., 2002) provide evidence that the processing of a given sequence is distributed in the brain in different forms (i.e. spatial and motor representations) with distinct neural networks to support the sequential procedures. Initially, the sequence is executed in a discrete manner for each element. With additional practice, the reliance on serial sensorimotor processes diminishes and connections are formed between the individual elements in the sequence (also see Bapi, Miyapuram, Graydon, and Doya (2006)). At this stage of practice, the sequence representation is coded in visual–spatial coordinates, which rely on attention, explicit knowledge, and working memory. This visual–spatial representation is thought to be abstract, and thus transferable to unpracticed sets of effectors resulting in relatively good effector transfer when the visual/spatial characteristics of the sequence are unchanged. Simultaneously, but more slowly, a second representation is developed which is coded in motor coordinates (e.g. agonist/antagonist muscle activation patterns and/or joint angles). Motor representations are more effector specific (Hikosaka et al., 2002) given that anatomical and neurological properties of the effector are being exploited to improve the performance (Jordan, 1995, Park and Shea, 2005). Thus, transfer to a new set of effectors based on the motor code is thought to be limited to the same activation pattern of homologous muscles. The motor code is not thought to rely on attention or working memory, but rather on implicit memory.

The model proposed by Hikosaka et al., 1999, Hikosaka et al., 2002 was developed based, to a large extent, on primate and human findings from moderately difficult multi-element discrete key press sequences (e.g. 2 × 5, 2 × 10 tasks¹) where patterns of transfer changed as practice progressed. Thus, the Hikosaka model proposed that reliance on the visual/spatial representation will gradually decrease over practice with the production of the sequential movements relying more heavily on the motor representation, which allows a more rapid and precise execution of the sequence. Recently, however, inter-manual transfer and inter-manual practice experiments have provided evidence that not all tasks follow this pattern of reliance on visual–spatial and motor representations. Kovacs, Muhlbauer, and Shea (2009), for example, had participants practice a 14-element movement sequence with either the left or the right limb for 1, 4 or 12 days (Experiments 1–3, respectively). Following the various amounts of practice, a delayed retention test and two effector transfer tests were administered. In one effector transfer test (spatial transfer), the visual–spatial coordinates were reinstated on the transfer test such that the participants moved to the same spatial locations as during acquisition, but the unpracticed limb was used. However, because the contra-lateral limb was used, a new unpracticed pattern of muscle activation and joint angles were required to achieve the target locations in the sequence. The other transfer test (mirror) also involved the unpracticed limb with a mirror presentation of the target positions, which required the same pattern of muscle activation of homologous muscles and the achievement of the same relative joint angles as during practice. Thus during the mirror transfer test, the motor coordinates were reinstated and the visual–spatial coordinates were altered, while during the spatial transfer test the visual–spatial coordinates were reinstated and the motor coordinates were altered. The results indicated that regardless of the amount of practice (1, 4, or 12 days) performance on the spatial transfer was superior to that on the mirror transfer test. In a related experiment using a similar movement sequence Panzer, Muehlbauer et al. (2009) found that inter-manual practice (practice with the right and left limbs) with the same visual–spatial coordinates resulted in enhanced learning of both the left and right limb sequences relative to when the same mirror coordinates were used during inter-manual practice.

Alternatively, recent experimental evidence suggests that the pattern of inter-manual transfer for shorter duration spatial–temporal movement sequences differs from that found for longer duration, multi-element sequences. Panzer, Krueger, Muehlbauer, Kovacs, and Shea (2009; Experiment 1) hypothesized that the production of rapid movements, which are thought to be pre-planned, may develop reliance on motor coordinates earlier in practice than longer duration movement sequences which appear to be structured into a series of subsequences and which are thought to require on-line/hierarchical control for the activation of the successive subsequences (e.g., Braden et al., 2008, Park and Shea, 2005, Wilde and Shea, 2006). Using a retention and transfer design similar to the one discussed above (Kovacs, Muehlbauer et al., 2009), Panzer, Krueger et al. (2009; Experiment 1) asked the participants to reproduce a waveform with 3 reversals and 1300 ms duration. Their findings after one practice session indicated that transfer to a mirror condition with the contra-lateral limb where the same pattern of homologous muscle activation and joint angles (motor coordinates) were required as those used during acquisition resulted in superior transfer compared with the non-mirror transfer of the contra-lateral limb where the same display (visual–spatial coordinates) was used. In Experiment 2, inter-manual practice was provided with the same visual–spatial coordinates or the same motor coordinates on the practice session with the right and left limbs. They found enhanced learning when the same motor coordinates were available during practice with each limb relative to when the same visual–spatial coordinates were reinstated.

In sum, the inter-manual practice and transfer findings, at least at first glance, appear to suggest that the amount of practice may not be as important as task characteristics (complexity) in determining the most effective inter-manual transfer and practice conditions. However, it is possible that the patterns of inter-manual practice and transfer are not dependent on the difficulty of the movement sequences per se, but on the control mechanisms utilized to produce the sequence (Glover, 2004). Pre-planning and on-line control of movement sequences have been shown to utilize different information and rely on different neural pathways (e.g., Glover, 2004, Hikosaka et al., 2002). Furthermore, planning determines the initial kinematic characteristics of the movement including timing and velocity while the on-line control system monitors and occasionally adjusts movement progress “in flight” but these adjustments are limited to the spatial characteristics of the movement (Glover, 2004). In other words, shorter duration movements with few elements predominantly rely on pre-planning while longer duration movements with more elements have an initial pre-planned component after which movement control is gradually taken over by the on-line control mechanism (Kovacs, Han, & Shea, 2009). Indeed, Smiley-Oyen and Worringham (2001) have demonstrated that movement trajectory, average velocity and number of targets influence planning. These findings suggest that sequential tasks with dissimilar spatial–temporal characteristics might also differentially rely on distinct control mechanisms. Note, however, that the dichotomy of these two control mechanisms is not so evident and the difficulty of the movement sequence may to some degree co-vary with the control processes. According to Glover (2004), the transitioning from one (pre-planned) to the other (on-line control) mechanism has a gradual crossover with a significant temporal overlap. Additionally Hikosaka et al. (1999) noted that reliance on visual–spatial coordinates is slower and less automated (on-line control) than reliance on motor coordinates which supports faster, more automatic (pre-planned) production of the movement sequence.

The primary goal of this experiment was to determine the pattern of inter-manual transfer of the same relatively simple dynamic arm movement sequence when visual information provided during movement production is manipulated with the intent to alter the degree to which on-line and pre-planned control is used by the participants to produce the movement sequence. The task involved moving a right lever by extending or flexing the right arm at the elbow in the horizontal plane or moving a left lever by flexing or extending the left arm. The lever movement was monitored by a potentiometer which was used to record the movement and provide feedback. In an on-line control condition the participants were provided a visual display of the goal movement sequence prior to, during, and after the movement sequence was completed. In addition, a cursor (dot) provided the current position of the limb throughout the production of the movement. Presumably, providing the goal movement pattern and cursor indicating the position of the lever will enhance the chances of participants using the discrepancy between the cursor and the goal movement pattern to effect on-line changes in the movement trajectories. In a pre-planned control condition, the participants were provided a visual display of the goal movement sequence prior to and after the movement sequence was completed, but not during the movement. In addition, the cursor (dot) which provided the current position of the limb was not displayed during the production of the movement. Presumably, eliminating this information during the production of the movement would reduce the participant’s ability to effect on-line corrections and encourage pre-planned control. Note that the pre- and post-movement information were identical for the two groups.

We predicted that the participants in the on-line control would perform the non-mirror transfer test better than the mirror transfer test with the opposite pattern of effector transfer for the participants in the pre-plan control condition. We make this prediction because we hypothesize that the participants provided feedback during the production of the movement will be more likely to engage in on-line control process which are thought to be coded in visual–spatial coordinates. Thus, reinstating visual–spatial coordinates on the transfer test (non-mirror test) should enhance the transfer performance relative to reinstating the motor coordinates (mirror test). Alternatively, we predict that the participants not provided feedback during the movement will be more likely to pre-plan their movement, which are thought to be coded in motor coordinates. Thus, reinstating the motor coordinates (mirror test) should enhance the transfer performance relative to reinstating the visual–spatial coordinates (non-mirror test). This finding would be consistent with the hypothesis that the control processes rather than stage of practice or sequence difficulty are the prime determiner of the salient coding reference system although control processes may also co-vary with these factors.