SEARCH

SEARCH BY CITATION

Keywords:

  • basal ganglia;
  • cerebellum;
  • Huntington's disease;
  • learning dissociation;
  • Parkinson's disease;
  • sensorimotor learning

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Visuomotor adaptation is often driven by error-based (EB) learning in which signed errors update motor commands. There are, however, visuomotor tasks where signed errors are unavailable or cannot be mapped onto appropriate motor command changes, rendering EB learning ineffective; and yet, healthy subjects can learn in these EB learning-free conditions. While EB learning depends on cerebellar integrity, the neural bases of EB-independent learning are poorly understood. As basal ganglia are involved in learning mechanisms that are independent of signed error feedback, here we tested whether patients with basal ganglia lesions, including those with Huntington's disease and Parkinson's disease, would show impairments in a visuomotor learning task that prevents the use of EB learning. We employed two visuomotor throwing tasks that were similar, but were profoundly different in the resulting visual feedback. This difference was implemented through the introduction of either a lateral displacement of the visual field via a wedge prism (EB learning) or a horizontal reversal of the visual field via a dove prism (non-EB learning). Our results show that patients with basal ganglia degeneration had normal EB learning in the wedge prism task, but were profoundly impaired in the reversing prism task that does not depend on the signed error signal feedback. These results represent the first evidence that human visuomotor learning in the absence of EB feedback depends on the integrity of the basal ganglia.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

When learning a visuomotor task such as throwing a dart, the ability to judge how far and in which direction the dart went in relation to the target leads to subsequent improvements in throwing accuracy. In this case, the visuomotor system relies on signed error signals to make parametrical adjustments to an internal model that helps guide the next throw [error-based (EB) learning] (Shadmehr et al., 2010; Wolpert et al., 2011). A classic EB example is the adaptation to wedge prisms, where subjects use the signed error to update their motor commands (Martin et al., 1996a; Fernandez-Ruiz & Diaz, 1999). In these experiments, subjects throw balls to a target while looking through refracting wedge prisms that produce a lateral displacement of the visual field (Fig. 1A). Then, based on the error feedback, subjects must adjust their throws to compensate for the optical perturbation, i.e. the adjustment is based on the sensorimotor error. Reward prediction errors, by contrast, are the net success or failure to reach a goal that modifies the valuation of the sensory states that result from the motor commands (Izawa & Shadmehr, 2011). Prism adaptation is only one example of EB learning, but other tasks such as visuomotor rotations, grip force adaptations and force field perturbations have also been studied (Flanagan & Wing, 1997; Krakauer, 2009; Shadmehr et al., 2010; Wolpert et al., 2011).

image

Figure 1. Task set-up and timeline. (A) Participants were asked to throw 25 balls to a target (black cross) 2 m in front and 20 cm to the left of their midline. However, during the prism phase they always perceived the target 20 cm to the right of their midline. (B) Diagram illustrating the three phases in each of the learning tasks.

Download figure to PowerPoint

Recently it has been shown that when the perceived errors do not provide feedback about the required direction of the behavioral change, rendering the EB mechanism useless, it is still possible to show improvements in sensorimotor tasks (Izawa & Shadmehr, 2011; Lillicrap et al., 2013). The neural bases of this visuomotor learning in the absence of error direction information are currently unknown. However, the basal ganglia (BG) have been related to the implementation of cognitive strategies that adjust the response, the acquisition of habits or the action of reinforcement learning (Redgrave et al., 1999; Fernandez-Ruiz et al., 2001; Frank et al., 2004; Yin & Knowlton, 2006; Grahn et al., 2009; Huang et al., 2011; Izawa & Shadmehr, 2011; Dezfouli & Balleine, 2012). Therefore, we tested the hypothesis that visuomotor learning is dependent upon processes carried out by or in a participant with the BG in the absence of feedback in a visuomotor adaptation task. To test this hypothesis we investigated patients with BG deficits, more specifically patients with Parkinson's disease (PD) and patients with Huntington's disease (HD), using a non-EB learning task. PD is characterized by the reduction of dopamine levels in the nucleus caudate and the putamen due to the loss of dopamine-producing neurons in the substantia nigra (Damier et al., 1999), while HD is characterized by the degeneration of the caudate nucleus and the putamen (Vonsattel et al., 1985). Degeneration in these areas results in distinct functional differences that have been characterized as either a hypokinetic disorder (PD) or a hyperkinetic disorder (HD; DeLong, 1990). Because these important differences can have specific cognitive consequences we decided to test both populations (Heindel et al., 1989; Litvan et al., 1998; Aretouli & Brandt, 2010). Our experiment included two similar visuomotor tasks, one that requires EB learning (visuomotor adaptation to wedge prisms) and the other that discourages the use of EB learning (visuomotor adaptation to reversing prisms). We predicted that both patient groups with BG deficits would perform normally in the visuomotor EB learning task but be impaired in the visuomotor non-EB learning task.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Participants

Twenty-four HD patients (mean age 43.5 ± 12 years; 14 females) and 17 PD patients (mean age 61.5 ± 8.26 years; seven females) were recruited from the Genetics Clinic at the Instituto Nacional de Neurologia y Neurocirugia ‘Manuel Velasco Suarez’ (see Supporting Information Tables S1 and S2 for detailed patient information). All HD patients included in this study had a molecular diagnosis of the disease with 38 or more CAG (glutamate) trinucleotide repetitions. All PD patients had a confirmed clinical diagnosis. PD patients were tested before having taken their first morning medication for their Parkinson's symptoms. All patients where right-handed and were paired with two groups of right-handed healthy control subjects matched as possible for age and gender [CHD (HD control) mean age 42.5 ± 11.9 years, 15 females; and CPD (PD control) 62 ± 6.1 years, eight females]. All procedures were in accordance with the ethical standards of the committees on human experimentation of both the Instituto Nacional de Neurología y Neurocirugía ‘Manuel Velasco Suarez’ and the Universidad Nacional Autónoma de México. The study was approved by the Research and Ethics committee of the Faculty of Medicine of the Universidad Nacional Autónoma de México. All subjects gave informed consent prior to the experiments in accordance with the Helsinki Declaration (Council for International Organizations of Medical Sciences and World Health Organization, 2002).

Experimental tasks

The tasks were throwing tasks in which subjects had to throw balls towards a target in front of them (Martin et al., 1996b; Fernandez-Ruiz & Diaz, 1999). The critical variable was that in the EB task a wedge displacing prism was introduced while in the non-EB task a reversing prism was introduced. These two kinds of prisms have profound differences in the context of the feedback they produce. An explanation of the effect these two different kinds of perturbations have on the feedback of the visuomotor system is given below.

Task descriptions

While seated, subjects rested their heads on a chin support that had attached an occluding panel with a 5 × 5-cm window so the subjects could only see with their right eye (Fig. 1A). This set-up occluded the view of the subject's hand during the experiments. Both tasks consisted of throwing clay balls (weight – 10 g) at a 12 × 12-cm cross drawn on a large sheet of parcel paper centred at shoulder level and placed 2 m away. Subjects were instructed to make overhand tosses during the whole experiment with the right hand.

Each experiment had three phases, a baseline, a test and a post phase, each consisting of 26 trials (Fig. 1B). After the baseline phase, subjects were informed that a prism was going to be introduced in the set-up. However, the nature of the prism was not revealed, and no further explanation was given. Then, a prism was fit by the experimenter in the viewing window. In the EB learning condition, subjects looked at the target through a wedge prism (20 diopters), which effectively displaced the perceived position of the target 40 cm (11.31°) to the right with respect to the real target (Fig. 2A). In the non-EB learning condition, subjects looked at the target through a dove prism that resulted in a horizontal reversing of the perceived position of the target by 40 cm (11.31o) to the right with respect to the real target (Fig. 2B). In each group half of the subjects were tested first in the EB condition and the other half started in the non-EB condition. These conditions varied only in the nature of the feedback (Fig. 2; see task rationale). During the post phase, the experimenter removed the prisms before the subject finished the last 26 trials (Fig. 1B). The location of the impacts was plotted sequentially by trial number (abscissa) vs. horizontal displacement (in centimetres) from a vertical line passing through the target centre (ordinate). Impacts to the left of the target were plotted as negative values and impacts to the right were plotted as positive values.

image

Figure 2. Task rationale. To hit the target (black cross) after the introduction of either prism, participants had to throw 40 cm left of the perceived target (grey cross). (A) After the introduction of wedge prisms in the error-based task, participants could use the visual feedback of the ball hitting the wall (signed error signal e) to adjust their responses in subsequent trials. (B) After the introduction of dove reversing prisms, if participants used e they would end up throwing the ball further away from the target, and therefore their visual feedback would be useless. In this condition, throwing to the same side of the target would be rewarded, although the error would still be highly variable and there would be no aftereffects.

Download figure to PowerPoint

Motor performance (MP) was defined as the variance shown during the baseline throws. The adaptation measure was defined as the mean difference between the final throw and the first throw of the prism condition. The aftereffect measure was defined as the distance to the target in the first throw of the post phase. All transitions between the different prism conditions were done while subjects remained with their eyes closed.

Task rationale

Note that the prism tasks were identical in the sense that subjects always perceived a target located 20 cm to the right of the midline, and to hit the target, they had to throw 40 cm to the left. However, these tasks have a fundamental difference. In the laterally displacing wedge prism task (Fig. 2A), subjects see the target to their right, but when they throw, they see the ball hitting even further to the right of the target (see trial 1 in Fig. 2A). To correct for that error, they need to throw more to the left, a response that is congruent with the direction of the error they perceive. On the other hand, in the reversing dove prism task (Fig. 2B), subjects also see the target to their right, but when they throw, they see the ball hitting to the left of the target (see trial 1 in Fig. 2B). To correct for that error, they must throw more to the left, a response that is incongruent with the direction of the error they perceive. The critical difference is that the error signal cannot be mapped onto appropriate motor command changes in the second task, i.e. the error signal is reversed, creating a maladaptive feedback that, if used, results in a counterproductive update of the internal model (see trial 2 in Fig. 2B; Lillicrap et al., 2013). Previous studies have shown that healthy subjects reduce the error using EB in the wedge prism condition (Fig. 2A; Martin et al., 1996b). However, the response pattern found in the reversing dove prism task is not consistent with EB (Fig. 2B), as subjects show a highly variable error correction followed by no aftereffects once the prism is withdrawn (Lillicrap et al., 2013).

Statistical analysis

One-way analyses of variance (anovas) were used for testing the motor performance, adaptation and aftereffect magnitudes. Repeated-measures anovas were used for within-groups results analyses between both kinds of prisms. If the homogeneity of the variances was not met as tested with the Levene statistic, then a Dunnett T3 post hoc test was applied. All statistics were run using IBM spss Statistics 20 (IBM Corp., Armonk, NY, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Motor performance

As expected, control groups had a better MP than the patient groups. The analysis showed significant group differences (F3,160 = 18.09, < 0.01). The post hoc test revealed differences between the CHD and HD groups (< 0.01), but only a trend between the CPD and PD groups (= 0.09; Fig. 3). No differences were found between the control groups (= 0.5).

image

Figure 3. Motor performance deficits in the patient groups measured as the standard deviation of the combined baseline for the EB and non-EB experiments. HD and PD are Huntington's disease and Parkinson's disease patients, respectively, and CHD and CPD are the control groups for HD and PD, respectively. Error bars are SEM. *< 0.01.

Download figure to PowerPoint

Non-error learning vs. error leaning in control groups

An important aspect in our experimental design was the assumption that the adaptation to wedge and dove prisms would require the implementation of different processes (Lillicrap et al., 2013). As an initial step to confirm that the adaptation processes were different for each prism, we compared the control groups' sensorimotor adaptation for both tasks (Fig. 4, darker symbols in each panel). Based on our previous work, we predicted that the control groups would be able to adapt to both tasks (Lillicrap et al., 2013). The repeated-measures analysis confirmed our hypothesis. The control groups showed no adaptation differences between the two learning tasks (F1,39 = 0.804, = 0.37; Fig. 4, darker symbols in the prism section of A and B, and the x-axis in C).

image

Figure 4. Visuomotor learning results. (A) Error distance in the EB learning task during baseline, prism (wedge prism) and generalization (aftereffects) trials. Note that control and patient data symbols overlap. (B) Same as (A), but for non-EB (dove reversing prisms). (C) Adaptation and aftereffect measures for all groups. Note that there were no differences in EB learning between controls (dark blue symbols) and patients (light blue symbols). However, there were significant impairments in both HD and PD once the signed error feedback was rendered useless (reddish symbols). Also note the lack of negative aftereffects in the non-EB task compared with the EB task. *< 0.01. Error bars are ± SEM. Each symbol represents the group average for that data point. Controls – dark symbols, patients – light symbols.

Download figure to PowerPoint

Another characteristic of EB learning is a smooth learning rate across trials, followed by significant negative aftereffects; therefore, if the reversing prism task could not be learned using EB, then a more variable error reduction during the adaptation phase was expected, followed by a lack of aftereffects. Analysis of the standard error of the distance to the target across the adaptation throws showed significant differences between EB and non-EB MP (F1,39 = 20.4, < 0.01). Finally, the aftereffect analysis did show a negative aftereffect in the EB task that was absent in the non-EB task, leading to significant aftereffect differences (F1,39 = 53.25, < 0.01; Fig. 4, darker symbols in the after effects section of A and B, and the y-axis in C).

Error-learning in patients with BG lesions

As EB learning has been shown to depend on the cerebellum and not on the BG integrity, we expected the patient groups to perform at normal rates in this task. The results confirmed this hypothesis. A one-way anova between the EB adaptation magnitudes obtained at the end of the prism phase showed that there were no significant differences (F3,78 = 0.17, = 0.91; Fig. 4A and C, light blue symbols). Analysis of the aftereffects showed similar results (F3,78 = 1.67, = 0.18), corroborating previous findings in HD and PD patients (Fig. 4A and C, light blue symbols; Fernandez-Ruiz et al., 2003).

Non-error learning in patients with BG lesions

The critical test was the analysis of the patients' performance in non-EB learning. Our main hypothesis was that non-EB learning depends on processes carried out by the BG or cortico-BG circuits. Therefore, deterioration of these systems due to the neurodegenerative processes occurring in HD or PD should lead to a detriment in non-EB learning. The results clearly confirm our hypothesis.

A one-way anova between the four groups adaptation magnitudes obtained at the end of the prism phase showed that there were significant differences (F3,78 = 7.06, < 0.01). Post hoc test analyses showed significant differences between CHD and HD groups (< 0.01), as well as significant differences between CPD and PD groups (= 0.05; Fig. 4B and C, x-axis). There were no differences between control groups, or between patient groups. The aftereffect analysis showed a general effect (F3,78 = 2.98, < 0.05), but none of the comparisons yielded a significant result in the post hoc analysis (Fig. 4C, y-axis).

EB learning and non-EB learning dissociation in the BG patient groups

The pattern of results between the control and patients groups suggested a single dissociation between EB and non-EB learning. However, a direct comparison between the performances of the patient groups on both tasks would confirm the dissociation within the same subjects. The analysis showed no differences between the HD and PD groups (F1,39 = 0.3, = 0.58). However, there was a significant adaptation difference between the EB and non-EB tasks (F1, 39 = 66.81, < 0.01). A similar analysis on the aftereffect results showed a similar pattern. There were no differences between the patient groups (F1,39 = 0.17, = 0.68), but significant differences between tasks (F1,39 = 36.85, < 0.01; Fig. 4C, light symbols).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

In this study we tested the hypothesis that in the absence of feedback that informs about the required direction of a behavioural change, then visuomotor learning is dependent upon processes carried out by or in participation with the BG. The results show that both HD and PD patients could not improve their performance once the signed error feedback was rendered useless by the reversing prisms. However, the same patients showed normal adaptation in the wedge prism signed EB task. Therefore, here we propose that the BG, or the cortico-basal circuits, participate in visuomotor learning processes that do not depend on a signed error feedback.

Error-based learning

It has been proposed that during sensorimotor learning such as prism adaptation, the visuomotor system uses the signed error feedback to update the controller to adjust to current environmental changes (Wolpert et al., 2011). Therefore, learning occurs on a trial by trial basis and, once the environmental perturbation is withdrawn, a negative aftereffect can be observed (Martin et al., 1996a; Fernandez-Ruiz & Diaz, 1999). Our results support the hypothesis that patients with BG lesions have normal EB learning performance, despite having significant deficits in motor control. This result suggests that the capacity to update internal models in response to external perturbations in open loop tasks is not dependent on the BG, as has been previously shown in similar studies (Weiner et al., 1983; Fernandez-Ruiz et al., 2003; Smith & Shadmehr, 2005; Marinelli et al., 2009). In contrast, other studies in patients with cerebellar damage suggest that this capacity is supported by cerebellar circuits (Weiner et al., 1983; Martin et al., 1996b; Smith & Shadmehr, 2005; Fernandez-Ruiz et al., 2007), a finding that has been replicated in many different tasks (Fiez et al., 1992; Jenkins et al., 1994; Jueptner et al., 1997).

Non-error based learning in control groups

In many sensorimotor learning tasks the signed error is used to adjust the motor command (Shadmehr et al., 2010; Wolpert et al., 2011; Lillicrap et al., 2013). To do so, not only must subjects know the magnitude of the error, but they also need to know the direction of the error. This concept is difficult to explain because almost always the error magnitude has an implicit sign that rarely has been dissociated. To try to explain it, imagine a scenario in which a subject throws darts at a dartboard, but with his or her eyes closed. If the subject is informed only that the previous dart hit 10 cm from the bullseye, will the subject be able to correct? The subject cannot know if the correction should be 10 cm to the left or to the right, or up or down? To be able to correct, the system should have a mechanism to map the sign of the error and not only its magnitude. If the error sign is not available, then the system has no information about where the motor adjustment should be done, regardless of the magnitude of the error. So, without the error sign, knowing the error magnitude is useless for EB learning (Shadmehr et al., 2010; Wolpert et al., 2011; Lillicrap et al., 2013).

In the dove prism experiment the error sign is reversed (Lillicrap et al., 2013). Therefore, if subjects miss to the right, through the dove prisms they will see the error to the left, and will try to correct to the right (Fig. 2B). However, to reduce the error they should correct to the left because the error-action correction mapping is reversed (Fig. 2B). In previous experiments we have shown that healthy subjects cannot update this parameter in one session, or even within days of continuing training (Lillicrap et al., 2013). Therefore, subjects cannot use the signed error signal information, and yet, they improve their response within the task trials. We think that this improvement is instantiated by mechanisms different from those observed during the typical adaptation to wedge prisms. For instance, the response to reversing prisms shows an increase in space exploration as measured by a larger throwing variance, a pattern that is maintained even after localizing the target. Note also that while control subjects were homogeneous in the adaptation magnitudes during the EB learning, they displayed a large variance in the adaptation magnitudes during the non-EB learning (Fig. 2C), even including some control subjects that could not learn the non-EB task, as has been reported previously (Lillicrap et al., 2013). Also, once the prism is withdrawn, there are no negative aftereffects; this lack of aftereffects contrasts with the negative aftereffects observed after wedge prism adaptation, in which many trials are needed to re-arrange the system to its previous value. This large aftereffect magnitude is characteristic of EB learning (Fernandez-Ruiz & Diaz, 1999; Moreno-Briseno et al., 2010; Izawa & Shadmehr, 2011). Thus, the results from the control groups suggest that the healthy subjects' behavioural improvement probably uses mechanisms that differ from those involved in signed error-learning (Abdelghani et al., 2008; Lillicrap et al., 2013).

Non-error based learning in patient groups

If EB learning is not the primary mechanism involved during the reversing prism task, then other learning mechanisms, including those instantiated by the BG, could be used to implement this non-EB learning task. Although it has been assumed that structures such as the BG could instantiate these non-EB process in human visuomotor learning, to our knowledge, this has never been proved before (Doya, 2000; Doyon et al., 2003; Krakauer, 2009; Shadmehr et al., 2010; Wolpert et al., 2011).

The critical test for probing this hypothesis was to find the effect of BG deterioration, caused by HD or PD, on non-EB learning. The results of this experiment show that both patient groups had a profound deficit when exposed to the dove reversing prisms. In contrast to their behaviour in the EB learning task, in the non-EB task they show a large throwing variability that remained far from the target by the end of the trials. Interestingly, during the non-EB task some subjects showed a gradual increase in error magnitude, suggesting a maladaptive response driven by EB learning. However, because the error signal was reversed, their throws ended up gradually going away from the target. This inability to show behavioural improvements once they cannot make use of the signed error signal could be due to the disruption of different mechanisms, including deficits in the implementation of cognitive strategies (Redgrave et al., 1999; Grahn et al., 2009), habit impairments (Dezfouli & Balleine, 2012) and deficits in reinforcement learning (Frank et al., 2004). These mechanisms have been extensively studied outside the sensorimotor learning field, and all involve the BG, including the cortico-BG loops (Owen et al., 1992; Schultz et al., 2000; Fernandez-Ruiz et al., 2001; Graybiel, 2005; Taylor & Ivry, 2012).

Kinds of sensorimotor learning

To understand the nature of this impairment it is useful to review previous findings on patients with BG lesions. Deficits in motor performance are hallmark characteristics of patients with lesions of the BG (Alexander & Crutcher, 1990; Graybiel et al., 1994). Therefore, it is logical to assume that they may also have deficits in motor learning. Initial studies in different patient populations confirmed the BG participation in learning motor skills (Heindel et al., 1988; Harrington et al., 1990). In relation to visuomotor learning, however, studies using representational feedback using a cursor or the like have led to conflicting results. For example, the introduction of a perturbation consisting of a 90° rotation in a reaching task resulted in learning impairments in PD patients (Contreras-Vidal & Buch, 2003), while the introduction of a 30o rotation led to normal learning (Marinelli et al., 2009). A plausible explanation for this discrepancy suggests that the mechanisms used to solve both perturbations are different (Marinelli et al., 2009). While the 30o perturbation could rely mainly on EB learning mechanisms, the 90o perturbation would invoke other mechanisms, such as the implementation of cognitive strategies, which are accompanied by increases in reaction times (Fernandez-Ruiz et al., 2011). Therefore, the involvement of cognitive processes would be more related to the participation of cortico-BG loops that have been well established in many different learning paradigms (Doyon et al., 1997, 2003; Toni & Passingham, 1999; Murray et al., 2000; Fernandez-Ruiz et al., 2001; Packard & Knowlton, 2002; Frank et al., 2004; Haruno et al., 2004; Shohamy et al., 2004; Pasupathy & Miller, 2005; Rosas et al., 2005; Palminteri et al., 2011).

Based on these experimental findings, theoretical advances in motor learning suggested the possible implementation of different mechanisms during sensorimotor learning, including cognitive, reinforcement and EB mechanisms (Doya, 2007; Huang et al., 2011; Wolpert et al., 2011). However, until now, there have been no specific experiments published to test the involvement of BG in human visuomotor learning. Our experiment specifically addressed this idea by showing dissociation between the lack of effect of BG lesions on an open-loop task mainly based on error learning and the profound disruption that occurs when the signed error signal is rendered ineffective. This disruption could be due to the interference with processes carried out by the BG or the cortico-BG loops, which may include action selection, habit learning or reinforcement learning (Redgrave et al., 1999; Schultz et al., 2000; Frank et al., 2004; Ito & Doya, 2011). At this point, however, linking the observed deficits to one specific mechanism would be unreliable, although the observed BG-dependent deficits are robust and provide firm ground for future hypothesis-driven experiments in this field.

Study limitations

The hypothesis we wanted to address was that visuomotor learning is dependent upon processes carried out by or in participation with the BG in the absence of feedback in a visuomotor adaptation task. The approach we used involved testing patients whose diseases are mainly characterized as deficits in the BG. However, this approach comes with a caveat – HD and PD affect other systems beyond the BG. A recent meta-analysis including 180 HD patients suggests that HD also results in significant grey matter decrease in the bilateral dorsolateral prefrontal cortex, the bilateral anterior insula as well as the characteristic bilateral striatum (Lambrecq et al., 2013). Besides the characteristic lesions of the substantia nigra, PD also affects other systems, including a significant decrease in grey matter volume in the temporal lobe (Braak et al., 1996; Tam et al., 2005; Pereira et al., 2012). Therefore, although both diseases have behavioural manifestations that result from BG lesions, our results should be considered discretely.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Here we show for the first time that HD and PD patients show significant visuomotor learning impairments when a signed error signal is not available to inform the system. These results suggest that it is possible to dissociate two kinds of sensorimotor learning mechanisms – one based on signed error feedback that could depend on cerebellar integrity (Doya, 2000), and another that is independent of signed error feedback that is impaired in patients with HD or PD. Further research is needed to explore the participation of BG in this kind of visuomotor learning.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

We thank all participants, especially the patients and their families, who freely decided to participate in this study. J.F-R. was supported by grants from CONACYT, 102314 COVECYT 127808 and PAPIIT-DGAPA IN202810. The authors declare no competing financial interests.

Abbreviations
BG

basal ganglia

CHD

control Huntington's disease

CPD

control Parkinson's disease

EB

error-based

HD

Huntington's disease

MP

motor performance

PD

Parkinson's disease

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
  • Abdelghani, M.N., Lillicrap, T.P. & Tweed, D.B. (2008) Sensitivity derivatives for flexible sensorimotor learning. Neural Comput., 20, 20852111.
  • Alexander, G.E. & Crutcher, M.D. (1990) Functional architecture of basal ganglia circuits: neural substrates of parallel processing. Trends Neurosci., 13, 266271.
  • Aretouli, E. & Brandt, J. (2010) Episodic memory in dementia: Characteristics of new learning that differentiate Alzheimer's, Huntington's, and Parkinson's diseases. Arch. Clin. Neuropsych., 25, 396409.
  • Braak, H., Braak, E., Yilmazer, D., de Vos, R.A., Jansen, E.N. & Bohl, J. (1996) Pattern of brain destruction in Parkinson's and Alzheimer's diseases. J. Neural Transm., 103, 455490.
  • Contreras-Vidal, J.L. & Buch, E.R. (2003) Effects of Parkinson's disease on visuomotor adaptation. Exp. Brain Res., 150, 2532.
  • Damier, P., Hirsch, E.C., Agid, Y. & Graybiel, A.M. (1999) The substantia nigra of the human brain. II. Patterns of loss of dopamine-containing neurons in Parkinson's disease. Brain, 122, 14371448.
  • DeLong, M.R. (1990) Primate models of movement disorders of basal ganglia origin. Trends Neurosci., 13, 281285.
  • Dezfouli, A. & Balleine, B.W. (2012) Habits, action sequences and reinforcement learning. Eur. J. Neurosci., 35, 10361051.
  • Doya, K. (2000) Complementary roles of basal ganglia and cerebellum in learning and motor control. Curr. Opin. Neurobiol., 10, 732739.
  • Doya, K. (2007) Reinforcement learning: computational theory and biological mechanisms. HFSP J., 1, 3040.
  • Doyon, J., Gaudreau, D., Laforce, R. Jr., Castonguay, M., Bedard, P.J., Bedard, F. & Bouchard, J.P. (1997) Role of the striatum, cerebellum, and frontal lobes in the learning of a visuomotor sequence. Brain Cognition, 34, 218245.
  • Doyon, J., Penhune, V. & Ungerleider, L.G. (2003) Distinct contribution of the cortico-striatal and cortico-cerebellar systems to motor skill learning. Neuropsychologia, 41, 252262.
  • Fernandez-Ruiz, J. & Diaz, R. (1999) Prism adaptation and aftereffect: specifying the properties of a procedural memory system. Learn. Memory, 6, 4753.
  • Fernandez-Ruiz, J., Wang, J., Aigner, T.G. & Mishkin, M. (2001) Visual habit formation in monkeys with neurotoxic lesions of the ventrocaudal neostriatum. Proc. Natl. Acad. Sci. USA, 98, 41964201.
  • Fernandez-Ruiz, J., Diaz, R., Hall-Haro, C., Vergara, P., Mischner, J., Nunez, L., Drucker-Colin, R., Ochoa, A. & Alonso, M.E. (2003) Normal prism adaptation but reduced after-effect in basal ganglia disorders using a throwing task. Eur. J. Neurosci., 18, 689694.
  • Fernandez-Ruiz, J., Velasquez-Perez, L., Diaz, R., Drucker-Colin, R., Perez-Gonzalez, R., Canales, N., Sanchez-Cruz, G., Martinez-Gongora, E., Medrano, Y., Almaguer-Mederos, L., Seifried, C. & Auburger, G. (2007) Prism adaptation in spinocerebellar ataxia type 2. Neuropsychologia, 45, 26922698.
  • Fernandez-Ruiz, J., Wong, W., Armstrong, I.T. & Flanagan, J.R. (2011) Relation between reaction time and reach errors during visuomotor adaptation. Behav. Brain Res., 219, 814.
  • Fiez, J.A., Petersen, S.E., Cheney, M.K. & Raichle, M.E. (1992) Impaired non-motor learning and error detection associated with cerebellar damage. A single case study. Brain, 115, 155178.
  • Flanagan, J.R. & Wing, A.M. (1997) The role of internal models in motion planning and control: evidence from grip force adjustments during movements of hand-held loads. J. Neurosci., 17, 15191528.
  • Frank, M.J., Seeberger, L.C. & O'Reilly, R.C. (2004) By carrot or by stick: cognitive reinforcement learning in parkinsonism. Science, 306, 19401943.
  • Grahn, J.A., Parkinson, J.A. & Owen, A.M. (2009) The role of the basal ganglia in learning and memory: neuropsychological studies. Behav. Brain Res., 199, 5360.
  • Graybiel, A.M. (2005) The basal ganglia: learning new tricks and loving it. Curr. Opin. Neurobiol., 15, 638644.
  • Graybiel, A.M., Aosaki, T., Flaherty, A.W. & Kimura, M. (1994) The basal ganglia and adaptive motor control. Science, 265, 18261831.
  • Harrington, D.L., Haaland, K.Y., Yeo, R.A. & Marder, E. (1990) Procedural memory in Parkinson's disease: impaired motor but not visuoperceptual learning. J. Clin. Exp. Neuropsyc., 12, 323339.
  • Haruno, M., Kuroda, T., Doya, K., Toyama, K., Kimura, M., Samejima, K., Imamizu, H. & Kawato, M. (2004) A neural correlate of reward-based behavioral learning in caudate nucleus: a functional magnetic resonance imaging study of a stochastic decision task. J. Neurosci., 24, 16601665.
  • Heindel, W.C., Butters, N. & Salmon, D.P. (1988) Impaired learning of a motor skill in patients with Huntington's disease. Behav. Neurosci., 102, 141147.
  • Heindel, W.C., Salmon, D.P., Shults, C.W., Walicke, P.A. & Butters, N. (1989) Neuropsychological evidence for multiple implicit memory systems: a comparison of Alzheimer's, Huntington's, and Parkinson's disease patients. J. Neurosci., 9, 582587.
  • Huang, V.S., Haith, A., Mazzoni, P. & Krakauer, J.W. (2011) Rethinking motor learning and savings in adaptation paradigms: model-free memory for successful actions combines with internal models. Neuron, 70, 787801.
  • Ito, M. & Doya, K. (2011) Multiple representations and algorithms for reinforcement learning in the cortico-basal ganglia circuit. Curr. Opin. Neurobiol., 21, 368373.
  • Izawa, J. & Shadmehr, R. (2011) Learning from sensory and reward prediction errors during motor adaptation. PLoS Comput. Biol., 7, e1002012.
  • Jenkins, I.H., Brooks, D.J., Nixon, P.D., Frackowiak, R.S. & Passingham, R.E. (1994) Motor sequence learning: a study with positron emission tomography. J. Neurosci., 14, 37753790.
  • Jueptner, M., Frith, C.D., Brooks, D.J., Frackowiak, R.S. & Passingham, R.E. (1997) Anatomy of motor learning. II. Subcortical structures and learning by trial and error. J. Neurophysiol., 77, 13251337.
  • Krakauer, J.W. (2009) Motor learning and consolidation: the case of visuomotor rotation. Adv. Exp. Med. Biol., 629, 405421.
  • Lambrecq, V., Langbour, N., Guehl, D., Bioulac, B., Burbaud, P. & Rotge, J.Y. (2013) Evolution of brain gray matter loss in Huntington's disease: a meta-analysis. Eur. J. Neurol., 20, 315321.
  • Lillicrap, T.P., Moreno-Briseno, P., Diaz, R., Tweed, D.B., Troje, N. & Fernandez-Ruiz, J. (2013) Adapting to inversion of the visual field: a new twist on an old problem. Exp. Brain Res., doi: 10.1007/s00221-013-3565-6 [Epub ahead of print].
  • Litvan, I., Paulsen, J.S., Mega, M.S. & Cummings, J.L. (1998) Neuropsychiatric assessment of patients with hyperkinetic and hypokinetic movement disorders. Arch. Neurol., 55, 13131319.
  • Marinelli, L., Crupi, D., Di Rocco, A., Bove, M., Eidelberg, D., Abbruzzese, G. & Ghilardi, M.F. (2009) Learning and consolidation of visuo-motor adaptation in Parkinson's disease. Parkinsonism Relat. D., 15, 611.
  • Martin, T.A., Keating, J.G., Goodkin, H.P., Bastian, A.J. & Thach, W.T. (1996a) Throwing while looking through prisms. 2. Specificity and storage of multiple gaze-throw calibrations. Brain, 119, 11991211.
  • Martin, T.A., Keating, J.G., Goodkin, H.P., Bastian, A.J. & Thach, W.T. (1996b) Throwing while looking through prisms. I. Focal olivocerebellar lesions impair adaptation. Brain, 119, 11831198.
  • Moreno-Briseno, P., Diaz, R., Campos-Romo, A. & Fernandez-Ruiz, J. (2010) Sex-related differences in motor learning and performance. Behav. Brain Funct., 6, 74.
  • Murray, E.A., Bussey, T.J. & Wise, S.P. (2000) Role of prefrontal cortex in a network for arbitrary visuomotor mapping. Exp. Brain Res., 133, 114129.
  • Owen, A.M., James, M., Leigh, P.N., Summers, B.A., Marsden, C.D., Quinn, N.P., Lange, K.W. & Robbins, T.W. (1992) Fronto-striatal cognitive deficits at different stages of parkinsons-disease. Brain, 115, 17271751.
  • Packard, M.G. & Knowlton, B.J. (2002) Learning and memory functions of the basal ganglia. Annu. Rev. Neurosci., 25, 563593.
  • Palminteri, S., Lebreton, M., Worbe, Y., Hartmann, A., Lehericy, S., Vidailhet, M., Grabli, D. & Pessiglione, M. (2011) Dopamine-dependent reinforcement of motor skill learning: evidence from Gilles de la Tourette syndrome. Brain, 134, 22872301.
  • Pasupathy, A. & Miller, E.K. (2005) Different time courses of learning-related activity in the prefrontal cortex and striatum. Nature, 433, 873876.
  • Pereira, J.B., Ibarretxe-Bilbao, N., Marti, M.J., Compta, Y., Junque, C., Bargallo, N. & Tolosa, E. (2012) Assessment of cortical degeneration in patients with Parkinson's disease by voxel-based morphometry, cortical folding, and cortical thickness. Hum. Brain Mapp., 33, 25212534.
  • Redgrave, P., Prescott, T.J. & Gurney, K. (1999) The basal ganglia: a vertebrate solution to the selection problem? Neuroscience, 89, 10091023.
  • Rosas, H.D., Hevelone, N.D., Zaleta, A.K., Greve, D.N., Salat, D.H. & Fischl, B. (2005) Regional cortical thinning in preclinical Huntington disease and its relationship to cognition. Neurology, 65, 745747.
  • Schultz, W., Tremblay, L. & Hollerman, J.R. (2000) Reward processing in primate orbitofrontal cortex and basal ganglia. Cereb. Cortex, 10, 272284.
  • Shadmehr, R., Smith, M.A. & Krakauer, J.W. (2010) Error correction, sensory prediction, and adaptation in motor control. Annu. Rev. Neurosci., 33, 89108.
  • Shohamy, D., Myers, C.E., Grossman, S., Sage, J., Gluck, M.A. & Poldrack, R.A. (2004) Cortico-striatal contributions to feedback-based learning: converging data from neuroimaging and neuropsychology. Brain, 127, 851859.
  • Smith, M.A. & Shadmehr, R. (2005) Intact ability to learn internal models of arm dynamics in Huntington's disease but not cerebellar degeneration. J. Neurophysiol., 93, 28092821.
  • Tam, C.W., Burton, E.J., McKeith, I.G., Burn, D.J. & O'Brien, J.T. (2005) Temporal lobe atrophy on MRI in Parkinson disease with dementia: a comparison with Alzheimer disease and dementia with Lewy bodies. Neurology, 64, 861865.
  • Taylor, J.A. & Ivry, R.B. (2012) The role of strategies in motor learning. Ann. NY Acad. Sci., 1251, 112.
  • Toni, I. & Passingham, R.E. (1999) Prefrontal-basal ganglia pathways are involved in the learning of arbitrary visuomotor associations: a PET study. Exp. Brain Res., 127, 1932.
  • Vonsattel, J.P., Myers, R.H., Stevens, T.J., Ferrante, R.J., Bird, E.D. & Richardson, E.P. Jr. (1985) Neuropathological classification of Huntington's disease. J. Neuropath. Exp. Neur., 44, 559577.
  • Weiner, M.J., Hallett, M. & Funkenstein, H.H. (1983) Adaptation to lateral displacement of vision in patients with lesions of the central nervous system. Neurology, 33, 766772.
  • Wolpert, D.M., Diedrichsen, J. & Flanagan, J.R. (2011) Principles of sensorimotor learning. Nat. Rev. Neurosci., 12, 739751.
  • Yin, H.H. & Knowlton, B.J. (2006) The role of the basal ganglia in habit formation. Nat. Rev. Neurosci., 7, 464476.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
ejn12288-sup-0001-TableS1-S2.docxWord document23K

Table S1. Huntington's disease patient information.

Table S2. Parkinson's disease patient 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.