Impact of explicit information on implicit motor-sequence learning following middle cerebral artery stroke.

Physical therapists spend considerable therapeutic time explicitly delivering instructions to their patients and clients. In an effort to guide the learner to an optimal motor solution, these directions commonly are centered on "how to" complete a movement task. Despite the large amount of time and effort dedicated to instructing individuals during rehabilitation, few studies have considered the impact of explicit information on the learning of implicit motor skills in individuals with, or without, neurologic damage. (1,2) In fact, no research has unequivocally established that verbal explicit instructions aid implicit motor skill learning in any population. This study sought to address this issue by investigating the impact of explicit information (EI) on implicit motor-sequence learning in people with stroke in the middle cerebral artery (MCA) distribution that affected the sensorimotor cortical (SMC) regions.

Explicit Versus Implicit Memory and Learning

Learning and memory are not singular processes, but are composed of many separate abilities. The broad categories of learning and memory can be subdivided into 2 main types--explicit and implicit. (3) Explicit learning may be assessed directly by testing memory for factual knowledge (eg, recognition and recall). In contrast, implicit learning is interred by observing changes in skilled movement relative to some baseline performance. In this case, improved performance is assumed to reflect the acquisition of knowledge about the task, which is then manifested as, for example, faster or more accurate movements.

Thus, the explicit and implicit learning and memory systems differ fundamentally. Explicit knowledge is represented as memory for facts, events, and episodes and may be formed very quickly (even following one exposure to explicit information). It is directly accessible to conscious recollection (3) and is used to guide high-level cognition when decisions are based on complex rules and information.

By contrast, the functions of the implicit system are highly distributed, supporting multiple behaviors, including skills and habits (eg, sequence learning), priming (eg, word completion), associative learning (eg, classic and operant conditioning), and nonassociative learning (eg, habituation). (3) The focus of our study was implicit motor learning that subserves the acquisition of motor skills. The hallmark of implicit motor learning is the capacity to acquire skill through physical practice without conscious recollection of what elements of performance improved. A classic example illustrating this process is learning to ride a bicycle. Improved performance is manifested by fewer falls, yet the ability to explicitly express "what" procedures are being used to avoid tailing is almost impossible. (4)

The Interaction Between Implicit and Explicit Learning

One of the most interesting features that separates the explicit and implicit learning and memory systems is their relative neuroanatomic isolation from one another. Strong evidence for this dissociation comes from the finding that individuals with medial temporal lobe damage demonstrate profound explicit learning deficits, but retain the ability to learn implicit motor skills. (5-7) Fortunately, it is almost impossible to completely disrupt the implicit learning system to the same degree. Because the implicit learning system is highly distributed--supported by the cerebellum, basal ganglia, and SMC areas (8-13)--the literature has shown that no single lesion or disease process completely abolishes the ability to implicitly learn and remember motor skills. Despite their neuroanatomic separation, it appears that explicit and implicit learning sometimes develop in parallel (14) and can profoundly affect one another. (1,15-17) This raises the possibility that one of these memory systems might be used to stimulate or inform the other. However, the mechanism for the interaction between explicit and implicit memory systems during learning is not yet understood.

Recently, it was demonstrated that extended, focused implicit practice could be used to promote the explicit memory function in an individual with profound explicit memory loss. (18-20) Less studied is the impact of prior explicit information on motor-sequence learning in cases of an impaired implicit memory system. There are many examples demonstrating that the cognitive demand of explicit instructions can disrupt the formation of the implicit motor plan. (15-17,21) This is true for rule learning (16) and for implicit motor tasks. (15) In these studies, it appears that EI interfered with participants' implicit motor performance. (22,23) These data suggest that explicit information was less helpful in the development of the motor plan for these tasks than was discovering a motor solution primarily through the implicit system. In contrast to the above referenced work, other data have been reported that demonstrate a benefit of EI for implicit motor skill learning. (14-24) These conflicting findings suggest that the impact of EI on implicit motor learning is dependent on the type, timing, and meaningfulness of the information provided. However, these factors have not been carefully considered.

Implicit Motor Learning After Stroke-Related Brain Damage

Another factor that may critically affect the impact of EI on implicit motor skill learning is lesion location following central nervous system damage. We have previously shown that EI aided implicit motor-sequence performance for individuals with stroke, (1) but we did not consider the locus of brain damage. (25) To date, only 2 studies have examined implicit motor skill learning in individuals with focal brain damage. These studies showed a benefit of EI following cerebellar stroke (26) and an interference effect after basal ganglia stroke. (21) However, no work has considered the impact of EI on implicit motor skill learning following local brain damage resulting from stroke in the SMC areas.

We expected that stroke in the SMC areas might negatively affect implicit motor skill learning. This is because the SMC regions are important for supporting move-merit initiation and fine motor coordination (primary motor cortex [M1] (27,28)), transitioning between movements (premotor cortex [PMC] (29)), and the selection of responses in a sequence when choices are based on internal or prelearned information (supplementary motor areas [SMA] (30)). Additionally, functional neuroimaging investigations commonly demonstrate both unilateral and bilateral activation in the M1, SMA, and PMC during implicit motor skill learning. (8,11,13,30-33)

Stroke presents an excellent model for studying the neural control of learning because it permits investigators to examine the relationships between lesion location and functional deficits. This type of analysis allows the formation of hypotheses about regional brain contributions to behavior. One difficulty inherent in using a stroke model to study motor skill learning centers on the need to separate the effects of impairments in motor execution from motto learning. In this context, asking individuals with MCA stroke to use the more involved contralesional upper extremity for task practice is problematic; differences between people with and without stroke might be inflated by impaired motor execution and not reflect motor learning abilities at all. Further, individuals who did not have sufficient contralesional motor ability would be summarily excluded from all motor learning studies. We chose to circumvent these pitfalls by requiring individuals with MCA stroke to practice our implicit motor learning task using the ipsilesional upper extremity. It is well known that the motor control of both upper extremities is affected by unilateral stroke. (34-36) The bilateral participation of the SMC areas during motor sequencing tasks has been demonstrated in functional neuroimaging (8) and by using transcranial magnetic stimulation that, when applied to the motor cortical areas, causes severe motor control deficits during motor-sequence practice. (37,38) To expand on these data, we chose to determine if the disrupted ipsilesional motor control noted following MCA stroke would be mirrored by deficits in implicit motor learning.

Therefore, the goal of this study was to examine the interaction between EI and implicit motor-sequence learning in individuals with unilateral damage in the SMC regions secondary to stroke in the MCA distribution. Based on our previous work, (1) we hypothesized that EI would benefit sequence learning in people with no known neurologic damage as well as in those people with stroke in the MCA distribution as they practiced an implicit motor-sequencing task.

Method

Participants

Ten people with first-time unilateral stroke in the MCA distribution affecting the SMC areas and 10 age-matched people with no known pathology or impairment (control participants) were recruited. All participants were randomly assigned to either a group that was provided with EI or a group that was not (EI and No-EI groups, respectively). Inclusion criteria for individuals with stroke were: (1) confirmation of unilateral damage in the MCA distribution affecting the SMC areas and (2) current clinical status being at least 6 months post-stroke. To ensure homogeneity between the participants with stroke and the control participants, all participants were right-hand dominant (determined by participant self-report) and did not demonstrate any evidence of dementia (score of at least 26 on the Mini-Mental State Exam). Exclusion criteria for all participants included the following: (1) acute medical problems; (2) uncorrected vision loss: (3) previous history of psychiatric admission; and (4) history of multiple strokes, transient ischemic attacks, or extensive cortical white matter disease. Individuals with stroke were recruited from the outpatient clinical services at the University of Southern California Healthcare Consultation Center, the Rancho Los Amigos National Rehabilitation Center, and the South Bay Stroke Support Group. Control participants were recruited from the local community. To protect the individual rights of participants, each person signed an approved institutional informed consent form and a medical records release form prior to testing. To characterize stroke severity and ensure that no baseline differences existed between groups, the upper-extremity motor portion of the Fugl-Meyer physical assessment was calculated for each participant. There were no differences in age, Mini-Mental State Exam scores, or Fugl-Meyer motor scores between groups. Descriptions of the participants with stroke and the control participants are presented in Table 1.

Lesion Location

Prior to each participant's inclusion in this study, review of an existing magnetic resonance imaging (MRI) or computed tomography (CT) scan confirmed lesion location. These medical records were obtained with the written consent of each participant. Existing MRI or CT scans were used to reconstruct each lesion on 3 representative axial brain slices. (39) Each brain scan was reconstructed using custom-designed software. Representative overlay images demonstrating the extent of each individual's brain lesion are illustrated in Figure 1.

[FIGURE 1 OMITTED]

Instrumentation and Task

All participants practiced the serial reaction time (SRT) task. (6) Four different colored circles (yellow, red, blue, and green) could he displayed on the computer screen (43.2 cm [17 in], color) placed directly in front of the participant. A standard keyboard was placed on the table directly in front of the computer screen with the most centered letters (v, b, n, and m) capped with the colors yellow, red, blue, and green, respectively.

Displaying 1 of the 4 colored circles on the screen generated stimuli for movement. Only 1 colored circle appeared at a time; the other circles were transparent. However, each colored circle always appeared in the same position and thus maintained its relative location on the screen and to the corresponding key. Following the appearance of 1 colored circle and prior to the display of the next colored circle, a large black asterisk, centered on the screen, served as a fixation point. Responses were made by pressing 1 of the 4 keys corresponding (in color and location) to the appropriately colored circle. A custom-designed computer software program (L Boyd, 2000, E-Prime software platform, version Beta 5.0) * controlled the appearance of the colored circles and recorded the participants' responses. Time data (response time [RT]) were stored after every key press for future analysis.

Procedure

Sequence practice. Participants were seated facing the computer screen with their hand resting on the keyboard. Participants with stroke used the hand ipsilateral to brain damage; the hand used by control participants was matched to that used by participants with stroke (Tab. 1). Four fingers of one hand were used to respond (all except the thumb). Following the cue to respond, participants pushed the key corresponding to the colored circle. Participants were instructed to "respond as quickly as possible."

The procedure for implicit repeating sequence practice was identical for both groups. All participants practiced the same fixed and repeating 10-element sequence (blue-yellow-red-blue-green-red-blue-red-green-yellow). This sequence was constructed to be ambiguous, such that there were minimal probability relationships among its elements. The beginning and end of each sequence were not marked, so that the transition between sequences was seamless. Each block of responses was composed of 10 repetitions of the sequence (100 responses). In each session, participants practiced the sequence a total of 50 times (5 blocks of practice). A short break of 1 to 2 minutes was provided at the end of each block of responses.

An initial block of random responses were practiced (100 responses). Next, 4 blocks of repeating-sequence practice and a second block of random responses were performed. Finally, participants practiced 1 last block of the repeating sequence. In stun, participants practiced the repeating sequence for 5 blocks (50 times through the sequence, for a total of 500 responses) and made random responses for 2 blocks (200 responses). This practice pattern (1 random block, 4 sequence blocks, 1 random block, 1 sequence block) was repeated on 3 consecutive days. On day 4, retention tests were given to assess learning of the SRT task. Retention was measured by performance of 1 block of the repeating sequence.

By random designation, half of the participants in each group were provided with explicit information regarding sequence patterns prior to practice, and half of the participants were kept unaware of the sequences being practiced. For those participants in the EI group, day 1 consisted of practice only. On day 2, participants in the EI group were informed that there was a repeating sequence in some of the practice trials. On day 3, participants in the EI group were explicitly instructed regarding the existence and composition of the repeating sequence. They also were provided with a schematic drawing of the repeating sequence that they were allowed to study, but not to physically practice. Days 1, 2, and 3 consisted of practice only for participants in the No-EI group; no explicit information regarding the sequence was provided. The delivery and content of explicit instructions for each group are detailed by day and group in Table 2.

Explicit testing. Three levels of explicit knowledge were tested: subjective awareness of the existence and composition of the sequence, recognition memory, and recall memory. Subjective memories were tested by asking participants if they noticed anything about the task. Recognition memory tests determined if participants would be able to correctly identify the repeated sequence after watching it be played on the screen. Recall was tested to ascertain if participants knew the repeated sequence well enough to correctly predict what element of it would come next when viewing a fragment of the 10 elements (ie, 3 elements). Table 3 provides the instructions and details of explicit tests.

Because of the differences in delivery of explicit information across groups, these tests were given at different time points. The EI group participated in 4 separate explicit test sessions (with only the first testing subjective awareness; Tab. 2). Explicit knowledge was tested for the No-EI group only after the retention test on day 4 (Tab. 2).

Outcome measures. Response time (reaction time + movement time) was the time between stimulus onset to key press and was measured and stored for each trial. As is standard procedure in SRT task data analyses, (1,6,8,14,40,4) we calculated the median RT for each 10-element sequence trial. Calculation of median RT values for each sequence trial reduces the sensitivity of this measure to very large or small values. Because RT data can be highly variable, the use of median RT as an outcome measure reflects a more conservative approach to data management. (42) Response times then were summarized by calculating the mean median for each block of responses. This procedure was performed for both random and repeated sequences. Finally, to allow comparison across participants and to eliminate the effect of grossly different RTs, a change score was calculated for each block of practice (RT change score=mean median RT from the second block of random sequence practice on day 1 minus mean median RT from repeating sequence blocks 1-15).

We chose to use the mean median RT from the second random block in all our change score calculations for 2 reasons: (1) at this time, the amount of practice and EI was equivalent between the EI and No-EI groups, and (2) we wanted to ensure that enough practice had occurred to greatly reduce (if not virtually eliminate) nonspecific learning effects. Early in practice, RT is often very long, but decreases rapidly as participants become familiar with the task. These shorter RTs are often due to learning the relationship between stimulus and response and do not relate to sequence learning per se. To ensure that we did not inflate change scores by the inclusion of these nonspecific learning effects, we represented random response performance by using RT from the second random block (block 5 on day 1). Learning was assessed by calculating the difference between the median RT from the second block of random responses on day 1 and the mean median RT from the retention test. Explicit testing was evaluated by calculating a percentage-correct score for the subjective responses, recognition, and recall (eg, 80%, or 4 of 5 participants, noticed the repeating sequence).

Data Analysis

Acquisition performance was assessed using a 3-factor (group [participants with stroke, control participants] x information (EI, No-EI) x day [1, 2, 3]) analysis of variance (ANOVA), with repeated-measures correction for day and RT change score as the dependent measure. Subsequently, post hoc tests were performed to identify the locus of interactions. These tests separately examined information of the sequence (information x day ANOVA) and between-group differences (group x day ANOVA).

Retention test data were used to reflect implicit motor-sequence learning. These data were examined in 2 ways. First, the mean median RT from the retention test was compared with that from the random responses tot each group and information condition (paired t tests). These results indicated whether implicit learning of the repeating sequence had occurred for each group. Second, the magnitude of RT change during the retention test was evaluated using a 2-factor (group x information) ANOVA. This analysis indicated whether normalized change scores were reliably different across group and information conditions. Post hoc 1-factor (information) ANOVAs were performed to determine the reliability of differences between groups that practiced with and without EI about the sequence.

Results

Acquisition Performance

Regardless of the EI factor, all participants in this study were able to decrease their response times across acquisition, as indicated by their overall improved ability to respond to the practiced sequence relative to a random sequence (Figs. 2A and 2B; see days 1, 2, and 3). Visual inspection of the data, however, demonstrated that, contrary to our hypotheses, the participants with stroke did not benefit from the provision of explicit information. A full-factor ANOVA (with repeated-measures correction) confirmed this observation and revealed a 3-way group X information X day interaction (F=8.17; df=2,32; P=.00). Post hoc analyses revealed that EI differently affected the performance of the 2 groups. When the information factor was evaluated separately for each group (group X day ANOVA), performance was different between the participants with stroke and the control participants in the EI group (F-7.00; df-2,16; P=.05), whereas performance was not different between the participants with stroke and the control participants in the No-EI group (P=.21). When the 2 groups were considered separately (information X day ANOVA), both demonstrated an information X day interaction. However, the direction, of this effect was opposite for the 2 groups. Providing the control participants with EI aided their performance (F=3.97; df=2,16; P=.04; Fig. 2A). In strong contrast to this result, EI impaired the ability of participants with stroke to reduce RTs (F=4.82; df=2,16; P=.02; Fig. 2B). These data are presented separately and in more detail in the following sections.

Control participants. Changes in RT for the repeated sequence across acquisition for the control participants are displayed in Figure 2A. Analysis of the change scores from days 1, 2, and 3 demonstrated improvements; both the EI and No-EI groups were faster in responding to the repeated sequence (main effect of day: F= 15.16; df=2,16; P=.00). As mentioned earlier, the information X day interaction was due to the larger decreases in RT for the control participants in the EI group compared with the control participants in the No-EI group. By the end of practice on day 3, the control participants in the EI group had decreased RT by 196 milliseconds, whereas the control participants in the No-EI group had improved performance (decreased RT) by only 87 milliseconds. This difference indicates a clear benefit of EI during SRT task acquisition for individuals without neurologic damage. Because of our small sample size, we confirmed this finding by calculating an effect size. (43) A large effect size (.72) verified that there was a meaningful difference between the control participants in the EI group and the control participants in the No-EI group at the end of day 3 of practice.

Participants with stroke. The participants with stroke also were able to improve acquisition performance with practice (main effect of day: F=17.86; df=2,16; P=.00; Fig. 2B). The provision of EI, however, had a vastly different impact on the participants with stroke compared with the control participants. When given EI, the participants with stroke in the EI group demonstrated less change in their RTs relative to the participants with stroke in the No-EI group. This was confirmed statistically with an information X day interaction (see above). The detrimental effect of EI oil individuals with stroke in the MCA distribution was most evident at the end of day 3 of practice, when the participants with stroke in the No-EI group demonstrated a 207-millisecond decrease in RT, whereas the participants with stroke in the EI group had decreased RT by only 128 milliseconds. This difference was meaningful, as demonstrated by a large effect size (43) (.97), verifying the substantially better performance of the participants with stroke in the No-EI group than the participants with stroke in the EI group at the end of day 3 of practice.

Retention Performance

To demonstrate that there were no baseline performance differences between the El and No-EI groups, random-sequence performance was compared for both the participants with stroke and the control participants. There was no difference between the participants with stroke and the control participants in the EI group in terms of random-response RT (Fig. 3A). Regardless of EI conditions, all of the individuals in this study demonstrated lower response times for the repeated sequence compared with the random sequence during the retention test (control participants in EI and No-EI groups, P=.01; participants with stroke in EI and No-EI groups, P=.05 and P=.03, respectively; Fig. 3). This finding demonstrates implicit motor-sequence learning of the repeated sequence, for both people with and without stroke affecting the SMC regions. Despite decreased RTs, neither the participants with stroke in the EI group nor those in the No-EI group were as fast in their responses as the control participants (Fig. 3A).

A group X information ANOVA with the RT change score from the retention test as the dependent measure revealed an interaction (F=7.20; df=1,16; P=.02; Fig. 3B). Post hoc tests demonstrated that this interaction was the result of differences between the participants with stroke in the El group and those in the No-EI group (F=5.20; df=1,9; P=-.05). During the retention test, the participants with stroke in the No-EI group had decreased RTs compared with participants with stroke in the EI group (261 versus 138 milliseconds, respectively). Interestingly, there was not an information effect in the control participants during the retention test (P=.20). This finding indicates that, for individuals without neurologic damage, the beneficial effect of explicit information was temporary and principally affected acquisition performance.

Explicit Knowledge

Participants not provided with El. The acquisition of explicit knowledge during implicit motor-sequence practice tot both participants with stroke and control participants in the No-EI group varied greatly (Tab. 4). Most striking was the finding that, at the conclusion of 3 days of practice and the retention test, only 20% (1 of 5) of the participants with SMC-area stroke noticed the presence of some repetition in their responses. In contrast, 80% (4/5) of the control participants in the No-EI group reported being aware of a repeating sequence for some of the trials. Despite this disparity, the control participants and the participants with stroke in the No-EI group were similarly poor in their ability to recognize (66% and 53%, respectively) and recall (40% and 33%, respectively) the practiced sequence. The poor recall performance of both the control participants and the participants with stroke (below 50%) indicates that they were guessing and unable to correctly recall the repeated sequence.

Participants provided with EI. Similar data were found at the conclusion of day 1 of practice when explicit knowledge was first tested in both participants with stroke and control participants provided with EI. Eighty percent of the control participants in the EI group stated that they noticed some degree of repetition in their responses for the SRT task (Tab. 5). Despite this, recognition and recall were below chance (40% each). By the end of day 2 of practice, recognition for the control participants in the EI group had improved to 73%; however, recall remained below chance (40%). Alter the study period, the control participants demonstrated a high degree of explicit knowledge of the SRT task prior to (recognition 86%, recall 73%) and at the end of (recognition 100%, recall 86%) practice on day 3.

The participants with stroke in the EI group demonstrated poor knowledge of the sequence at the end of day 1 of practice (20%; Tab. 5). Recognition was at a chance level (46%), and recall was below chance (27%). Over day 2 of practice, when the participants with stroke explicitly knew that a sequence existed, they showed improved recognition (53%) and recall (47%). The pre-implicit practice, explicit study period at the beginning of day 3, substantially improved sequence recognition (80%), but not recall (47%). Following sequence practice on day 3, recognition was well above chance (73%), and recall was just at chance level (53%).

Discussion

There is little agreement in the literature concerning the impact of EI on implicit motor-sequence learning. Further, few studies have addressed the question of the effect of EI on implicit motor-sequence learning with regard to the neural substrates that may mediate this information transfer. This was one question addressed by this study: How might EI affect implicit motor-sequence learning following focal stroke in the neural regions that likely subserve learning and may mediate information transfer between the explicit and implicit memory systems?

We hypothesized that EI would benefit both groups' implicit motorsequence learning. We learned, however, that this was not true for individuals with stroke affecting the SMC areas. Participants with stroke in the EI group were unable to make use of EI during acquisition performance. Indeed, they demonstrated poorer learning of the SRT task than did participants with stroke in the No-EI group. This pattern suggests an interference effect of EI on implicit motorsequence learning, and that explicit information might have blocked formation of the implicit motor plan.

Regions within the SMC have long been considered critical for most motor output (including the M1, PMC, and SMA). It has been demonstrated that ipsilateral M1 is active during the execution of complex, repetitive finger movements. (37,38,44) Further, neuroimaging data have shown that once EI is gained for implicit tasks, bilateral PMCs are active even for unimanual tasks. (30,45) This finding suggests that the PMC has a strong role in regulating sequence production when learners have access to EI. Interestingly, the PMC has strong connections with the prefrontal regions associated with explicit memories (ie, the dorsolateral prefrontal cortex) and is richly and reciprocally interconnected with the caudate nucleus of the basal ganglia. (46-48) It is quite likely that damage to, and in regions associated with, the PMC results in disrupted integration of EI into planned sequences of movement. Similarly, we also have reported that following basal ganglia damage, EI has a large interference effect on implicit motor-sequence learning. (21) Perhaps under normal circumstances, both the basal ganglia and the PMC are highly active in the integration of El into the representations of movement (eg, motor plans) that are being learned. It may be that disruption of either of these regions leads to diminished ability, to take advantage of EI during implicit motorsequence practice.

Explicit information affected acquisition performance differently for both groups of participants. Explicit information clearly benefited the control participants in the EI group, as demonstrated by the larger decreases in RT for these participants compared with the control participants in the No-EI group and participants with stroke in both the EI and No-EI groups. The beneficial impact of El on the acquisition performance of individuals without neurologic damage corroborates previous findings by Curran and Keele. (24) However, Curran and Keele's findings contradict work by Reber and Squire, (7) who did not find a benefit of prior EI for a group who memorized the sequence before practice.

There are 2 logical explanations for this disparity. First, no pretest was administered in the study by Rebec and Squire, (7) and it cannot be assumed that the participants had explicitly learned the sequence. We controlled for this factor by using a pretest to demonstrate that both the participants with stroke and the control participants in the EI group showed a high degree of recognition memory for the repeated sequence prior to practice at the beginning of day 3 of practice (Tabs. 2 and 5). A second difference centered on the timing of the provision of EI. In our study, participants in the EI group had an initial 5 blocks of sequence practice without any EI; another 5 blocks were practiced with partial EI (of the existence of a sequence). Finally, at the beginning of day 3 of practice, full EI was provided. Thus, all participants in the El group had considerable amounts of implicit SRT task practice before they were encouraged to incorporate EI into their performance. It has been hypothesized that explicit and implicit memories develop in parallel, but that awareness of EI does not occur until a certain degree of motor success has been achieved. (14,49) If this is true, then perhaps it is not just awareness of EI that occurs following some motor success; more importantly, EI may only be incorporated into, and benefit, the motor plan after some degree of implicit task ability has been gained.

It has been demonstrated that when sequences are practiced unilaterally under purely implicit conditions, SMC regions are more active contralaterally. (30,50) Therefore, it has been assumed that the ipsilateral SMC function is not directly impaired by stroke. In our study, the participants with stroke in both the EI and No-EI groups initially showed very long RTs, which reflects disrupted motor output. This detail is interesting, as the individuals with stroke were practicing using the arm ipsilateral to brain damage, and thus their undamaged cortical hemisphere and less-involved upper extremity. As the SRT task was unimanual, this finding suggests that bilateral hemispheric activity is used for executing and learning motor sequence plans.

Indeed, other work has shown that ipsilesional deficits alter stroke affect rehabilitation outcomes. (51-53) Our data demonstrate that, in addition to impairing motor control, (34,54) MCA stroke also negatively affects the ability to use EI during implicit motor task practice, even when using the ipsilesional upper extremity. As use of the ipsilesional arm invoked the undamaged hemisphere, the performance deficits we recorded strongly suggest that bilateral hemispheric function is necessary during implicit motor-sequence learning. Our findings, along with the findings of other researchers, (35,54) imply that, in the clinical arena, therapeutic interventions also must be directed to the less-involved upper extremity following MCA stroke.

The findings of this study might inform clinical practice in several ways. We documented that, regardless of information condition at the time of the retention test, the performance of the participants with stroke was never as fast as that of the control participants. Thus, this study adds to the growing body of literature demonstrating that the ipsilesional upper extremity is not completely spared from the effects of MCA stroke. (34-36,51-55) Despite this finding, it is critical to remember that all of the individuals with MCA stroke in our study demonstrated the capacity for implicit motor-sequence learning. This is an important fact, and supports other recent work by Pohl et al. (41) It appears that, with sufficient practice, improved acquisition performance can persist (retention performance), and the motor task can reliably be considered learned.

The predominant pattern of stroke is in the distribution of the MCA (56) and affects the SMC areas and the basal ganglia. Currently in clinical practice, it is common for little attention to be paid to the locus of brain damage (other than identifying which side of the brain is affected). It is clear that individuals ill this study who had stroke affecting the SMC regions did not benefit from El during SRT learning; more accurately, EI interfered with learning in this group. It is very difficult to extend laboratory data directly into clinical practice; however, we believe that consideration of lesion location, and increased awareness of the type and timing of EI about the task, might benefit implicit motor learning following stroke. However, several factors limit the generalizability of our findings. We are not able to form any conclusions regarding the impact of lesion in the right versus left hemisphere on implicit motor skill learning. Additionally, because of the unique nature of the implicit learning and memory systems, it may be problematic to extend these findings to all of motor learning. Last, due to our small sample size, it may be premature to alter clinical practice solely on the basis of these data. Therefore, it may be prudent to wait for future data to confirm or refute our findings before making dramatic changes in rehabilitation practice.

Conclusion

Consideration of the locus of brain damage in conjunction with the specific behavioral function associated with the region of stroke during formation of the rehabilitation plan may be one means of optimizing recovery of function poststroke. Bilateral SMC areas are likely important for accurate formation of the motor plan for movement. For the participants in this study, it appears that, following SMC stroke, EI did not benefit SRT task learning and in fact it degraded both performance and learning. We believe that other alternative methods of prescriptive instructions (other than verbal explicit information) might have been more beneficial for implicit learning of motor skills in this sample. This idea is not new (57); however, its implementation into the clinical environment would require substantial reconceptualization. For example, it is possible that explicit instructions should be used to focus the learner's attention rather than provide information about the task. (58) Making such a paradigm shift in the rehabilitation setting may prove initially difficult for therapist and client alike, yet it may yield far more beneficial long-term results.

Table 1.
Participant Characteristics

                                              Age (y)
                          Lesion
                          Side      Sex       [bar]X  SD

Participants with stroke
  EI (c)                  3 right,  2 male,
                          2 left    3 female  59.0    10.5
  No EI                   1 right,  4 male,
                          4 left    1 female  58.6    19.2

                                              Age (y)
                          Hand
                          Used      Sex       [bar]X  SD

Control participants
  EI                      3 right,  1 male,
                          2 left    4 female  55.4    11.0
  No EI                   3 right,  2 male,
                          2 left    3 female  57.4    16.1

                                                     Upper-
                                                     Extremity
                          Poststroke                 Fugl-Meyer
                          Duration                   Motor
                          (mo)          MMSE (a)     Score (b)

                          [bar]X  SD    [bar]X  SD   [bar]X  SD

Participants with stroke
  EI (c)                  33.4    18.9  29.0    1.2  30.2    21.2
  No EI                   48.0    30.1  27.8    1.8  26.8    18.7

                                                     Upper-
                                                     Extremity
                          Poststroke                 Fugl-Meyer
                          Duration                   Motor
                          (mo)          MMSE (a)     Score (b)

                          [bar]X  SD    [bar]X  SD   [bar]X  SD

Control participants
  EI                                    29.8    0.4
  No EI                                 29.6    0.5

(a) MMSE = Mini-Mental State Exam.

(b) Maximum upper-extremity Fugl-Meyer motor score = 66.

(c) EI = explicit information.

Table 2.
Explicit Information and Knowledge Testing Conditions by Group and Day

Group                            Day 1              Day 2

EI (a)  Information condition    "Respond as fast   Partial explicit
                                   as possible"       awareness:
                                                      "Sequence
                                                      exists"
        Explicit knowledge test  Subjective,        Recognition and
                                   recognition and    recall
                                   recall
No-EI   Information condition    "Respond as fast   "Respond as fast
                                   as possible"       as possible"
        Explicit knowledge test  None               None

                                 Day 3              Day 3
Group                            Prepractice        Postpractice

EI (a)  Information condition    Full explicit
                                   information:
                                   verbal instruc-
                                   tions, study
                                   session, and
                                   pretest
        Explicit knowledge test  Prepractice test   Postpractice test
                                   of recognition     of recognition
                                   and recall         and recall
No-EI   Information condition    "Respond as fast
                                   as possible"
        Explicit knowledge test  None               None

                                 Day 4
Group                            Retention Test

EI (a)  Information condition    "Respond as fast
                                   as possible"
        Explicit knowledge test  None
No-EI   Information condition    "Respond as fast
                                   as possible"
        Explicit knowledge test  Subjective,
                                   recognition
                                   and recall"

(a) EI = explicit information.

Table 3.
Explicit Knowledge Test Conditions and Questions

Explicit
Test
Condition    Explicit Test Questions

Subjective   "Did you notice anything      If yes, "What was it?"
  awareness    about the task?"
                                           If no, "There was a
                                             repeating
                                             sequence. Can you
                                             tell me what it
                                             was?"
Recognition  Watch 3 sequences:            "Is this a sequence
  memory       1 True                        that you
               2 False                       recognize?"
Recall       Display 3 separate 4-element  "What color comes
  memory       sequence fragments (ie,       next?" (forced
               yellow-red-blue-?)            choice)

Table 4.
Explicit Knowledge of Participants Not Provided With Explicit
Information (a)

                        Subjective   Recognition   Recall
                    N   % Noticed    % Correct     % Correct

Control
  participants      5   80 (4/5)     66.0          40.0
Participants with
  stroke            5   20 (1/5)     53.0          33.0

(a) Scores below 50% indicate responding at or below chance (ie,
guessing).

Table 5.
Explicit Knowledge of the Participants Provided With Explicit
Information (a)

                                Subjective %   Recognition %   Recall %
                            N   Noticed        Correct         Correct

Day 1
  Control participants      5   80 (4/5)        40.0           40.0
  Participants with stroke  5   20 (1/5)        46.6           26.6
Day 2
  Control participants      5                   73.0           40.0
  Participants with stroke  5                   53.3           46.6
Day 3 pretest
  Control participants      5                   86.6           73.0
  Participants with stroke  5                   80.0           46.6
Day 3 posttest
  Control participants      5                  100.0           86.0
  Participants with stroke  5                   73.3           53.3

(a) Scores below 50% indicate responding at or below chance (ie,
guessing). Explicit knowledge tests of participants provided with
explicit information were performed at the end of each day of practice.
In addition, a pretest of explicit knowledge was administered at the
beginning of day 3. Subjective awareness of the sequence was assessed
only at the end of day 1.

* Psychology Software Tools Inc, 2050 Ardmore Blvd, Suite 200, Pittsburgh, PA 15221.

References

(1) Boyd LA, Winstein CJ. Implicit motor-sequence learning in humans following unilateral stroke: the impact of practice and explicit knowledge. Neurosci Lett. 2001;298:65-69.

(2) Knowlton BJ, Mangels JA, Squire LR. A neostriatal habit learning system in humans. Science. 1996;273:1399-1402.

(3) Squire LR. Memory and Brain. New York, NY: Oxford University Press; 1987.

(4) Polanyi M. Personal Knowledge: Towards a Post-Critical Philosophy. London, England: Routledge; 1958.

(5) Milner B. Amnesia following operation on the temporal lobes. In: Whitty CM, Zagwell OL, eds. Amnesia. London, England: Butterworth; 1966:103-133.

(6) Nissan MJ, Bullemer P. Attentional requirements of learning: evidence from performance measures. Cognit Psychol. 1987;19:1-32.

(7) Reber PJ, Squire LR. Encapsulation of implicit and explicit memory in sequence learning. J Cognit Neurosci. 1998;10:248-263.

(8) Grafton ST, Hazeltine E, Ivry RB. Motor sequence learning with the nondominant left band: a PET functional imaging study. Exp Brain Res. 2002; 146:369-378.

(9) Doyon J. Penhune V, Ungerleider LG. Distinct contribution of the cortico-striatal and cortico-cerebellar systems to motor skill learning. NeuropSychologia. 2003; 41:252-262.

(10) Grafton ST, Hazeltine E, Ivry R. Functional mapping of sequence learning in normal humans. J Cognit Neurosci. 1995;7:497-510.

(11) Honda M, Deiber M-P, Ibanez V, et al. Dynamic cortical involvement in implicit and explicit motor sequence learning: a PET study. Brain. 1998:121:2159-2173.

(12) Seitz RJ, Roland PE. Learning of sequential finger movements in man: a combined kinematic and positron emission tomography (PET) study. Eur J Neurosci. 1992;4:154-165.

(13) Hazeltine E, Grafton ST, Ivry R. Attention and stimulus characteristics determine the locus of motor-sequence encoding: a PET study. Brain. 1997;120:123-140.

(14) Willingham DB, Goedert-Eschmann K. The relationship between implicit and explicit learning: evidence for parallel development. Psychol Sci. 1999;6:531-534.

(15) Green TD, Flowers JH. Implicit versus explicit learning processes in a probabilistic, continuous fine-motor catching task. J Motor Behav. 1991;23:293-300.

(16) Reber AS. Implicit learning of synthetic languages: the role of instructional set. J Exp Psychol Hum Learn Mem. 1976;2:88-94.

(17) Wulf G. Weigelt C. Instructions about physical principles in learning a complex motor skill: to tell or not to tell. Res Q Exert Spree. 1997;68:362-367.

(18) Glisky EL, Schacter D. Acquisition of domain-specific knowledge in organic amnesia: training for computer-related work. Neuropsychologia. 1987;25:893-906.

(19) Glisky EL, Schacter DL. Long-term retention of computer learning by patients with memory disorders. Neuropsychologia. 1988;26:173-178.

(20) Glisky EL, Schacter DL. Extending the limits of complex learning in organic amnesia: computer training in a vocational domain. Neuropsychologia. 1989;27:107-120.

(21) Boyd LA, Winstein CJ. Implicit learning of a complex tracking task: the effect of explicit knowledge and focal stroke in the basal ganglia or cerebellum. Soc Neurosci Abstr. 2011;27:638.10.

(22) Bliss CB. Investigations in reaction time and attention. Studies from the Yale Psvchology Laboratory. 1892;1:1-55.

(23) Boder DP. The influence of concomitant activity and fatigue upon certain forms of reciprocal hand movements and its fundamental components. Comparative Psychology Monographs. 1935;11.

(24) Curran T, Keele SW. Attentional and nonattentional forms of sequence learning. J Exp Psychol Learn Mem Cognit. 1993;19:189-202.

(25) Hazeltine E. Ipsilateral sensorimotor regions and motor sequence learning. Trends Cognit Sci. 2001;5:281-282.

(26) Molinari M, Leggio MG, Solida A, et al. Cerebellum and procedural learning: evidence from focal cerebellar lesions. Brain. 1997;120: 1753-1762.

(27) Berridge KC, Whishaw IQ. Cortex, striatum, and cerebellum: control of serial order in a grooming sequence. Exp Brain Res. 1992;90: 275-290.

(28) Barone P, Joseph JP. Prefrontal cortex and spatial sequencing in macaque monkey. Exp Brain Res. 1989;78:447-464.

(29) Mushiake H, Inase M, Tanji J. Neuronal activity in the primate premotor, supplementary, and precentral motor cortex during visually guided and internally determined sequential movements. J Neurophysiol. 1991;66:705-718.

(30) Toni I, Krams M. Turner R, Passingham RE. The time course of changes during motor sequence learning: a whole-brain fMRI study. Neuroimage. 1998;8:50-61.

(31) Grafton ST, Hazeltine E. Ivrt R. Abstract and effector-specific representations of motor Sequences identified with PET. J Neurosci. 1998;18:9420-9428.

(32) Harrington DL, Ran SC, Haaland KY, et al. Specialized neural systems underlying representation of sequential movements. J Cognit Neurosci. 2000;12:56-77.

(33) Doyon J, Owen AM. Petrides M. et al. Functional anatomy of visuomotor skill learning in human subjects examined with positron emission tomography. Eur J Neurosci. 1996;8:637-648.

(34) Winstein CJ, Pohl PS. Effects of unilateral brain damage on the control of goal-directed hand movements. Exp Brain Res. 1995;105: 163-174.

(35) Winstein CJ, Merians AS, Sullivan KJ. Motor learning after unilateral brain damage. Neuropsychologia. 1999:37:975-987.

(36) Pohl PS, Winstein CJ, Onla-or S. Sensory-motor control in the ipsilesional upper extremity after stroke. NeuroRehabilitation. 1997;9: 57-69. [Corrigendum, NeuroRehabilitation, 1997;9:245-249]

(37) Chen R, Gerloff C, Hallett M, Cohen LG. Involvement of the ipsilateral motor cortex in finger movements of different complexities. Ann Neurol. 1997;41:247-254.

(38) Chen R, Cohen LG, Hallett M. Role of the ipsilateral motor cortex in voluntary movement. Can J Neurol Sci. 1997:24:284-291.

(39) DeArmond SJ, Fusco MM, Dewey MM. Structure of the Human Brain: A Photographic Atlas. 3rd ed. New York, NY: Oxford University Press: 1989.

(40) Willingham DB. Nissan MJ, Bullemer P. On the development of procedural knowledge. J Exp psychol Learn Meta Cognit. 1989;15: 1047-1060.

(41) Pohl PS, McDowd JM, Filion DL, et al. Implicit learning of a perceptual-motor skill after stroke. Phys Ther. 2001;81:1780-1789.

(42) Rosen B. Fundamentals of Biostatics. Pacific Grove, Calif: Duxbury Thomson Learning; 2000.

(43) Thomas JR, Salazar W. Landers DM. What is missing in p <.05? Effect size. Res Q Exerc Sport. 1991;62:344-348.

(44) Shibasaki H, Sadato N. Lyshkow H, et al. Both primary motor cortex and supplementary, motor area play an important role in complex finger movement. Brain. 1993;116:1387-1398.

(45) Schlaug G, Knorr U, Seitz RJ. Inter-subject variability of cerebral activations in acquiring a motor skill: a study with positron emission tomography. Exp Brain Res. 1994;98:523-534.

(46) Selemon LD, Goldman-Rakic PS. Longitudinal topography and interdigitation of corticostriatal projections in the rhesus monkey. J Neurosci. 1985;5:776-794.

(47) Cavada C, Goldman-Rakic PS, Topographic segregation of corticostriatal projections from posterior parietal subdivisions in the macaque monkey. Neuroscience. 1991;42:683-696.

(48) Cavada C, Goldman-Rakic PS. Posterior parietal cortex in rhesus monkey, I: parcellation of areas based on distinctive limbic and sensory corticocortical connections. J Comp Neurol 1989;287:393-421.

(49) Brooks DJ. The role of the basal ganglia in motor control: contributions from PET. J Neurol Sci. 1995;128:1-13.

(50) Boecker H, Dagher A, Ceballos-Baumann O, et al. Role of the human rostral supplementary motor area and the basal ganglia in motor sequence control: investigations with [H.sup.2] [sup.15.O] PET.J Neurophysiol. 1998;79:1070-1080.

(51) Wade DT, Wood VA, Hewer RL. Recovery after stroke: the first three months. J Neurol Neurosurg Psychiatry. 1985;48:7-13.

(52) McDowell F, Louis S. Improvement in motor performance in paretic and paralyzed extremities following nonembolic cerebral infarction. Stroke 1971;2:395-399.

(53) Olsen TS. Improvement of function and motor impairment after stroke. J Neurol Rehabil. 1989;3:187-192.

(54) Pohl PS, Winstein CJ. Practice effects on the less-effected upper extremity after stroke. Arch Phys Med Rehabil. 1999;80:668-675.

(55) Pohl PS. Luchies CW, Stoker-Yates J, Duncan PW. Upper extremity control in adults post stroke with mild residual impairment. Neurorehabil Neural Repair. 2000;14:33-41.

(56) Brust, JCM. Circulation in the brain. In: Kandel ER, Schwartz JH, Jessell TM, eds. Principles of Neural Science. New York, NY: McGraw-Hill; 2000:1306-1316.

(57) Gentile AM. Implicit and explicit processes during acquisition of functional skills. Scand J Occup Ther. 1998;5:7-16,

(58) Wulf G, Hob M. Prinz W. Instructions for motor learning: differential effects of internal versus external focus of attention. J Motor Behav. 1998;30:169-179.

LA Boyd, PT, PhD, is Assistant Professor, Department of Physical Therapy and Rehabilitation Sciences, University of Kansas Medical Center, 3056 Robinson, Mail Stop 2002, 3901 Rainbow Blvd, Kansas City, KS 66160-7601 ([e-mail]). Address all correspondence to Dr Boyd.

CJ Winstein, PT, PhD, FAPTA, is Associate Professor, Department of Biokinesiology and Physical Therapy and Department of Neurology, Keck School of Medicine, University of Southern California, Los Angeles, Calif.

Both authors provided concept/idea/research design, writing, data analysis, and consultation (including review of manuscript before submission). Dr Boyd provided data collection, project management, fund procurement, and participants. Dr Winstein provided facilities/equipment and institutional liaisons.

This research was approved by the institutional review boards of University of Southern California and Rancho Los Amigos National Rehabilitation Center.

This study was supported by funding from the Neurology Section of the American Physical Therapy Association (Patricia Leahy Scholarship) and the Foundation for Physical Therapy (PODS Levels I and II) awarded to Dr Boyd.

This article was received October 18, 2002, and was accepted June 25, 2007.