SEARCH

SEARCH BY CITATION

Keywords:

  • levodopa;
  • learning;
  • Parkinson's disease

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Authors Roles
  8. Financial Disclosures
  9. Michael Lasarev: none.
  10. References
  11. Supporting Information

Studies in animals and in people with Parkinson's disease (PD) demonstrate complex effects of dopamine on learning motor tasks; its effect on retention of motor learning has received little attention. Recent animal studies demonstrate that practicing a task in the off state, when initially learned in the on state, leads to progressive deterioration in performance. We measured the acquisition and retention of 3 different motor tasks in the presence and absence of levodopa. Twenty individuals with Hoehn and Yahr Stage 1.5 to 3 PD practiced the tasks daily for two 4-day weeks, one half practicing on l-dopa the first week and off the second week. The other half practiced off l-dopa both weeks. The tasks were (1) alternate tapping of 2 keys, (2) moving the body toward 2 targets on a posturography device, and (3) mirror drawing of a star. For the tapping and body movement tests, those who practiced on the first week had a progressive decline in performance with practice during week 2, while subjects off during week 1 maintained or improved. In contrast, for the mirror task, subjects on l-dopa initially had much more difficulty completing the task compared to subjects who practiced off. Both groups improved with practice the first week and had flat performance the second week. These data suggest that performance of speed-accuracy tasks learned in the on state may progressively worsen if subsequently practiced in the off state. In addition, performance, but not learning, of some tasks may be impeded by l-dopa. © 2013 International Parkinson and Movement Disorder Society.

Levodopa and, presumably, elevations of brain dopamine from l-dopa, have complex effects on learning new motor tasks in animals with dopamine deficiency and in humans with Parkinson's disease (PD).

Many studies demonstrate that implicit motor sequence learning becomes impaired as PD progresses,[1, 2] and that in early PD patients implicit learning consolidation is impaired.[3, 4] The effects of l-dopa on motor learning are mixed, in part due to the heterogeneity of motor learning tasks studied and to severity of subjects' PD. l-Dopa had little effect on implicit motor learning on serial reaction time[5, 6] and probabilistic learning.[7] l-Dopa impaired explicit motor sequence learning in early PD[8] but restored learning on a different explicit motor learning sequencing task.[9] l-Dopa has also had a negative impact on feedback-based learning requiring error-correction[10] and on tasks requiring “reversal learning,” or unlearning of previously learned responses.[11-14] l-Dopa also has differential effects on feedback-based learning, with impaired learning seen with l-dopa when negative reinforcement is used.[15]

The bulk of studies on learning in humans have evaluated the effect of l-dopa over 1 day of testing, thereby focusing on the acquisition phase of learning. Little is known regarding how practice in the off state affects performance of new tasks initially learned in the on state. Parkinsonian rodents have demonstrated retention of motor learning acquired in the on state that subsequently decayed with practice in the off state.[16] Repeated practice of a task learned on but subsequently practiced off has not been examined in humans.

In this study we examined motor learning and retention over the course of two 4-day weeks, to determine whether tasks learned on l-dopa may be progressively reversed by subsequent practice off l-dopa. In addition, we asked whether performance of a new task is improved or impaired by l-dopa, in 2 types of tasks: a hand and a postural movement task requiring speed-accuracy tradeoffs for achieving the target and a visual-motor transformation task requiring inhibition or reversal of previously learned behavior.

Patients and Methods

  1. Top of page
  2. ABSTRACT
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Authors Roles
  8. Financial Disclosures
  9. Michael Lasarev: none.
  10. References
  11. Supporting Information

Participants

Twenty subjects were recruited from patients associated with a university Movement Disorders program (see Table 1 for subject demographics). Inclusion criteria included mild to moderate idiopathic PD consistent with established guidelines.[17] All participants had been treated with l-dopa for at least 1 year, reported a good response to the medication, and had been on stable doses for at least 1 month. Disease duration ranged from 2 to 16 years, averaging 6.8 years. All subjects were taking l-dopa at least 3 times a day except 1 who was on twice-daily dosing. l-Dopa response was established by history and confirmed via chart review. Exclusion criteria included cognitive impairment, psychosis, or other musculoskeletal or systemic illness that could affect participation in the study. All gave informed consent to a protocol approved by the OHSU Institutional Review Board.

Table 1. Subject demographics
 Group 1 (on week 1, off week 2)Group 2 (off both weeks)t Testa
  1. a

    Two-tail t test, heteroskedastic errors.

  2. b

    One-tail t test, homoskedastic errors.

  3. on, on medication; off, off medication; SD, standard deviation; LED, levodopa equivalent dose; MOCA, Montreal Cognitive Assessment; UPDRS, Unified Parkinson's Disease Rating Scale.

Gender, n (M, F)10 (7 M, 3 F)10 (9 M, 1 F) 
Age, y, mean (SD)67.2 (5.3)65.5 (8.3)0.47
Hoehn & Yahr, mean (SD)2.2 (.4)2 (0)0.17
LED, mean (SD)871 (441)826 (270)0.80
MOCA, mean (SD)27 (2.3)28 (1.4)0.54
UPDRS, mean (SD)Week 1: 24 (9.8)Week 1: 30 (8.7)0.08
 Week 2: 31 (10)Week 2: 31 (7.6)0.42
t Test week 1 versus 2b<0.010.175809 

Tasks

Finger Tapping (Upper Extremity Speed-Accuracy Task)

Subjects alternately tapped between 2 manual counters for 1 minute, repeated for 5 sets, with 1-minute breaks between sets. They were asked to tap as quickly and accurately as possible. Outcome measured was number of taps per minute. The finger tapping task improves with practice[18] and has been used in prior studies of individuals with PD as a measure of bradykinesia and l-dopa response.[19]

Center of Mass Movement Velocity (Whole-Body Speed-Accuracy Task)

Subjects moved their body center of mass (CoM) to targets indicated on the Neurocom Equitest posturography device. This system is comprised of 2 force plates connected to a computer with a monitor visible to the subject. The force plate recorded shifts in the center of pressure under the feet as the subject leaned toward 1 of 2 predetermined, square targets in space, using a computer generated stick figure image as visual guidance. The 2 targets required the subject to alternately lean forward to the left and forward to the right for 15 repetitions daily. The outcome was movement velocity, or average speed of center of gravity movement toward the target. Other studies in individuals with PD have demonstrated that practice of this task resulted in improved movement velocity.[20] Movement velocity results are presented as summed scores for speed of movement (degrees per second) toward each of the 2 targets for each trial, based on previous studies demonstrating no significant difference in laterality of performance on this task in individuals with PD.[20]

Mirror Drawing (Visual-Motor Transformation Task)

Subjects used a computer mouse to draw a star-shaped figure on a computer screen. The star figure was comprised of a series of dots on the screen; each dot appeared in fixed sequence after the subject clicked on a dot with the mouse to make the next dot appear. The mouse was programmed “backward,” requiring the subject to move his or her hand in the opposite direction of the cursor on the screen to successfully complete the figure. Thus, the subject had to inhibit the normal manner of drawing a line to each succeeding dot and institute the opposite movement in response to the visual cue. The outcome measured was the average time, in seconds, to complete the task. The task was repeated 9 times each day. PD patients have demonstrated motor learning on mirror-reversed vision tasks in some previous studies,[21, 22] but not in another when participants practiced for 1 or 3 days.[23]

Protocol

Subjects were randomized into 2 groups: (1) on-off: learn the tasks for a week in the on l-dopa state after taking their usual morning dose of medication 1 hour before testing; or (2) off-off: learn the tasks for a week in the practical off l-dopa state, defined as taking their last l-dopa dose before midnight the night before testing. Randomization was assigned via irregular group assignments in a pattern established at the beginning of the study; eg, the first 3 subjects were in the on-off group, the second 2 subjects in the off-off group, etc. Both groups practiced the motor tasks for a second week in the off l-dopa state. Thus, subjects in the off-off group practiced off l-dopa both weeks. The subjects participated in 2 sequential weeks of testing consisting of 4 days each week. Most subjects had a 3-day break over the weekend between week 1 and week 2, but there were exceptions due to scheduling issues and holidays. One subject in the on-off group had a 1-day break, 1 subject in the on-off group and 2 subjects in the off-off group had a 2-day break, and 2 subjects in each group had a 4-day break. Most subject visits occurred at 8 am in the morning. Subjects resumed their usual medication regimen following testing each morning. A Unified Parkinson's Disease Rating Scale (UPDRS) motor score was taken each day of testing, and a Montreal Cognitive Assessment (MOCA) was obtained the final day of testing.

Analysis

On any given testing day, each measured response consisted of multiple trials on the tasks: 9 trials of mirror drawing, 5 trials of finger tapping, or 15 trials on the CoM movement velocity task. The analysis proceeded by first computing the median response among the trials of 1 task and then analyzing these medians for changes over time, changes between groups, and whether the changes over time differed between groups. Generalized estimating equations[24, 25] (GEEs) were used for all analyses to account for the correlated nature of the measurements across time. All models used an autoregressive lag-1 (AR1) correlation structure and employed a robust sandwich estimator of the variance for all hypothesis tests and confidence intervals. Different probability distributions were used depending upon the type of response: finger tapping (counts per minute) was modeled as Poisson; body movement (CoM velocity) was modeled as normal; and mirror drawing (a latency time) was modeled according to a gamma distribution. Secondary questions regarding retention of learning from week 1 to week 2 were considered by applying the above methods to responses from only the last day's testing (of week 1) and the first day of testing during week 2. Statistical significance was held to 0.05 and all tests were 2-sided.

Results

  1. Top of page
  2. ABSTRACT
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Authors Roles
  8. Financial Disclosures
  9. Michael Lasarev: none.
  10. References
  11. Supporting Information

Twenty-one subjects passed telephone screening and were enrolled in the study. One subject withdrew during the first week due to scheduling conflicts, leaving a total of 10 subjects per group. The 2 groups were comparable with regard to age, average l-dopa dose, Hoehn & Yahr stage, MOCA score, and UPDRS off medication (Table 1). Subjects in the on-off group had a significantly lower average UPDRS score for week 1 when they were tested on l-dopa compared to week 2, off l-dopa, suggesting a clinically significant response to l-dopa. Subjects in both groups had minimal dyskinesia, reflecting relatively mild to moderate PD in both groups.

Finger Tapping

Week 1

The 2 groups had the same average tapping speed on day 1. Both the on-off group and the off-off group improved their performance on the finger tapping task during week 1, with no significant difference (P = 0.96) between the 2 groups at any point during the 4 testing sessions (Fig. 1; Supporting Table 1). The rate of improvement did not differ between the 2 groups (P = 0.23), with both groups experiencing a 3.9% (95% CI, 2.7%-5.0%; P < 0.001) increase in the mean number of taps per minute per day.

image

Figure 1. Estimated mean response derived from fitted models relating taps per minute (A), CoM velocity (B), and star drawing latency (C) to testing session (days 1–4) during each week for the 2 groups of interest (on-off, black dotted line, solid black triangle; off-off, gray solid line, solid gray square). Plotting symbol denotes mean response; 95% confidence interval for mean is shown by error bars. Note reverse direction of vertical axis in (C). Slight horizontal offset has been included to minimize overstriking of points.

Download figure to PowerPoint

Week 2

Both groups started week 2 with an average of 135 (95% CI, 122–150) taps per minute during their first session, a value comparable to the performance of both groups at the conclusion of the first week's testing. However, the performance of the on-off group declined as they practiced off l-dopa during week 2, with a decrease in tapping speed of 2% per day (95% CI, 0.1%-3.3%; P = 0.039). In contrast, the off-off group continued to show an increase of 4.4% per day (95% CI, 2.4%-6.3%; P < 0.001) in mean number of taps per minute during the first 2 days of week 2, then leveled off during days 3 and 4 (Fig. 1).

CoM Movement Velocity

Week 1

The 2 groups started week 1 with no difference in movement velocity (P = 0.221). During week 1, both on-off and off-off groups demonstrated a significant increase in movement velocity that did not differ between groups (P = 0.60) (Fig. 1; Supporting Table 2). For both groups, each additional day of testing was associated with a 0.35-degree per second increase in the mean combined response (95% CI, 0.11-0.60 degrees/s increase; P = 0.004). The 2 groups did not differ significantly on any given testing day (P = 0.221), averaging only 1.35 degrees per second lower in the off group than the on group.

Week 2

On the first day of the second week, when both groups practiced without l-dopa overnight, average performance was comparable between groups (P = 0.82). Over the following 3 days, the 2 groups demonstrated a pattern of performance similar to that observed on the finger tapping task. The on-off group, who trained when on l-dopa, had a significant worsening of performance, with each day of practice associated with a decrease of 0.47 degrees per second (95% CI, 0.15-0.78 degrees/s decrease; P = 0.004). The off-off group, who practiced the task when off l-dopa during week 1, had an insignificant (P = 0.87) decrease of 0.02 degrees per second for each additional day of practice. Although the 2 groups were not significantly different at any of the 4 specific testing days during this second week, changes in performance over time did differ between the groups (P = 0.017; test of group:day interaction).

Mirror Drawing

Week 1

At the beginning of week 1, subjects on l-dopa took almost 3 times as long to complete the mirror drawing of the star than those off l-dopa (2.86; 95% CI, 1.35-6.25 times as long; P = 0.006) (Supporting Table 3). From this point, both groups demonstrated significant improvements in performance, although the 2 rates of improvement significantly differed between groups (P = 0.004; test of group:day interaction). The on-off group decreased their time to complete the mirror drawing by 24% each additional day of testing (95% CI, 14%-33%; P < 0.001) while the off-off group decreased their time by 8.6% (95% CI, 5%-12%; P < 0.001) each additional day of testing. The rate of improvement was significantly different between groups (P = 0.004). These trends are displayed in Figure 1.

Week 2

At the beginning of week 2, both groups had mirror drawing times comparable to their performance at the end of week 1 (P = 0.73) and did not improve further with practice. In this second week, the pattern of performance over time was essentially the same for both groups (P = 0.95; test of group:day interaction), implying parallel profiles of performance over time for the 2 groups. On any given day, the on-off group, who had practiced on l-dopa the week prior, had average latencies that were 1.5 (95% CI, 0.98-2.29) times as great as those for the off-off group (P = 0.059).

Discussion

  1. Top of page
  2. ABSTRACT
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Authors Roles
  8. Financial Disclosures
  9. Michael Lasarev: none.
  10. References
  11. Supporting Information

We have examined the effects of l-dopa on motor learning for 3 tasks: a fine motor task involving limb muscles, a postural adjustment task involving axial muscles, and a mirror drawing task requiring reversing normal limb movements. A unique aspect of our study is that we examined the retention of motor learning during continued practice.

Tapping speed and the CoM speed improved with practice over 4 days whether practicing on l-dopa or off l-dopa. These 2 tasks are similar in that they require pointing to a target with different parts of the body, and represent an implicit motor learning process in which subjects improve spatial accuracy with repetition.[26] l-Dopa had no significant effect on the rate of improvement with practice for either task. The improvements persisted for 3 to 4 days after the first 4 days of practice ended, indicating that the practiced motor skills were retained and represent motor learning.[27] The baseline performances of finger tapping and moving CoM to a target were similar in the subjects on and off l-dopa. We attribute the lack of difference in baseline performance in the 2 groups to the relatively mild parkinsonism in our subjects.

The lack of effect of l-dopa on learning these motor tasks is consistent with some but not all studies of l-dopa on learning motor tasks. As in our study, l-dopa has had little effect on implicit motor learning on such exercises as serial reaction time[5, 6] and probabilistic learning.[7] The phase and duration of learning involved may also play a role: in explicit learning using a motor sequence task, l-dopa inhibited early learning and had little effect on later or consolidation phase learning, as we believe occurred with several days of practice in our study.[28, 29] l-Dopa has also been observed to have differential effects depending on whether positive or negative rewards are used with impaired learning in presence of l-dopa observed when negative reinforcement is used.[15] Because our subjects were given feedback on their performance, we might have expected enhanced learning in the presence of l-dopa; however, this feedback was not specifically reward-based. Finally, in tasks involving incremental, feedback-based learning that requires error-correction, l-dopa has had a negative impact[10] that was not seen in our study.

The mirror drawing task differed from the other 2 tasks; it required an inhibition of the normal visual-motor program and the execution of a reverse movement. At baseline, performance of the mirror drawing task was much slower in the subjects on l-dopa than the subjects off l-dopa. Despite their initial poor performance on l-dopa, the on l-dopa group improved their performance with practice over the 4 days of the first week, just barely achieving the performance of the subjects who practiced while off. Because the subjects learning mirror drawing on l-dopa started with a much worse baseline, their rate of learning was faster than the subjects starting off l-dopa. We do not believe that dyskinesia explains the poor performance of this task on the on state because dyskinesia scores were very low, dyskinesia effects were not observed on the other 2 tasks and the mirror drawing improved with practice in the on state.

The pattern we observed with mirror drawing is consistent with other studies in which l-dopa has been shown to interfere with acquisition of new motor skills requiring inhibition of previously learned behavior[11-14] or ignoring distractors that were previously relevant.[30] One explanation for these observations is the “dopamine overdose” hypothesis, suggesting that the dorsal to ventral progression of striatal dopaminergic denervation in PD results in l-dopa improving the performance of activities dependent upon the more dopaminergically denervated dorsal striatum and overdosing and disrupting the executive functions mediated by the less denervated ventral striatum.[26, 31-33]

Retention of learning was examined the second week when the subjects continued to practice the same tasks on 4 consecutive days, but now all subjects had l-dopa withheld overnight before each practice session. Both the subjects learning on l-dopa and those learning off l-dopa demonstrated retention of the previous week's learning by tapping and moving the CoM at the same speed on day 1 of week 2 as they had on the last day of week 1, consistent with motor learning having taken place over the previous week's practice session. The surprising finding was a decay in performance of finger tapping and body movement tasks, when these tasks were practiced off l-dopa during the second week, after initially practicing the tasks on l-dopa the first week. In contrast, the subjects who had learned the tasks off l-dopa either continued to slowly improve or to plateau in their performance on finger tapping and CoM movement the second week.

The decline in performance of tasks learned on when practiced off is a novel observation in parkinsonian humans. However, a similar phenomenon has been observed in parkinsonian rodents. Dopamine-deficient mice, treated with l-dopa, progressively improved their performance on a rotarod with daily practice for 5 days whereas untreated dopamine-deficient mice did not.[16] The improvement in rotarod performance progressively deteriorated when the l-dopa-treated mice subsequently practiced without l-dopa treatment 1 to 10 days later.[16] These observations were explained by an active unlearning or aberrant learning in the dopamine-deficient state.[16, 34] Likewise, rats trained to poke their nose in a hole indicated by a light on either side of their head progressively lost their speed and accuracy at this task unilaterally when the contralateral medial forebrain bundle was lesioned.[35] This observation was interpreted as indicting a loss of the dopamine reward signal. This explanation was supported by the observation that the rats had a similar decay in performance if the food reward for performing the task was stopped.

The progressive decay in performance of our human subjects in the 2 speed-accuracy tasks is similar to the progressive deterioration in performance of tasks learned with l-dopa or normal dopamine in these 2 rodent studies. But which rodent study hypothesis best fits our observations? The human subjects could learn without acute l-dopa treatment possibly because they were not as severely dopamine-depleted as the dopamine-deficient mice. This observation suggests that the human subjects did not have aberrant learning as postulated by Beeler et al.[16] and Beeler[34] as an explanation for the decay in performance with practice after learning on. The human subjects could also learn without l-dopa indicates that acute l-dopa was not required as a reward for learning as postulated by Dowd and Dunnett.[35] However, if the subjects learned on with l-dopa, the reward mediated by l-dopa may have been present, and when this endogenous reward was absent, the subjects' performance deteriorated in accord with Dowd and Dunnett.[35]

The clinical consequences of this decay in performance of tasks learned on and subsequently performed off is unknown. However, our observations raise questions about what treatment conditions are optimal for physical therapy and other learning to take place to have the largest impact on patients' performance in daily life when they are both on and off. The condition in which the learned performance takes place may influence the retention of learned motor tasks. Our results with mirror drawing also suggest that certain types of learning, especially that requiring executive function, may be learned more easily in the off condition, another consideration for the clinician.

Limitations of this study include its relatively small sample size, which in part reflects the substantial time commitment required of our subjects, who were willing to delay taking their morning dose of l-dopa for 8 days in 2 weeks. We note, however, that the study was sufficiently powered to find differences between the learning conditions. Our results may be difficult to generalize across the entire spectrum of disease severity because our subjects primarily had mild-moderate PD.

In conclusion, we studied the effect of l-dopa on acquisition and retention with continued practice of 2 simple speed-accuracy tradeoff motor tasks and a mirror drawing task. Simple motor tasks were learned on or off but continued practice off l-dopa produced worsening performance of tasks learned when on l-dopa. In contrast, we found that in mirror drawing, a more complex motor task requiring inhibition of a learned motor behavior, l-dopa impaired performance but did not inhibit learning and the improvement in performance was maintained when subsequently practiced off.

Acknowledgments

  1. Top of page
  2. ABSTRACT
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Authors Roles
  8. Financial Disclosures
  9. Michael Lasarev: none.
  10. References
  11. Supporting Information

We thank the subjects for their participation in the study, Anna Lovelace, B.A. and Elizabeth Murdock, B.S. research assistants for help with the execution of the protocol, James McNames PhD, for development of the mirror drawing task and Roger Albin, MD, PhD for helpful discussion about mechanisms.

Authors Roles

  1. Top of page
  2. ABSTRACT
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Authors Roles
  8. Financial Disclosures
  9. Michael Lasarev: none.
  10. References
  11. Supporting Information

Elise Anderson: protocol design and execution, data collection and analysis, manuscript writing of first draft and revision. Fay Horak: protocol design, manuscript review and critique. Michael Lasarev: statistical analysis design and execution. John Nutt: protocol design, manuscript review and critique.

Financial Disclosures

  1. Top of page
  2. ABSTRACT
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Authors Roles
  8. Financial Disclosures
  9. Michael Lasarev: none.
  10. References
  11. Supporting Information

Elise Anderson: none.

Fay Horak: Stock Ownership in medically-related fields; APDM. Intellectual Property Rights: Patents and Patents pending: “Movement Monitoring System” (US 2011/0213278 A1); European Patent No. 04736776.8 “A device for conditioning balance and motor co-ordination”; Quantification of turning with body worn inertial sensors (US 2008). Consultancies: none. Expert Testimony: none. Advisory Boards: NCMRR of NIH Employment: OHSU, APDM. Partnerships: none. Contracts: none. Honoraria: Singapore Health Ministry, Reliance Hospital in Mumbai India, U of Sao Paulo, Alaska Physical Therapy Association Royalties: BESTest Educational DVD from OHSU. Grants: NIH, VA. PD alliance Other: none.

Michael Lasarev: none.

  1. Top of page
  2. ABSTRACT
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Authors Roles
  8. Financial Disclosures
  9. Michael Lasarev: none.
  10. References
  11. Supporting Information

John Nutt: Stock Ownership in medically-related fields - none. Intellectual Property Rights: none. Consultancies: 1. Elan Pharmaceuticals; 2. Lundbeck Inc.; 3. ONO Pharma; 4. SynAgile Corp; 5. Prexa Inc.; 6. US World Med.; Ceregene. Expert Testimony: none. Advisory Boards: none Employment: OHSU, VA. Partnerships: none. Contracts: none. Honoraria: Speaking at American Academy of Neurology. Royalties: none. Grants: 1. National Parkinson Foundation; 2. NIH; 3. Michael J. Fox Foundation; Ceregene. Other: none.

References

  1. Top of page
  2. ABSTRACT
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Authors Roles
  8. Financial Disclosures
  9. Michael Lasarev: none.
  10. References
  11. Supporting Information
  • 1
    Doyon J. Motor sequence learning and movement disorders. Curr Opin Neurol 2008;21:478-483.
  • 2
    Siegert JS, Taylor KD, Weatherall M, Abernethy DA. Is implicit sequence learning impaired in Parkinson's disease? A meta-analysis. Neuropsychology 2006;20:490-495.
  • 3
    Marinelli L, Crupi D, Di Rocco A, et al. Learning and consolidation of visuo-motor adaptation in Parkinson's disease. Parkinsonism Relat Disord 2009;15:6-11.
  • 4
    Mochizuki-Kawai H, Kawamura M, Hasegawa Y, et al. Deficits in long-term retention of learned motor skills in patients with cortical or subcortical degeneration. Neuropsychologia 2004;42:1858-1863.
  • 5
    Muslimovic D, Post B, Speelman JK, Schmand B. Motor procedural learning in Parkinson's disease. Brain 2007;130:2887-2897.
  • 6
    Ghilardi MF, Feigin AS, Battaglia F, et al. L-Dopa infusion does not improve explicit sequence learning in Parkinson's disease. Parkinsonism Relat Disord 2007;13:146-151.
  • 7
    Shiner T, Seymour B, Wunderlich K, et al. Dopamine and performance in a reinforcement learning task: evidence from Parkinson's disease. Brain 2012;135:1871-1883.
  • 8
    Feigin A, Ghilardi MF, Carbon M, et al. Effects of levodopa on motor sequence learning in Parkinson's disease. Neurology 2003;60:1744-1749.
  • 9
    Tremblay PL, Bedard MA, Langlois D, Blanchet PJ, Lemay M, Parent M. Movement chunking during sequence learning is a dopamine-dependent process: a study conducted in Parkinson's disease. Exp Brain Res 2010;2005:375-385.
  • 10
    Shohamy D, Myers CE, Geghman KD, Sage J, Gluck MA. L-Dopa impairs learning, but spares generalization, in Parkinson's disease. Neuropsychologia 2006;44:774-784.
  • 11
    Cools R, Barker RA, Sahakian BJ, Robbins TW. Enhanced or impaired cognitive function in Parkinson's disease as a function of dopaminergic medication and task demands. Cereb Cortex 2001;11:1136-1143.
  • 12
    Swainson R, Rogers RD, Sahakian BJ, Summers BA, Polkey CE, Robbins TW. Probabilistic learning and reversal deficits in patients with Parkinson's disease or frontal or temporal lobe lesions: possible adverse effects of dopaminergic medication. Neuropsychologia 2000;38:596-612.
  • 13
    Graef S, Biele G, Krugel LK, et al. Differential influence of levodopa on reward-based learning in Parkinson's disease. Front Hum Neurosci. 2010;4:1-13.
  • 14
    Marzinzik F, Wotka J, Wahl M, Krugel LK, Kordsachia C, Klostermann F. Modulation of habit formation by levodopa in Parkinson's disease. PLoS One 2011;6:e27695.
  • 15
    Frank MJ, Seeberger LC, O'Reilly RC. By carrot or by stick: cognitive reinforcement learning in parkinsonism. Science 2004;306:1940-1943.
  • 16
    Beeler JA, Cao ZF, Kheirbek MA, et al. Dopamine-dependent motor learning: insight into levodopa's long duration response. Ann Neurol 2010;67:639-647.
  • 17
    Gelb DJ, Oliver E, Gilman S. Diagnostic criteria for Parkinson disease. Arch Neurol 1999;56:33-39.
  • 18
    Nutt JG, Lea ES, Van Houten L, Schuff RA, Sexton GJ. Determinants of tapping speed in normal control subjects and subjects with Parkinson's disease: differing effects of brief and continued practice. Mov Disord 2000;15:843-849.
  • 19
    Nutt JG, Carter JH, Lea ES, Sexton GJ. Evolution of the response to levodopa during the first 4 years of therapy. Ann Neurol 2002;51:686-693.
  • 20
    Jessop RT, Christopher Horowicz C, Dibble LE. Motor learning and Parkinson disease: refinement of movement velocity and endpoint excursion in a limits of stability balance task. Neurorehabil Neural Repair 2006;20:459.
  • 21
    Agostino R, Sanes JN, Hallett M. Motor skill learning in Parkinson's disease. J Neurol Sci 1996;139:218-226.
  • 22
    Harrington DL, Haaland KY, Yeo RA, Marder E. Procedural memory in Parkinson's disease: impaired motor but not visuoperceptual learning. J Clin Exp Neuropsychol 1990;12:323-339.
  • 23
    Schnider A, Gutbrod K, Hess CW. Motion imagery in Parkinson's disease. Brain 1995;118:485-493.
  • 24
    Liang KY, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika 1986;73:13-22.
  • 25
    Hardin JW, Hilbe JM. Generalized Estimating Equations. Boca Raton, FL: Chapman & Hall/CRC; 2003.
  • 26
    Ghilardi MF, Eidelberg D, Silvestri G, Ghez C. The differential effect of PD and normal aging on early explicit sequence learning. Neurology 2003;60:1313-1319.
  • 27
    Schmidt RA, Wrisberg CA. Motor Learning and Performance With Web Study Guide: A Situation-Based Learning Approach. 4th ed. Champaign IL: Human Kinetics; 2007.
  • 28
    Kwak Y, Muller M, Bohnen NI, Dayalu P, Seidler RD. Effect of dopaminergic medications on the time course of explicit motor sequence learning in Parkinson's disease. J Neurophysiol 2010;103:942-949.
  • 29
    Kwak Y, Muller M, Bohnen NI, Dayalu P, Seidler RD. L-DOPA changes ventral striatum recruitment during motor sequence learning in Parkinson's disease. Behav Brain Res 2012;230:116-124.
  • 30
    Moustafa AA, Sherman SJ, Frank MJ. A dopaminergic basis for working memory, learning and attentional shifting in Parkinsonism. Neuropsychologia 2008;46:3144-3156.
  • 31
    Cools R, Lewis SJ, Clark L, Barker RA, Robbins TW. L-DOPA disrupts activity in the nucleus accumbens during reversal learning in PD. Neuropsychopharmacology 2007;32:180-189.
  • 32
    Gotham AM, Brown RG, Marsden CD. Frontal cognitive function in patients with PD on and off levodopa. Brain 1988;111:299-321.
  • 33
    Kulsievsky J. Role of dopamine in learning and memory implications for the treatment of cognitive dysfunction in patients with Parkinson's disease. Drugs Aging 2000;16:365-379.
  • 34
    Beeler J. Preservation of function in Parkinson's disease: what's learning got to do with it? Brain Res 2011;1423:96-113.
  • 35
    Dowd E, Dunnett SB. Movement without dopamine: striatal dopamine is required to maintain but not to perform learned actions. Biochem Soc Trans 2007;35:428-432.

Supporting Information

  1. Top of page
  2. ABSTRACT
  3. Patients and Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Authors Roles
  8. Financial Disclosures
  9. Michael Lasarev: none.
  10. References
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
mds25702-sup-0001-suppInfo.docx16KSupplementary Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.