HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when Rapid Responses are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Shea, C. H Articles by Wilde Braden, H. Search for Related Content PubMed PubMed Citation Articles by Shea, C. H Articles by Wilde Braden, H.

Age-Related Effects in Sequential Motor Learning

CH Shea, PhD, is Professor, Department of Health and Kinesiology, Texas A&M University, College Station, TX 77843-4243 (USA)
JH Park, PhD, is Assistant Professor, Physical Education Department, Seoul University, Seoul, South Korea
H Wilde Braden, PT, PhD, is Assistant Professor, Department of Physical Therapy, The University of Texas Health Science Center, San Antonio, Tex

(cshea{at}tamu.edu). Address all correspondence to Dr Shea

Submitted October 27, 2004; Accepted November 17, 2005

	Abstract

 
Background and Purpose. When learning multi-element movement sequences, participants organize individual elements into subsequences. Imposing this type of structure on the elements leads to the efficient production of sequences because the processing of all but the first elements in a subsequence can be completed prior to their execution. The primary purpose of this study was to determine whether older adults organize lengthy movement sequences with the same efficiency as young adults. Subjects and Methods. Participants were young adults (N=8, 19–23 years of age) and older adults (N=8, 65–68 years of age). The task required participants to move a lever as quickly as possible to targets sequentially projected on a tabletop. At various stages during practice, random practice blocks were inserted between the repeated sequence blocks. Repeated and random sequence retention tests were administered after 24 hours. Results. The results indicated that the young adults performed the repeated sequences substantially faster than the older adults and that this difference increased over practice. On the retention tests, there were no differences in response time for the random sequence blocks, but the young performers were substantially faster than the older performers when repeated sequences were used. No differences were detected in the interview or on the recognition (²=1.22, P>.05) and completion (²=0.89, P>.05) tests designed to determine explicit or implicit knowledge of the sequences. Discussion and Conclusion. Analysis of the sequence structure indicated that the older adults did not organize their responses into subsequences as effectively as the young adults. The failure of older adults to optimally organize movement sequences may contribute to the overall slowing of sequential movement production. [Shea CH, Park JH, Wilde Braden H. Age-related effects in sequential motor learning. Phys Ther. 2006;86:478–488.]

Key Words: Movement skills in older adults • Physical therapy • Sequential motor learning

	Introduction

Top Abstract Introduction Explanations for Age-Related... Structure of Movement Sequences Implicit Versus Explicit... Predictions and Design Features... Method Results Discussion and Conclusions References

 
In the clinic, physical therapists spend substantial amounts of time training and retraining older adults to regain safety, balance, and function with activities of daily living skills that require efficient motor responses. Despite the time that physical therapists spend on these motor skill activities, few studies have assessed the manner in which older adults process, organize, and plan such sequential motor skills in order to create the most efficient outcome.¹ The present study assessed the ability of older adults to effectively process, organize, and plan sequential motor tasks. The role of implicit and explicit processing also was explored.

	Explanations for Age-Related Deficits: Motor Slowing Versus Processing Deficits

Top Abstract Introduction Explanations for Age-Related... Structure of Movement Sequences Implicit Versus Explicit... Predictions and Design Features... Method Results Discussion and Conclusions References

 
Aging appears to affect, to various extents, all aspects of human endeavors, including cognitive and motor functions. Age-related effects on cognitive performance have been studied a great deal.^1–3 The pattern, perhaps oversimplified, that emerges is that older participants generally perform more poorly than do young adults on cued and free recall tasks but that recognition tasks typically are not affected.^4–6 The effects of aging on basic motor function also have been extensively studied.^7–10 It is clear that neuromuscular capabilities functionally and structurally decline with advancing age, limiting maximal but typically not submaximal motor capabilities.⁷ However, decrements in motor skill performance and the ability of aging adults to learn new skills or relearn old motor skills, or both, have not been studied as extensively; the effects of aging on the various processes involved in retrieving, organizing, executing, and learning motor tasks are not clearly understood.

The strategy used in the present study was to compare performance when participants were required to respond as rapidly as possible to random stimulus sequences (for which the contributions of advance planning and organization of the movement patterns are minimized) with performance on repeated movement sequences (for which participants typically engage in additional cognitive processing activities designed to optimize movement speed through the advance planning and effective organization of the elements in the movement sequences). Because the motor demands are very similar between random and repeated sequences but the processing demands are different, this design should provide a reasonable (within-participant) comparison of the motor and processing accounts of aging deficits.

	Structure of Movement Sequences

Top Abstract Introduction Explanations for Age-Related... Structure of Movement Sequences Implicit Versus Explicit... Predictions and Design Features... Method Results Discussion and Conclusions References

 
Sequential tasks are particularly interesting theoretically because detailed analyses of the elements comprising a sequence and the transitions between the elements often reveal markers indicating the presence of subsequences (termed "motor chunks")¹¹ and the transition from one subsequence to another.^12–15 The patterns of element duration (time from 1 element to another) are used to infer the structure of a movement sequence.^12,16 Young adults typically organize the elements comprising the sequence into a series of relatively contiguous subsequences.^14,15 This strategy reduces processing and memory demands and permits the parallel processing of the sequence information and the articulatory commands responsible for producing the elements in the sequence.^17,18

A popular way to study the structure of sequential movements over the last 20 years has been to have participants sequentially depress keys corresponding to visual signals presented on a computer monitor.^12,17 In serial reaction time experiments, it has been relatively common to include a control group,¹³ control segments,¹⁹ or control blocks^14,20 composed of randomly presented elements throughout practice. The reduction over practice in response time for the random sequence indicates general improvements in the capability to respond. The reduction in performance time for the repeated sequences relative to the random sequences is used as an index of the effectiveness of the structure imposed on the sequences.

	Implicit Versus Explicit Learning of Motor Sequences

Top Abstract Introduction Explanations for Age-Related... Structure of Movement Sequences Implicit Versus Explicit... Predictions and Design Features... Method Results Discussion and Conclusions References

 
Research over the last 20 years has demonstrated that motor skill learning can occur at both explicit and implicit levels, at least for some tasks. Explicit learning is characterized by effortful, conscious, and verbally describable activities that are directed at learning a particular task. In contrast, implicit learning is characterized by a dissociation between performance of the task and indicators of conscious effort or knowledge.^21–25 Several studies from the motor behavior literature have shown that learners can acquire knowledge of stimulus regularities in the environment without directing conscious effort to this specific aspect of the task and without explicit knowledge of these regularities.^19,26–28 The finding that the learners are not aware that they have acquired this knowledge and are not able to verbalize the regularities upon prompting, even though their performance indicates that they exploit the regularities in the stimulus environment, has been taken as convincing evidence of implicit learning. This finding has been reported for a number of tasks, including sequence learning,^13,28 visual target search,^29,30 computer-implemented control tasks,³¹ and pursuit tracking tasks.^{19,26,27,32–34}

Hoyer and Lincourt³⁵ concluded in their review of learning and aging that experiments with older participants consistently have revealed "age-related differences in determining higher-order dependencies." This apparent interaction between aging and implicit or explicit learning led Willingham⁸ to pose the following question: "Does the implicit/explicit distinction capture the age-related deficit on some motor skill tasks?" This question was proposed in part because in serial reaction time tasks with repeated sequences, fewer older participants than young participants exhibited awareness of the repeated sequences. However, the older adults were successful in improving their performance, presumably demonstrating implicit sequence knowledge. Although Willingham concluded that age-related processing deficits cannot be explained on the basis of the implicit/explicit distinction alone, it is important in studies of age-related learning and performance characteristics to attempt to determine whether the contrasted age groups differentially rely on implicit memory systems.

	Predictions and Design Features of the Present Study

Top Abstract Introduction Explanations for Age-Related... Structure of Movement Sequences Implicit Versus Explicit... Predictions and Design Features... Method Results Discussion and Conclusions References

 
The present study was designed to investigate age-related deficits in the performance and learning of movement sequences. Through comparison of sequence structure and performance on random and repeated sequences across age groups, it should be possible to determine whether age-related performance and learning differences arise from general factors related to motor slowing, from more specific processing-related deficits associated with planning and organizing movement sequences, or from both. If differences are observed between age groups only on the repeated sequence, then processing deficits are implicated as the source of the differences. We used various postexperiment methods to determine the degree to which participants acquired explicit knowledge of movement sequences.

There are 2 relatively unique design features in this study. First, we used a continuous, dynamic arm movement task that allows both discrete and continuous measurements of performance. Many of the earlier sequential learning experiments related to aging used serial reaction time (key-pressing) tasks designed to minimize motor control. Second, we assessed delayed retention on the repeated and random sequences in an attempt to isolate specific processing and memory coding deficits, if any, that are related to learning the repeated sequences and that are a result of the aging process.

	Method

Top Abstract Introduction Explanations for Age-Related... Structure of Movement Sequences Implicit Versus Explicit... Predictions and Design Features... Method Results Discussion and Conclusions References

 
Participants

Young adults (N=8; 19–23 years of age) and older adults (N=8; 65–68 years of age) participated in the study. All participants had no history of neurologic disease or musculoskeletal dysfunction. The participants had no prior experience with the experimental task and were not aware of the specific purpose of the study. All participants were right-hand dominant, as determined by self-report prior to the experiment. Informed consent was obtained prior to participation in the experiment. Participants were recruited from university employees and faculty. All of the older participants were still actively involved in their profession, and all had completed graduate degrees in their respective fields. The older participants would be classified as "higher ability" according to the classification used by Cherry and Stadler.³⁶

Apparatus

The apparatus consisted of a horizontal lever affixed at 1 end to a nearly frictionless vertical axle. The axle, which rotated freely on ball bearing supports, allowed the lever to move in the horizontal plane over the table surface (Fig. 1A). Near the distal end of the lever, a vertical handle was attached. The position of the handle could be adjusted so that when the participant rested a forearm on the lever, the elbow aligned over the axis of rotation and the participant could comfortably grasp the handle (palm vertical). Extending from the lever was a pointer. The horizontal movement of the lever was monitored (100 Hz) by a potentiometer that was attached to the lower end of the axle. The data were used online to determine when target positions were achieved and were stored for later analysis on an IBM-compatible computer. The targets and total movement time were displayed on the tabletop by a projection system mounted above the table.

View larger version (20K): [in this window] [in a new window]   Figure 1. (A) Illustration of the apparatus and targets. The target positions were projected onto the tabletop from above. The illumination of the target (filled circle) indicated the next target in the sequence. (B) Example of a time series for a young participant’s responses to the repeated sequence midway through the practice session. Circles indicate positions of targets.

A 16-element sequence (targets 2, 3, 4, 3, 2, 3, 2, 1, 2, 3, 2, 3, 4, 3, 2, and 1) was repeated in acquisition blocks 2 to 4, 6 to 8, 10 to 12, and 14 to 16. Blocks consisted of 10 repetitions of the sequence, resulting in 160 targets. An example of the movement pattern created by this sequence is shown in Figure 1B. Random sequences were used in blocks 1, 5, 9, and 13. In these blocks, the targets were illuminated in a quasi-random order created by reordering the targets in the repeated sequence from repetition to repetition. Thus, the same targets were presented in the random and repeated sequences. This design is important so that lower-order sequence information related to the frequency with which specific targets were presented could be accounted for by performance on random sequence blocks. Participants were not provided any information about the random or repeated sequences. A rest interval of 30 seconds was provided after each block.

Repeated and random sequence tests were conducted approximately 24 hours after the completion of the acquisition session. The task and sequence for the repeated and random sequence tests were exactly the same as on day 1. These retention tests provided a measure of learning. After completing the repeated and random sequence retention tests, participants took a short break, and then an interview and 2 postexperiment tests (completion and recognition) were administered. The order of the 2 postexperiment tests was counterbalanced across participants to avoid possible testing and relearning effects. A computer and standard keyboard were used for the completion and recognition tests. For the completion test, participants had to use the "1," "2," "3," and "4" number keys, and for the recognition test, they were directed to use "Y" for "Yes" and "N" for "No."

Interview
The interview began with the experimenter asking participants to describe the task and any strategies that they used. If sequence information was forthcoming, participants were queried in order to determine the degree of precision with which they could describe the repeated sequences. Next, participants were asked to rate their confidence in having noticed a sequence in some of the blocks during the practice and testing phases of the experiment. Participants rated their confidence from 1 (highly confident that a sequence was not present in any blocks) to 7 (highly confident that a sequence was used in some blocks). Participants who had not provided sequence information earlier in the interview were encouraged to describe the sequence(s) if they believed any were used. At the end of the interview, participants were informed that a repeating sequence was used in 12 of 16 acquisition blocks and in 4 of 6 test blocks.

Completion test
In the completion test, participants were asked to watch 5 sequences of target illuminations projected on the tabletop. The participants were reminded that the numbers "1" to "4" represented the target positions. The program started from the nth element (for example, the 9th element) in the sequence and stopped after the completion of 3 repetitions of the entire sequence. Participants were asked to respond with the number (1–4) of the next target in the sequence to be illuminated. When the next target was selected, the sequence continued. Completion of the test involved stopping at random locations in the sequence a total of 5 times. This procedure was repeated a second time but involved starting at a different point in the sequence.

Recognition test
In the recognition test, participants were seated in front of the apparatus with their arms positioned as if they were to start a practice block. Five different demonstration sequences consisting of 5 repetitions of different repeated sequences were shown. The demonstration sequence involved sequentially illuminating the targets on the tabletop at 500-millisecond intervals. After each presentation, the participants were instructed to answer with "Yes" if the sequence that they just viewed was the one that they had practiced during acquisition; otherwise, they were to respond "No." Only 1 of the 5 demonstration sequences was exactly the same as the sequence that they had learned (sequence 3). Two of the sham sequences had changes in only 2 elements in the sequence (sequences 2 and 4), and the other 2 had additional changes (sequences 1 and 5). No feedback was provided concerning whether or not their answers were correct or incorrect. This procedure was repeated a second time with a different order of demonstration programs. The order in which the sequences were presented was randomized—the labels "1" to "5" denote only the sequence number.

Data Analysis

The dependent variable for acquisition and retention analyses was element duration. Element duration was computed as the elapsed time from "hitting" the previous target (crossing the target edge) to "hitting" the current target. This measure characterized the time required to transition from 1 target to the next. To analyze acquisition performance on the random and repeated sequences, we analyzed mean element duration with a 2 (practice group: young or older) x 16 (blocks: 1–16) analysis of variance with repeated measures on block. Performance on the repeated and random sequence retention tests was analyzed with a 2 (practice group: young or older) x 3 (test: repeated or random) x 16 (element: 1–16) analysis of variance with repeated measures on test and element. Element was included in the retention test analysis in order to determine whether the response structures used in the repeated and random sequence retention tests differed. The postexperiment recognition and completion tests were analyzed with separate practice group x response frequency chi-square analyses.

	Results

Top Abstract Introduction Explanations for Age-Related... Structure of Movement Sequences Implicit Versus Explicit... Predictions and Design Features... Method Results Discussion and Conclusions References

 
Mean element duration is displayed by group across acquisition and test blocks in Figure 2. Mean element durations for young and older participants by acquisition blocks (blocks 2, 3, 4, and 6) early in practice are displayed in Figure 3. This illustration shows the manner in which element durations changed as a function of "chunking" elements into subsequences. Results of the postexperiment tests are provided in Figure 4.

View larger version (14K): [in this window] [in a new window]   Figure 2. Mean element duration for young and older adults during acquisition (blocks 1–16) and retention (Rep=repeated sequence, Ran=random sequence). Repeated sequences were used in all blocks except blocks 1, 5, 9, and 13, in which random sequences were presented. Error bars represent standard errors.

View larger version (18K): [in this window] [in a new window]   Figure 3. Examples of patterns of element durations for young and older participants early in practice (blocks 2, 3, 4, and 6). Block 5 was not included because a random sequence was presented in this block. Error bars represent standard errors.

View larger version (35K): [in this window] [in a new window]   Figure 4. Results of confidence ratings (A), recognition tests (B), and completion tests by repetition (C). On the recognition tests, demonstration sequence 3 was the sequence used during testing (correct response). Demonstration sequences 2 and 5 differed from the correct sequence in only 2 of the 16 elements. Error bars represent standard errors.

Participants in both groups demonstrated shorter element durations on the repeated sequence blocks than on the random sequence blocks, and performance on the repeated sequences (Fig. 3, blocks 2–4, 6–8, 10–12, and 14–16) improved over practice, with the largest improvements early in practice. However, the older participants performed the repeated sequences substantially more slowly than the young participants, although the performance of the older participants on the random sequences did not differ from that of the young participants.

Retention: Repeated and Random Sequences

As in acquisition, the young participants performed the repeated sequences on the retention test more quickly than the older participants, but the performance time on the random sequences did not differ across age groups. The main effects of age (F_1,16=4.63, P<.05), test (F_2,16=81.29, P<.01), and elements (F_15,240=11.72, P<.01) were significant. In addition, the age x block (F_1,16=32.89, P<.01), age x element (F_15,240=5.08, P<.01), block x element (F_30,480=7.44, P<.01), and age x block x element (F_30,480=10.33, P<.01) interactions were significant.

Simple main-effects analysis failed to detect differences between the young and older participants in performance on the random sequences. The performances of the young and older participants did differ on the repeated sequences. The performance of the young group was significantly faster than that of the older group on the repeated sequences. Simple main-effects analysis of the age x element interaction indicated that the young participants responded more slowly to elements 3, 6, 11, and 16 than all other elements in the sequence on the repeated sequence retention test. On the other hand, for the older participants, no elements were responded to more slowly than all other elements in the sequence on this test. No element differences were detected on the random sequences for either age group.

Interview and Postexperiment Tests

During the interview process, all participants indicated that they were aware of the repeated sequences. Indeed, most of the participants provided the experimenter with detailed information about the repeated sequences, and this information did not differ across age groups. Although the older participants verbally indicated slightly less confidence in the presence of the repeated sequences than the young participants, the analysis of the confidence rating indicated no differences between the age groups (t₁₄=1.1, P>.05). Similarly, no differences in the patterns of responses for the young and older participants were detected on the completion tests (²=1.22, P>.05) or the recognition tests (²=0.89, P>.05) (Fig. 4).

	Discussion and Conclusions

Top Abstract Introduction Explanations for Age-Related... Structure of Movement Sequences Implicit Versus Explicit... Predictions and Design Features... Method Results Discussion and Conclusions References

 
Performance on the random sequence blocks was used as an index of general performance capabilities unrelated to the processing and organization of sequence information. Young and older participants executed the random sequences similarly during both acquisition and random retention testing (Fig. 2). The patterns of improvement across acquisition on the random sequences also were similar for the 2 groups. This finding suggests that there are very small, if any, differences in nonspecific performance and learning characteristics between young and older adults when advance planning and response organization requirements are relatively low. Importantly, this finding also suggests that the degree of motor slowing for older participants, if any, cannot account for differences in performance and learning across age groups for this type of controlled movement sequence task.

In acquisition, however, all participants performed the repeated sequences more quickly than the random sequences. This result suggests that sequence order information acquired during practice provided a basis on which to enhance response production. Furthermore, the advantages evident for the repeated sequences relative to the random sequences provide an index of the effectiveness of the advance processing of the sequence order information that is independent of nonspecific cognitive and motor processing. In fact, because the random sequences comprised the same distribution of targets as the repeated sequences, lower-order target probability information also was the same across the random and repeated sequence blocks and thus cannot explain the performance advantages observed on the repeated sequences. The differences in acquisition between the young and older participants on the repeated sequences, however, were relatively large by the end of acquisition (young: =302 milliseconds, SE=9.4; older: =440 milliseconds, SE=8.8). Although the older participants performed the repeated sequences more quickly than the random sequences, they were relatively ineffective (at least compared to the young participants) in using the sequence order information to decrease element duration. The results of the repeated and random sequence retention tests mirrored the findings at the end of acquisition. As in acquisition, the participants performed the random sequences in retention more slowly than the repeated sequences, with no difference between the young and older participants in performance on the random sequences. On the repeated sequence retention test, the young participants performed the repeated sequence, which was practiced during acquisition, substantially faster than the older participants.

Implicit Versus Explicit Learning

One potential explanation for the age-related performance differences on the repeated sequences centers on the distinction between implicit learning and explicit learning. More specifically, there may have been differences across age groups in the explicit or implicit types of learning that occurred. Implicit learning is characterized, for example, by improvement in performance when a repeated sequence is used but the participant lacks explicit knowledge of the sequence. We expected a relatively complex 16-element movement sequence, with random sequences interpolated after every 3 blocks, to have been learned at an implicit level, especially for the older participants. Nissen et al,³⁷ for example, using a shorter 12-element serial reaction time task, found participants to be unaware of the repeated sequence, although their performance demonstrated that they had acquired some sequence knowledge. Likewise, Cherry and Stadler³⁶ found that high-ability older participants implicitly learned a serial reaction time task with levels of performance attainment and levels of implicit knowledge similar to those of young participants.³⁸ However, the participants, young and older, in the present study were able to demonstrate explicit knowledge of the sequence order. In the postexperiment interview, participants in both age groups indicated that they were aware of the repeated sequence. In most cases (5 older and 6 young participants), without prompting, participants indicated explicit knowledge of the sequence order. They indicated their sequence knowledge by either spontaneously sequentially pointing to the targets or moving their arm through the movement path. Other participants did not volunteer this information but could, upon prompting during the interview process, describe the movement sequence, the pattern of movement reversals, or both. The analysis of the confidence ratings failed to detect differences between the young (=6.75, SE=0.75) and older (=6.75, SE=0.46) participants. The results of the completion and recognition tests also support the notion that participants acquired substantial sequence knowledge and that this knowledge did not differ across young and older age groups (Figs. 4B and 4C). On the completion test, the responses of the older participants were actually correct more often, on average, than those of the young participants, but this difference was not significant.

However, it is important to take some care in interpreting the distinction between implicit learning and explicit learning with respect to response production. Typically, researchers have assumed that when participants are capable of expressing explicit knowledge, they must be using the explicit knowledge to produce the movements.^26,31 If explicit knowledge and implicit knowledge both accrue as the result of experience but their dominance in terms of guiding movements changes over practice,^4,8 it is possible that participants are capable of explicitly expressing accurate information about the sequence but use implicit knowledge and rules to guide their responses. This notion is consistent with reports from both young and older participants in this study, who stated that they seemed to perform more slowly when they were thinking about what to do than when they simply responded. In addition, it is possible that participants rely on implicit knowledge to produce movements but can generate explicit knowledge about the sequence by actually producing the movement pattern. Indeed, a number of participants in the interview described the movement sequence by making hand or arm movements rather than through verbal description.

Structure of Movement Sequences

A second potential explanation for the age-related performance differences is that older participants may have been as effective in acquiring sequence knowledge as young participants but were simply less effective in using this information. The data do provide support for this explanation for a number of reasons. The structure imposed on the elements in the sequence by the young participants but not the older participants appears to be a likely candidate for the locus of the differences between age groups in performance on the repeated movement sequence. The development of this structure for the young participants was seen clearly early in acquisition. The differences between age groups in terms of overall time required to complete the sequence and the time required to complete the individual elements comprising the sequence were relatively small on the first block of practice with the repeated sequence (Fig. 3, block 2). However, on the next block (block 3), the young participants began to show a decrease in duration for some elements but not others.

This pattern of selective improvement suggests that participants were beginning to impose a structure on the elements comprising the sequence. That is, young participants began grouping elements into functional subsequences (motor chunks).^11,12,18 Subsequences are characterized by a relatively slow response to 1 element followed by a faster response to 1 or more subsequent elements.^12,17 The response to the first element is slower because of the additional processing required to retrieve, organize, and initiate the subsequence, but the subsequent elements are produced more rapidly because the articulatory activities required for their execution are completed in advance. On block 3 (Fig. 3), elements 3, 6, 8, 12, and 14 were produced somewhat more slowly than 1 or more of the following elements. Indeed, the major decreases in movement time across blocks for the young participants compared with the older participants from block 2 to block 3 accrued from the elements within the subsequences and not from the elements that marked the beginning of the subsequences. On block 4, decreases in movement time again were accomplished by further decreasing the time required to traverse the elements within the subsequences. Interestingly, there appeared to be further consolidation of subsequences in block 4. This consolidation process is best seen for element 8, which appeared initially in blocks 2 and 3 as if it were the beginning of an independent subsequence but, by the end of block 4, appeared to be consolidated into the preceding subsequence. Consolidation also was seen in block 6, in which a rather dramatic decrease for element 14 was observed. It is also important that there was little difference between the young and older participants for elements 3, 6, 11, and 16—elements that marked the beginning of subsequences. Further, the repeated sequence continued for 160 elements (10 repetitions of the 16-element sequence) within each block. This design allowed participants to group elements from the end of a sequence with elements at the beginning of the sequence. Indeed, such grouping appeared to occur with elements 16, 1, and 2 being grouped together in a single subsequence.

Importantly, there was little indication that the older participants attempted to impose a structure on the sequence; therefore, they treated each element in the sequence as an individual response (Fig. 3). Knowledge of the sequence order allowed them to respond more quickly when elements were repeated than when the sequence order was random but did not permit the efficiency of movement processing and execution provided by the development of an effective response structure. The end result was substantially slower response in performance of the repeated sequence for the older participants than for the young participants.

Theoretical and Clinical Considerations

The present findings are particularly interesting for theoretical reasons because the results suggest that a particular class of processing related to structuring of a sequence may be responsible for at least some of the age-related deficits associated with producing dynamic arm movement sequences. Recently, the study of sequential movements was viewed in terms of independent, perhaps parallel, processing mechanisms^17,2039–41: one processing mechanism responsible for planning and organizing the elements in the sequence and the other responsible for the articulatory activities required to effect the planned action. Verwey,¹⁷ for example, proposed a cognitive mechanism, which plans and represents the sequence, and a motor mechanism, which formulates the specific commands required to carry out the desired sequence. An interesting feature of Verwey’s parallel dual-processor model is the proposal that the cognitive and motor mechanisms not only are independent but also can operate in parallel. Thus, when a learned movement sequence is represented and executed as a series of subsequences (motor chunks), the planning of the next subsequence can be carried out while the current subsequence is being executed. This model is different from other, more serial dual-processor models,^20,40,42 in which the processing related to the sequence organization is completed prior to the initiation of the movement sequence (ie, preprogrammed) and, therefore, processing at one level is relatively independent of processing at other levels. In Verwey’s terms, the present data suggest that the age-related processing deficits are localized to the cognitive processor and could be a result of any or all of the following: reduced memory capacity, slowed cognitive processing, interference or simultaneity problems for processing activities, which would reduce the capabilities of the cognitive processor. Each of these types of potential problems has been noted in the cognitive functioning of aging adults.^1,2 Salthouse,¹ for example, argued that decreases in processing speed associated with aging lead to impairments in cognitive functioning by means of the "limited time" mechanism and the "simultaneity" mechanism. That is, performance is degraded because important processing cannot be completed in a timely manner (limited time) and because the results of advance processing may no longer be available when needed for subsequent processing or execution (simultaneity). In the present study, the failure of the older participants to develop a movement structure based on subsequences may have been a result of their inability to maintain the overall movement structure information, the advance planning instructions, or both while engaging in processing related to individual subsequences.

Finally, our findings indicate the need to develop clinical learning protocols for older adults that can enhance their ability to structure sequential responses. Two potential training protocols come to mind. First, Park et al¹⁵ recently demonstrated subtle advantages of part-whole practice of movement sequences. Part-whole practice involves practicing 1 or more parts of a movement sequence before practicing the whole movement sequence. The vast majority of research on part-whole practice has failed to find important advantages of this type of practice.^43,44 However, Park and colleagues, using sequences similar to those used in the present study, found a number of subtle differences between part-whole practice and whole practice that suggest that part-whole practice may be an effective method for enhancing a participant’s ability to structure movement sequences by reducing the overall processing demands to a more manageable level. This strategy allows subsequences to be developed without the interference associated with simultaneously holding the other parts of the sequence in short-term memory. Second, Verwey¹¹ introduced brief delays between the presentations of items in a sequence early in practice. He noted that the delays provided additional time for the participants to "chunk" the items together into functional subsequences, resulting in a movement structure different from that developed by participants not exposed to the delays. Perhaps these types of training protocols could assist older participants in adopting an effective movement structure. Evidence regarding the processing of motor sequences by older adults may be an important consideration for therapists when designing rehabilitation approaches involving learning and relearning skilled motor tasks.

	Footnotes

 
All authors provided concept/idea/research design, writing, and data collection. Dr Shea provided data analysis, facilities/equipment, and project management. Dr Park and Dr Wilde Braden provided subjects. Dr Shea and Dr Wilde Braden provided consultation (including review of manuscript before submission).

The study was approved by the Institutional Review Board at Texas A&M University.

	References

Top Abstract Introduction Explanations for Age-Related... Structure of Movement Sequences Implicit Versus Explicit... Predictions and Design Features... Method Results Discussion and Conclusions References

 

1. Salthouse TA. The processing-speed theory of adult age differences in cognition. Psychol Rev.130: 1996;403423.
2. Salthouse TA, Miles JD. Age and time-sharing aspects of executive control. Mem Cognit. 2002;30:572–582.[ISI][Medline]
3. Stelmach GA, Nahom A. Cognitive-motor abilities of the elderly driver. Human Factors. 1992;34:53–65.[ISI][Medline]
4. Willingham DB. A neuropsychological theory of motor skill learning. Psychol Rev. 1998;105:558–584.[ISI][Medline]
5. Craig FM. Memory changes in normal aging. Current Directions in Psychological Science. 1994;3:155–158.[ISI]
6. Welford AT. Between bodily changes and performancesome possible reasons for slowing with age. Exp Aging Res. 1984;10:73–88.[ISI][Medline]
7. Delbono O. Neural control of aging skeletal muscle. Aging Cell. 2003;2:21–29.[ISI][Medline]
8. Willingham DB. On the creation of classification systems of memory. Behav Brain Sci. 1994;17:426–427.[ISI]
9. Gutman GM. The effects of age and extraversion on pursuit rotor reminiscence. J Gerontol. 1965;20:346–350.[ISI]
10. Ruch FL. The differential effects of age upon human learning. J Gen Psychol. 1934;11:261–286.
11. Verwey WB. Evidence for the development of concurrent processing in a sequential key pressing task. Acta Psychol. 1994;85:245–262.
12. Povel D, Collard R. Structural factors in patterned finger tapping. Acta Psychol. 1982;52:107–123.[Medline]
13. Nissen MJ, Bullemer P. Attentional requirements of learning: evidence from performance measures. Cognit Psychol. 1987;19:1–12.[ISI]
14. Park JH, Shea CH. Sequence learning: response structure and effector transfer. Q J Exp Psychol. 2004;36:51–61.
15. Park JH, Wilde H, Shea CH. Part-whole practice of movement sequences. J Mot Behav. 2004;36:51–61.[ISI][Medline]
16. Rosenbaum DA, Saltzman E. A motor-program editor. In: Prinz W, Sanders A, eds. Cognition and Motor Processes VIII. Berlin, Germany: Springer; 1984:93–106.
17. Verwey WB. Evidence for a multistage model of practice in a sequential movement task. Journal of Experimental Psychology: Human Perception and Performance. 1999;25:1693–1708.[ISI]
18. Verwey WB. Concatenating familiar movement sequences: the versatile cognitive processor. Acta Psychol. 2001;106:69–95.[Medline]
19. Pew RW. Levels of analysis in motor control. Brain Res. 1974;71:393–400.[ISI][Medline]
20. Keele SW, Jennings P, Jones S, et al. On the modularity of sequence representation. J Mot Behav. 1995;27:17–30.[ISI]
21. Berry DC, Dienes Z. Implicit Learning: Theoretical and Empirical Issues. Hove, United Kingdom: Lawrence Erlbaum; 1993.
22. Reber AS. Implicit learning and tacit knowledge. Journal of Experimental Psychology: General. 1989;3:219–235.
23. Seger CA. Implicit learning. Psychol Bull. 1994;115:163–196.[ISI][Medline]
24. Shanks DR, St John MF. Characteristics of dissociable human learning systems. Behav Brain Sci. 1994;17:367–395.[ISI]
25. Stadler MA, Frensch PA. Handbook of Implicit Learning. 1998 Thousand Oaks, Calif: Sage Publications;.
26. Wulf G, Schmidt RA. Variability of practice and implicit motor learning. Journal of Experimental Psychology: Learning, Memory, and Cognition. 1997;23:987–1006.[ISI]
27. Shea CH, Wulf G, Whitacre CA, Park JH. Surfing the implicit wave. Q J Exp Psychol. 2001;54:841–862.[ISI]
28. Willingham DB, Nissen MJ, Bullemer P. On the development of procedural knowledge. Journal of Experimental Psychology: Learning, Memory, and Cognition. 1989;15:1047–1060.[ISI][Medline]
29. Lewicki P. Nonconscious acquisition of information. Am Psychol. 1992;47:796–801.[Medline]
30. Lewicki P, Czyzewska M, Hoffman H. Unconscious acquisition of complex procedural knowledge. Journal of Experimental Psychology: Learning, Memory, and Cognition. 1987;13:523–530.[ISI]
31. Berry DC, Broadbent DE. Interactive tasks and the implicit-explicit distinction. Br J Psychol. 1988;79:251–272.[ISI]
32. Magill RA. Is conscious awareness of environmental information necessary for skill learning? Paper presented at: 8th European Congress of Sport Psychology; June 22, 1991; Cologne, Germany.
33. Magill RA, Hall KG. Implicit learning in a complex tracking skill. Paper presented at30th Annual Meeting of the Psychonomic Society; November 10, 1989; Atlanta, Ga.
34. Magill RA, Schönfelder-Zohdi B, Hall KG. Further evidence for implicit learning in a complex tracking task. Paper presented at31st Annual Meeting of the Psychonomic Society; November 12, 1990; New Orleans, La.
35. Hoyer WJ, Lincourt AE. Aging and the development of learning. In: Stadler MA, Frensch PA, eds. Handbook of Implicit Learning. Thousand Oaks, Calif:Sage Publications; 1998:445–470.
36. Cherry KE, Stadler MA. Implicit learning of a nonverbal sequence in younger and older adults. Psychol Aging. 1995;10:379–394.[ISI][Medline]
37. Nissen MJ, Willingham DB, Hartman M. Explicit and implicit remembering: when is learning preserved in amnesia? Neuropsychologia. 1989;27:341–352.[ISI][Medline]
38. Howard DV, Howard JH. Age differences in learning serial patterns: direct versus indirect measures. Psychol Aging. 1989;4:357–364.[ISI][Medline]
39. Verwey WB. Buffer loading and chunking in sequential keypressing. Journal of Experimental Psychology: Human Perception and Performance. 1996;22:544–562.[ISI]
40. Klapp ST. Motor response programming during simple and choice reaction time: the role of practice. Journal of Experimental Psychology: Human Perception and Performance. 1995;21:1015–1027.[ISI]
41. Rosenbaum DA, Saltzman E, Kingman A. Choosing between movement sequences. In: Kornblum S, Requin J, eds. Preparatory States and Processes. Hillsdale, NJ: Lawrence Erlbaum; 1984119–134.
42. Schmidt RA. A schema theory of discrete motor skill learning. Psychol Rev. 1975;32:225–252.
43. Winstein CJ, Gardner ER, McNeal OR, et al. Standing balance training: effect on balance and locomotion in hemiparetic adults. Arch Phys Med Rehabil. 1989;70:755–762.[ISI][Medline]
44. Wightman DC, Lintern G. Part-task training for tracking and manual control. Human Factors. 1985;27:267–283.[Medline]

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when Rapid Responses are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Shea, C. H Articles by Wilde Braden, H. Search for Related Content PubMed PubMed Citation Articles by Shea, C. H Articles by Wilde Braden, H.