Brain Advance Access originally published online on November 22, 2006
Brain 2007 130(1):256-264; doi:10.1093/brain/awl315
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mixing pro- and antisaccades in patients with parkinsonian syndromes
1 Institut National de la Santé et de la Recherche Médicale Boulevard de l'Hôpital, Paris, France 2 Université Pierre and Marie Curie Boulevard de l'Hôpital, Paris, France 3 Fédération de Neurophysiologie Clinique, Hôpital de la Salpêtrière Boulevard de l'Hôpital, Paris, France
Correspondence to: Bertrand Gaymard, INSERM U679, Hôpital de la Salpêtrière, 47 Boulevard de l'Hôpital, 75651 Paris Cedex 13, France E-mail: gaymard{at}ccr.jussieu.fr
 |
Summary |
|---|
 Top Summary IntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Prosaccades and antisaccades were investigated in three groups of patients with parkinsonian syndromes, Parkinson's disease, corticobasal degeneration (CBD) and progressive supranuclear palsy (PSP), and in a control group. Saccade tasks were performed in single-task blocks (i.e. either blocks of prosaccades or blocks of antisaccades) and in mixed-task blocks (i.e. in blocks of randomly interleaved pro- and antisaccades). Saccade latencies and directional errors (misdirected saccades) were analysed in each subject, and we concentrated more specifically on the comparison of error rates in single tasks and in repeated trials of mixed tasks (i.e. mixing costs). The performance of each group in single tasks was largely consistent with previous studies, with normal antisaccade error rates in Parkinson's disease and CBD patients and increased antisaccade error rates in PSP patients. In contrast, a double dissociation was observed in mixed tasks. Parkinson's disease and CBD patients showed a marked increase in prosaccade and antisaccade error rates in repeated trials of mixed tasks, illustrated by increased mixing costs, whereas PSP patients showed similar error rates in single and repeated trials of mixed tasks, i.e. normal mixing costs. These results demonstrate that: (i) antisaccade performances may be differentially affected in mixed tasks and single tasks; (ii) the region of the dorsolateral prefrontal cortex which is crucial for reflexive saccade inhibition does not seem to be involved in the additional processes required in mixed-task conditions; (iii) the study of interleaved pro- and antisaccades may increase the accuracy of the differential diagnosis between these parkinsonian syndromes.
Key Words: antisaccade; prosaccade; task mixing; corticobasal degeneration; Parkinson's disease; progressive supranuclear palsy
Abbreviations: CBD, corticobasal degeneration; DLPFC, dorsolateral prefrontal cortex; FEF, frontal eye field; PSP, progressive supranuclear palsy
Received April 26, 2006. Revised September 15, 2006. Accepted October 10, 2006.
|
Introduction |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
In daily life, the ability to control reflexive responses such as the orientation of gaze is essential. It may be critical to rapidly look at, and thus identify, a potential danger, but it may also be useful to inhibit a reflexive glance and instead look in another direction. However, the same cue may elicit opposite responses depending on the behavioural context. Optimal performances may thus require the ability to flexibly alternate between activation and suppression of prepotent responses. Oculomotor tasks have been successfully used to study task mixing, in which two opposite stimulus–response rules must be held on-line and selected on a trial-to-trial basis. Prosaccades (saccades towards a visual target) and antisaccades (saccades away from a visual target) may be tested in distinct blocks of trials (single tasks of prosaccades and single tasks of antisaccades), or randomly interleaved within a block of trials (mixed task of pro- and antisaccades). Performance evaluation is based on the analysis of saccade latency and directional error rates, i.e. the percentage of saccades triggered in the direction opposite to that of the instructed direction. Globally, performances are usually slightly worse in mixed tasks than in single tasks (Meiran et al., 2000
), but more detailed comparisons reveal that differences between these two types of task designs are subtended by several processes.
Performances in single tasks may be compared to performances in repeated trials of mixed tasks, i.e. trials preceded by a trial with the same instruction (e.g. two successive prosaccades). Any differences between these two conditions illustrate the increased difficulty of performing a mixed task in which two appropriate stimulus–response rules must be kept on-line and the appropriate rule selected moments before the stimulus appears, and is referred to as the mixing cost. In a mixed pro- and antisaccade task for example, two opposite task sets must be handled simultaneously. Healthy subjects have roughly similar performances in single tasks and in repeated trials of mixed tasks: prosaccade and antisaccade latencies and error rates are slightly higher in repeated trials of mixed tasks than in single tasks, resulting in negligible mixing costs for pro- and antisaccade error rates and latencies (Cherkasova et al., 2002; Fecteau et al., 2004; Reuter et al., 2006).
Within mixed tasks, repeated trials may be compared with switch trials (i.e. trials preceded by a trial with a different instruction, e.g. a prosaccade preceded by an antisaccade). This provides a switch cost that results from at least two components: persistence of the previous instruction, known as task set inertia, which may influence the response to the next trial, and active task set reconfiguration, required on a trial-to-trial basis (Rogers and Monsell, 1995; Meiran et al., 2000; Wylie and Allport, 2000). In healthy subjects, a moderate switch cost has been observed for prosaccade latencies and error rates, and for antisaccade error rates, but a paradoxical switch benefit has been reported for antisaccade latencies (Manoach et al., 2002; Reuter et al. 2006).
The neural substrate of the additional processes required when performing a mixed task rather than a single task is poorly known. Most previous studies have either concentrated on switch costs or combined mixed and switch costs (Sohn et al., 2000; Rushworth et al., 2002; Erickson et al., 2005). In brain imaging studies, comparisons between single tasks and mixed tasks have revealed the activation of a fronto-parietal network and the basal ganglia (Crone et al., 2006). Parietal cortex activations are probably related to increased attentional demand required when two instructions must be held on-line (Bunge et al., 2002; Barber et al., 2005). Within the prefrontal cortex, the main areas activated are the dorsolateral prefrontal cortex (DLPFC) and the dorsomedian prefrontal cortex (DMPFC) (Rushworth et al., 2002; Crone et al., 2006). The DLPFC is thought to be important for working memory (Goldman-Rakic, 1995) and inhibitory processes such as overcoming dominant, prepotent responses (Barber, 2005), or discarding previous task configurations (Sohn et al., 2000). Medial frontal areas such as the supplementary motor area (SMA) and the pre-SMA could be more specifically involved in task switching per se (Rushworth et al., 2001, 2002; Wagner et al., 2004; Crone et al., 2006), allowing rapid task set reconfiguration (Crone et al., 2006). Within the basal ganglia, the striatum has been found to be more active during switched than during repeated trials (Cools et al., 2004; Crone et al., 2006), a result consistent with switching difficulties observed in patients with idiopathic Parkinson's disease and Huntington's disease (Cools et al., 2001; Aron et al., 2003). Very few studies, however, have specifically investigated the neuroanatomical basis of the sustained activities involved in mixing cost. Recently, a dissociation was found in schizophrenic subjects performing oculomotor tasks, consisting of antisaccade deficits but normal mixing and switch costs (Manoach et al., 2002). In an event-related brain imaging study, a particular pattern of areas including the anterior PFC and the cingulate cortex was found to be more activated during mixed tasks than single tasks and to be related to mixing cost (Braver et al., 2003). In a recent electrophysiological study on healthy subjects, different neuronal activities were recorded during repeated trials of a mixed task and during a single task (Goffaux et al., 2006). Accordingly, it seems reasonable to assume that the neural substrates for pro- and antisaccades differ depending on whether or not these tasks are performed in separate or interleaved trials.
We have previously shown that the analysis of pro- and antisaccades may help for the early differential diagnosis of parkinsonian syndromes, namely progressive supranuclear palsy (PSP), corticobasal degeneration (CBD) and idiopathic Parkinson's disease (Vidailhet et al., 1994). Patients affected by these neurodegenerative disorders may exhibit similar clinical symptoms at an early stage, principally parkinsonism (i.e. akinesia and rigidity related to damage of the basal ganglia). However, the involvement of additional structures in CBD and PSP patients may result in specific symptoms that may be crucial for the diagnosis. Widespread cognitive deficits including apraxia, aphasia or alien hand syndrome are typical of CBD and are associated with the involvement of cortical parietal areas (Nagahama et al., 1997). Postural instability, working memory deficits, aphasia and increased distractibility resulting from cortical frontal dysfunction are typical of PSP (Litvan et al., 1999). The different patient groups also frequently exhibit different patterns of oculomotor symptoms. Markedly increased prosaccade latency with a normal antisaccade error rate is frequently encountered in CBD patients. In this disease, the correlation between prosaccade impairments and an apraxia score suggests that increased prosaccade latency results from posterior parietal cortex dysfunction (Vidailhet et al., 1994). In PSP patients, reduced saccade velocity and an increased antisaccade error rate are key symptoms respectively associated with the involvement of brainstem oculomotor structures and the DLPFC (Pierrot-Deseilligny et al., 1989). In contrast, patients with Parkinson's disease show normal prosaccade latency and velocity, and usually have a normal antisaccade error rate (Mosimann et al., 2005). Subtle cognitive impairments similar to frontal lobe symptoms, including task-switching difficulties, may be observed in Parkinson's disease (Cools et al., 2001).
The main goal of this study was to demonstrate that, in these patients with different patterns of cerebral dysfunctions, performances in pro and antisaccades could significantly differ according to the task design, i.e. whether performed in single or in mixed tasks. We mainly concentrated on the comparison between error rates in repeated trials of mixed tasks and in single tasks. Based on current anatomical and physiological data, we hypothesized that predominant DLPFC dysfunction in PSP patients would result in markedly impaired saccade inhibition, but to a similar extent in single tasks and mixed tasks, with normal mixing cost. In contrast, we expected increased mixing costs in CBD and Parkinson's disease patients. Normal DLPFC function would result in normal error rates in single tasks, but fronto-parietal cortex and basal ganglia dysfunction in CBD and fronto-median cortex and basal ganglia dysfunction in Parkinson's disease would result in a decreased ability to handle two instructions on-line, and thus in increased error rates in repeated trials of mixed tasks. If confirmed, this hypothesis would improve our knowledge on the anatomical basis of task mixing and hopefully provide new tools for the differential diagnosis of parkinsonian syndromes. We therefore compared performances of prosaccades and antisaccades, performed either in single tasks or mixed tasks in patients with Parkinson's disease, CBD and PSP.
|
Material and methods |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Subjects
All subjects gave their informed consent to be included in this study, which was approved by the local ethics committee.
Patients
The PSP group consisted of 12 patients (eight male and four female, mean age 66 ± 9.9 years, mean disease duration 2.7 ± 1.6 years) with a mean PSP score [Litvan's criteria (Litvan et al., 1996)] of 8.6 (range 7–10). Although saccade velocity was decreased, all PSP patients exhibited the full range of horizontal eye movements. The CBD group consisted of eight patients (four female and four male, mean age 76 ± 5.4 years, mean disease duration 2.3 ± 1.4 years) with a mean CBD score [Litvan's criteria (Litvan et al., 1997)] of 7.0 (range: 5–8). Six patients had right ideomotor apraxia and two patients had left ideomotor apraxia. Two patients in the PSP group and one in the CBD group were receiving L-dopa medication. The Parkinson's disease group consisted of 15 patients (11 male and four female, mean age 64.3 ± 8.8 years, mean disease duration 6.6 ± 3.4 years). Motor disability was assessed by the motor subscale of the Unified Parkinson's Disease Rating Scale (UPDRS) (mean score 15.2). All Parkinson's disease patients were receiving L-dopa medication (anticholinergic medication: 3 patients; dopamine agonist: 5 patients) and were tested in the ON condition.
Control group
Ten healthy volunteers (mean age 64 ± 9 years) were studied as a control group. These subjects had no history of neurological or psychiatric disorders and were free of any medication.
Oculomotor tests
Eye movements were recorded by horizontal electro-oculography with our standardized protocol as described previously (sampling frequency: 200 Hz, bandwidth 0–100 Hz) (Ploner et al., 2005). Visual cues were green and red LEDs subtending a visual angle of 0.18° and with a luminance of 5 cd/m2. They were displayed on a curved ramp located 95 cm in front of the subject, seated in total darkness with the head immobilized. The following three paradigms were performed in all subjects and in the same chronological sequence: single tasks of prosaccades, single tasks of antisaccades and mixed tasks of pro- and antisaccades. In our experience with elderly patients, the best way of ensuring a good level of understanding is to start with the easiest task and progressively add more complex instructions. Clinical studies with elderly patients must also avoid non-specific impairments due to excessive fatigue. For this reason, breaks were regularly interleaved throughout each recording session, and the total duration of a recording session did not exceed 40 min.
Single task of prosaccades
Subjects were instructed to look at a green central fixation point (3500–4200 ms), then to look as quickly as possible at a 25° lateral red target (1000 ms) that appeared randomly on the right or on the left side. A 200 ms gap was interposed between central target offset and lateral target onset. Thirty-six prosaccades were recorded. Mean prosaccade latency was calculated by averaging prosaccade latencies in each direction.
Single task of antisaccades
Subjects were instructed to look at a red central fixation point (3500–4200 ms), then to trigger a saccade as soon as possible in the opposite direction to a randomly right or left 25° lateral red target that appeared 200 ms after fixation point offset. An initial training block of trials was performed to ensure that the instructions were understood. The presence of correct antisaccades and/or of corrective saccades was taken as evidence that the instructions had been correctly understood. Thirty-six antisaccades were recorded. The antisaccade error rate, defined as the percentage of directional errors (i.e. saccades triggered towards the lateral target), was determined in each subject.
Mixed task of pro- and antisaccades
In this task, the central fixation point initially consisted of two vertically aligned and contiguous red and green LEDs. After a fixation time of 3500–4200 ms, one of these two LEDs was turned off. The remaining LED stayed on during 500 ms, and subjects were instructed that its colour was the code for programming the appropriate response to the lateral target: a green LED required a prosaccade and a red LED an antisaccade. A 200 ms gap was used as in the previous tasks. The cue–target interval lasted 700 ms and the interval between two targets was between 5 and 6 s. We checked verbally that the instruction had been correctly understood, and a training task consisting of one block of mixed trials was then performed. The triggering of correct antisaccades or the presence of corrective saccades was taken as evidence that the instructions had been correctly understood. Forty-eight prosaccades and antisaccades were recorded. We determined the latency of correct prosaccades and correct antisaccades and prosaccade and antisaccade error rates, as in single tasks.
Analysis
Saccades with a latency below 80 ms or above 1000 ms, and/or an amplitude below 1° were rejected, but this represented <1% of all trials. Mean latency was determined only for correct antisaccades but was not calculated in subjects with error rates above 85% (i.e. in five subjects). Directional errors were defined as saccades initially directed towards the hemifield away from the target following a prosaccade instruction, or towards the target following an antisaccade instruction.
In each subject, we calculated mean pro- and antisaccade latencies and error rates in single tasks. A similar analysis was done globally in mixed tasks, i.e. for all pro- and antisaccade trials, then we selectively analysed repeated prosaccade and repeated antisaccade trials in order to provide a mixing cost for latencies and error rates, defined as performance in repeated trials in mixed tasks minus performance in single tasks. There are two possible methods for analysing repeated trials. One method is to select all trials that are preceded by an N-1 trial with a similar instruction. A second, more restrictive method is to take into account only correctly executed N-1 trials with the same instruction. In subjects with low error rates, these two methods are likely to provide similar results (Cherkasova et al., 2002; Reuter et al., 2006). In this study, however, using post hoc analysis, we found that antisaccade error rates were different in CBD patients whether or not only correct responses in the N-1 trial were taken into account. We therefore chose to select within repeated trials only those in which the N-1 response was correct. According to this method, correct repeated trials, i.e. two successive correct antisaccades, are required for the calculation of mixing cost for antisaccade latency. Since this occurred rarely in the PSP group, the mixing cost for antisaccade latency was not analysed.
We first compared saccade latencies, error rates and mixing costs across all four groups using a Kruskal–Wallis test. Significant results were further analysed by multiple comparisons using a Mann–Whitney test. The significance level was 5%.
Within each group, we made the following comparisons using a Wilcoxon test: prosaccade latencies versus antisaccade latencies in single tasks and in mixed tasks; prosaccade and antisaccade latencies in single tasks versus prosaccade and antisaccade latencies in mixed tasks; antisaccade error rates in mixed tasks versus prosaccade error rates in mixed tasks; antisaccade error rates in single tasks versus antisaccade error rates in mixed tasks; mixing costs for antisaccade error rates versus mixing costs for prosaccade error rates.
|
Results |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Single tasks
Pro- and antisaccade latencies
The comparison with the control group showed an increased prosaccade latency in the CBD group only (P = 0.0001) (Fig. 1A). Antisaccade latency was longer than prosaccade latency in all groups (control group: P = 0.005; Parkinson's disease: P = 0.001; CBD: P = 0.012; PSP: P = 0.008) and, compared with controls, antisaccade latency was increased in both the CBD group (P = 0.001) and the PSP group (P = 0.009) (Fig. 1B). However, antisaccade latency was significantly longer in CBD than in PSP patients (P = 0.011). The comparison between the CBD and the Parkinson's disease groups showed significantly increased prosaccade (P = 0.001) and antisaccade (P = 0.0001) latencies in the CBD group. The comparison between the PSP and the Parkinson's disease groups did not provide significant differences for pro- and antisaccade latencies.
|
P < 0.05. Error bars indicate the SEM. C = controls; CBD = corticobasal degeneration; PD = Parkinson's disease; PSP = progressive supranuclear palsy.Pro- and antisaccade error rates
No prosaccade directional errors were observed in any subject. Compared with controls, the antisaccade error rate was significantly increased in the PSP group only (P = 0.0001) (Fig. 2B).
|
Mixed tasks: global results
Pro- and antisaccade latencies
The comparison with the control group (Fig. 1A and B) showed increased prosaccade latency in the CBD group only (P = 0.006), and increased antisaccade latency in the CBD group (P = 0.0001) and in the PSP group (P = 0.013). However, antisaccade latency was longer in the CBD group than in the PSP group (P < 0.05) and the Parkinson's disease group (P = 0.001). Antisaccade latency was longer than prosaccade latency in all groups (controls: P = 0.007; CBD: P = 0.012; Parkinson's disease: P = 0.002; PSP: P < 0.05).
Pro- and antisaccade error rates
In mixed trials, the comparison with controls showed a higher prosaccade error rate in the CBD group (P = 0.0001) and the Parkinson's disease group (P = 0.005), but not in the PSP group (P = 0.582) (Fig. 2A), and higher antisaccade error rates in all patient groups (CBD: P = 0.001; Parkinson's disease: P = 0.001; PSP: P = 0.0001) (Fig. 2B). The comparison between the CBD group and the Parkinson's disease group showed a higher error rate in the CBD group for prosaccades (P = 0.002) but not for antisaccades (P = 0.325). The error rate was higher for antisaccades than for prosaccades in the control group (P = 0.007), the Parkinson's disease group (P = 0.001) and the PSP group (P = 0.002), but not in the CBD group (P = 0.07).
Mixed tasks: repeated trials
Mixing cost for latency
Mixing cost for prosaccade latency was not different across all groups (Kruskal–Wallis: P = 0.635). High antisaccade error rates were reached in some patients, resulting in a small number of correct antisaccade trials. For this reason, mixing cost for antisaccade latency was not analysed.
Mixing cost for error rates
The mixing cost for prosaccade error rate was significantly different across all groups (Kruskal–Wallis: P = 0.003). Comparison with the control group revealed significant differences in the Parkinson's disease (P < 0.05) and CBD (P = 0.01) groups, but not in the PSP group (P = 0.176) (Fig. 3). Comparisons between patient groups revealed that it was significantly higher in the CBD group than in the Parkinson's disease (P = 0.02) and PSP (P = 0.005) groups, but not different in the PSP and Parkinson's disease groups (P = 0.495).
|
The mixing cost for antisaccade error rates was significantly different across all groups (Kruskal–Wallis: P = 0.009). Comparison with the control group revealed significant differences in the Parkinson's disease (P = 0.01) and CBD (P = 0.01) groups, but not in the PSP group (P = 0.428). Comparisons between patient groups revealed that it was not significantly different in CBD and Parkinson's disease patients (P = 0.821), but it was significantly increased in CBD compared with PSP patients (P = 0.031), and in Parkinson's disease compared with PSP patients (P = 0.015). Mixing cost for error rate was not different between pro- and antisaccades in any group except in the Parkinson's disease group (P < 0.05).
For the sake of clarity, we now present a summary of the main results of each group (Table 1).
|
In controls, mixed tasks resulted in increased prosaccade latencies but identical antisaccade latencies, few prosaccade directional errors and no increase in the antisaccade error rate. The mixing cost for error rates was similar for prosaccades and for antisaccades.
In the Parkinson's disease group, performances in single tasks were similar to those of controls for both latencies and error rates. In the mixed tasks, error rates increased significantly compared with controls for prosaccades and for antisaccades, with a greater increase for antisaccades. The mixing cost for error rates was increased compared with controls for both pro- and antisaccades, and was higher for antisaccades than for prosaccades.
In the CBD group, pro- and antisaccade latencies were markedly increased in both single tasks and mixed tasks. The antisaccade error rate was normal in single tasks, but both pro- and antisaccade error rates increased significantly compared with controls in mixed tasks. The mixing cost for error rates was higher than that of controls for both pro- and antisaccades, but was not different for prosaccades versus antisaccades. Comparison between the CBD and Parkinson's disease groups showed that the mixing cost for error rates was lower in the Parkinson's disease group for prosaccades but was similar for antisaccades.
In the PSP group, the antisaccade error rate was markedly increased compared with controls in the single task, but error rates did not worsen in the mixed tasks. The mixing cost for error rates did not differ from that of controls for either pro- or antisaccades, and was identical for pro- and antisaccades.
|
Discussion |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
This study investigates the oculomotor performances of patients with parkinsonian syndromes in single tasks of prosaccades and single tasks of antisaccades and, for the first time, in blocks of randomly interleaved pro- and antisaccades (mixed task). Our main focus was to determine the influence of the task design on error rates and therefore compare performances in repeated trials of mixed tasks to performances in single tasks. The main finding is that the task design had a marked influence on error rates in Parkinson's disease and CBD patients but not in PSP patients or control subjects. Patients with Parkinson's disease and CBD had normal error rates in single tasks but significantly higher error rates in repeated trials of mixed tasks, whereas PSP patients showed similar results in both tasks. We will discuss first our patients' results in single tasks, then global performances in the mixed condition, then compare these two tasks by an analysis of mixing costs.
Single tasks
Our patients' results in single tasks confirm and extend previous findings. Oculomotor studies of Parkinson's disease have consistently reported normal reflexive saccade latency (Rascol et al., 1989; Vidailhet et al., 1994) and either normal (Lueck et al., 1990; Fukushima et al., 1994; Vidailhet et al., 1994; Rivaud-Péchoux et al., 2000; Kingstone et al., 2002; Mosimann et al., 2005), or slightly increased (Chan et al., 2005) antisaccade error rates. As already reported (Vidailhet et al., 1994), our patients with CBD had increased prosaccade latencies and normal antisaccade error rates, but we show here that antisaccade latencies were also markedly increased. Increased saccade latency has mainly been reported in patients with dysfunction of the parietal eye field (PEF), the frontal eye field (FEF) or the superior colliculus (for a review, see Gaymard et al., 1998). Whereas collicular dysfunction impairs all saccade types, PEF and FEF dysfunctions have more specific effects on reflexive and voluntary saccades, respectively (Leigh and Kennard, 2004). In a recent study on CBD patients, regional atrophy was observed in both PEF and FEF regions but not in the upper midbrain, which contains the superior colliculus (Boxer et al., 2006). The correlation we previously found in CBD patients between increased prosaccade latency and an apraxia score supports the hypothesis that increased prosaccade latency is associated with parietal cortex dysfunction (Vidailhet et al., 1994). Although a common mechanism could result in the increase of both pro- and antisaccade latencies, we propose that increased antisaccade latency in CBD may result from FEF dysfunction, since a similar oculomotor pattern, namely increased antisaccade latency and a normal antisaccade error rate, has been reported after focal lesions of the FEF (Rivaud et al., 1994; Gaymard et al., 1999).
In our PSP patients, normal prosaccade latencies and markedly increased antisaccade error rates are consistent with previous reports (Pierrot-Deseilligny et al., 1989; Vidailhet et al., 1994). Although various impairments of prosaccade latency may be encountered in this disease, likely depending on the degree of involvement of the superior colliculus in the degenerative process, a high antisaccade error rate is a key feature for the diagnosis of PSP and is related to DLPFC dysfunction (Pierrot-Deseilligny et al., 1989, 1991). In addition, we found an increased latency of correct antisaccades, although any comparison with controls must be treated with caution since both groups had markedly different antisaccade error rates. Increased difficulty in reflexive saccade inhibition may result in additional processing time being required in order to trigger a correct antisaccade. Indeed, in primates, DLPFC inactivation, which impairs reflexive saccade inhibition, increases correct antisaccade latencies even in the absence of FEF dysfunction (Condy et al., 2006).
Mixed tasks
Global results
Single tasks and mixed tasks were performed in the same order in all our subjects, i.e. from the easiest to the most complex task, to ensure a better understanding of the instructions. Therefore, any within-group comparison between tasks should be analysed cautiously, since non-specific factors such as training or fatigue may have influenced performance. However, comparison between groups showed clear differences. The most striking result provided by a global analysis of mixed tasks comes from patients in the Parkinson's disease and CBD groups. In these patients, antisaccade error rates were normal in the single antisaccade task but markedly increased in the mixed task (Fig. 2). This dissociation demonstrates that, compared with single tasks, performing antisaccades in a mixed-task design recruits additional cerebral structures. This result therefore emphasizes that the task design is of major importance for the study of the neural substrate of antisaccades and may shed light on conflicting results regarding the role of cerebral areas in antisaccade generation. Although it seems clear that the DLPFC is associated with prosaccade inhibition (Ploner et al., 2005; Condy et al., 2006), several other frontal and parietal areas whose roles are not yet clearly determined have been found to be activated during antisaccade tasks (Munoz and Everling, 2004). An involvement of the supplementary eye field (SEF) in reflexive saccade inhibition has been proposed on the basis that this area has been reported to be activated during the performance of correct antisaccades, although this was observed in a mixed task of interleaved pro- and antisaccades (Schlag et al., 1997). Interestingly, no deficit in reflexive saccade inhibition was reported in patients with SEF lesions performing single tasks of antisaccades (Gaymard et al., 1990; Husain et al., 2003). The hypothesis that SEF activation at least partially reflects task shifting should be further examined.
Mixing costs
Mixing costs can be used to evaluate the ability to perform a task in which two task sets instead of one must be handled simultaneously, without taking into account the need to switch between different stimulus–response rules. Hence, performing mixed tasks results in an increased cognitive load, especially increased demands on working memory, vigilance, sustained attention, motivation and response selection (Rogers and Monsell, 1995; Los, 1996).
Mixing costs for pro- and antisaccade error rates were low in our control group, in agreement with previous studies (Cherkasova et al., 2002; Reuter et al., 2006). A similar low mixing cost for error rates was observed in PSP patients: saccade inhibition was severely impaired in single tasks but performances did not worsen in repeated trials of mixed tasks. Although high error rates were reached in single tasks, most patients in this group were not restricted by a ceiling effect, since a 100% error rate was observed in only two of 12 patients. Impaired inhibition of reflexive saccades in PSP patients is associated with the involvement of the DLPFC in the degenerative process (Pierrot-Deseilligny et al., 1989). Human and non-human primate studies have been able to demonstrate that a circumscribed region of Brodmann area 46 is crucial for reflexive saccade inhibition (Ploner et al., 2005; Condy et al., 2006). It may thus be inferred from this study that this critical DLPFC subregion is not essential for task mixing, in accordance with a previous study performed on schizophrenic subjects (Manoach et al., 2002).
The opposite pattern was observed in Parkinson's disease and CBD patients. Although both CBD and Parkinson's disease patients performed worse in repeated trials of mixed tasks than in single tasks, they were not equally affected. CBD patients had similar error rates in repeated trials of pro- and antisaccades, whereas Parkinson's disease patients made significantly fewer errors on prosaccade trials. Reuter et al. (2006) recently demonstrated that, in addition to task switching (switching between pro- and antisaccades), response switching (e.g. switching between a rightward and a leftward saccade) also increases error rates but only in the case of two successive antisaccades, without affecting prosaccades. These authors have proposed that the selectivity of response switching on antisaccades may result from the different mechanisms that lead to the generation of prosaccades and antisaccades. Whereas prosaccade triggering would result from a direct visuomotor transformation, antisaccade triggering would require the active selection of an appropriate motor programme that encodes saccade direction. Thus, following an antisaccade, persistence of a motor programme would be more likely to affect a subsequent motor programme, i.e. the triggering of an antisaccade rather than a prosaccade. These hypotheses most likely apply to our repeated tasks, in which right or left saccades were randomly required. Normal visuomotor transformation in patients with Parkinson's disease, assessed by unaffected prosaccade triggering, would result in lower prosaccade than antisaccade error rates in repeated trials. In contrast, we suggest that dysfunction in an oculomotor fronto-parietal network in CBD patients would impair the normal process of direct visuomotor transformation required for prosaccade triggering, as suggested by the markedly increased prosaccade latencies. In these patients, prosaccade and antisaccade triggering would thus both require the active selection of a specific motor programme, and would consequently be equally affected by the motor switching required in our repeated tasks.
This study, performed in patients with degenerative diseases, does not enable us to define precisely the neural structures involved in oculomotor mixing tasks. However, it is noticeable that the greatest deficits were observed in patients with CBD, a disease in which the degenerative process involves the parietal and the frontal cortices and the basal ganglia. In Parkinson's disease, the most affected structures are the basal ganglia, although a slight involvement of the DMPFC (Cunnington et al., 2001) may participate in mixing difficulties. The distribution of these neural structures is in accordance with the few recent studies that have investigated the pattern of activation in subjects performing non-switch trials in mixed tasks compared with single tasks (Braver et al., 2003; Erickson et al., 2005). However, further studies on patients with focal cerebral lesions are needed in order to investigate this issue.
Provided our results are confirmed at earlier stages of disease, the different patterns of impairments observed in mixed tasks could have clinical implications for the diagnosis of these parkinsonian syndromes, especially for CBD and PSP patients, who may present overlapping clinical features. We found a particular contrast in the performance of mixed trials between PSP and CBD patients, for both pro- and antisaccade responses, resulting in markedly different mixing costs for error rates (Table 1). The most striking difference concerned directional errors on prosaccade instruction (i.e. triggering of an erroneous antisaccade), which occurred far less frequently in PSP patients than in CBD patients (PSP: mean 4%, SEM: 4; CBD: mean 33%, SEM 22; P < 0.0001). More generally, our results suggest that the use of paradigms with randomly interleaved pro- and antisaccade instructions could increase the accuracy of the diagnosis of parkinsonian syndromes (Table 1).
|
References |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Aron AR, Watkins L, Sahakian BJ, Monsell S, Barker RA, Robbins TW. (2003) Task-set switching deficits in early-stage Huntington's disease: implications for basal ganglia function. J Cogn Neurosci 15:629–42.
Barber AD and Carter CS. (2005) Cognitive control involved in overcoming prepotent response tendencies and switching between tasks. Cereb Cortex 15:899–912.
Boxer AL, Geschwind MD, Belfor N, Gorno-Tempini ML, Schauer GF, Miller BL, et al. (2006) Patterns of brain atrophy that differentiate corticobasal degeneration syndrome from progressive supranuclear palsy. Arch Neurol 63:81–6.
Braver TS, Reynolds JR, Donaldson DI. (2003) Neural mechanisms of transient and sustained cognitive control during task switching. Neuron 39:713–26.[CrossRef][ISI][Medline]
Bunge SA, Hazeltine E, Scanlon MD, Rosen AC, Gabrieli JD. (2002) Dissociable contributions of prefrontal and parietal cortices to response selection. Neuroimage 17:1562–71.[CrossRef][ISI][Medline]
Chan F, Armstrong IT, Pari G, Riopelle RJ, Munoz DP. (2005) Deficits in saccadic eye-movement control in Parkinson's disease. Neuropsychologia 43:784–96.[CrossRef][ISI][Medline]
Cherkasova MV, Manoach DS, Intriligator JM, Barton JJ. (2002) Antisaccades and task-switching: interactions in controlled processing. Exp Brain Res 144:528–37.[CrossRef][ISI][Medline]
Condy C, Wattiez N, Rivaud-Pechoux S, Tremblay L, Gaymard B. (2006) Antisaccade deficit after inactivation of the principal sulcus in monkeys. Cereb Cortex In press.
Cools R, Barker RA, Sahakian BJ, Robbins TW. (2001) Mechanisms of cognitive set flexibility in Parkinson's disease. Brain 124:2503–12.
Cools R, Clark L, Robbins TW. (2004) Differential responses in human striatum and prefrontal cortex to changes in object and rule relevance. J Neurosci 24:1129–35.
Crone EA, Wendelken C, Donohue SE, Bunge SA. (2006) Neural evidence for dissociable components of task-switching. Cereb Cortex 16:475–86.
Cunnington R, Egan GF, O'Sullivan JD, Hughes AJ, Bradshaw JL, Colebatch JG. (2001) Motor imagery in Parkinson's disease: a PET study. Mov Disord 16:849–57.[CrossRef][ISI][Medline]
Erickson KI, Colcombe SJ, Wadhwa R, Bherer L, Peterson MS, Scalf PE, et al. (2005) Neural correlates of dual-task performance after minimizing task-preparation. Neuroimage 28:967–79.[CrossRef][ISI][Medline]
Fecteau JH, Au C, Armstrong IT, Munoz DP. (2004) Sensory biases produce alternation advantage found in sequential saccadic eye movement tasks. Exp Brain Res 159:84–91.[ISI][Medline]
Fukushima J, Fukushima K, Miyasaka K, Yamashita I. (1994) Voluntary control of saccadic eye movement in patients with frontal cortical lesions and parkinsonian patients in comparison with that in schizophrenics. Biol Psychiatry 36:21–30.[CrossRef][ISI][Medline]
Gaymard B, Pierrot-Deseilligny C, Rivaud S. (1990) Impairment of sequences of memory-guided saccades after supplementary motor area lesions. Ann Neurol 28:622–6.[CrossRef][ISI][Medline]
Gaymard B, Ploner CJ, Rivaud S, Vermersch AI, Pierrot-Deseilligny C. (1998) Cortical control of saccades. Exp Brain Res 123:159–63.[CrossRef][ISI][Medline]
Gaymard B, Ploner CJ, Rivaud-Péchoux S, Pierrot-Deseilligny C. (1999) The frontal eye field is involved in spatial short-term memory but not in reflexive saccade inhibition. Exp Brain Res 129:288–301.[CrossRef][ISI][Medline]
Goldman-Rakic PS. (1995) Cellular basis of working memory. Neuron 14:477–85.[CrossRef][ISI][Medline]
Goffaux P, Phillips NA, Sinai M, Pushkar D. (2006) Behavioural and electrophysiological measures of task switching during single and mixed-task conditions. Biol Psychol 72:278–90.[CrossRef][ISI][Medline]
Husain M, Parton A, Hodgson TL, Mort D, Rees G. (2003) Self-control during response conflict by human supplementary eye field. Nat Neurosci 6:117–8.[CrossRef][ISI][Medline]
Kingstone A, Klein R, Morein-Zamir S, Hunt A, Fisk J, Maxner C. (2002) Orienting attention in aging and Parkinson's disease: distinguishing modes of control. J Clin Exp Neuropsychol 24:951–67.[ISI][Medline]
Leigh RJ and Kennard C. (2004) Using saccades as a research tool in the clinical neurosciences. Brain 127:460–77.
Litvan I, Agid Y, Calne D, Campbell G, Dubois B, Duvoisin RC, et al. (1996) Clinical research criteria for the diagnosis of progressive supranuclear palsy (Steele-Richardson-Olszewski syndrome): report of the NINDS-SPSP international workshop. Neurology 47:1–9.
Litvan I, Agid Y, Goetz C, Jankovic J, Wenning GK, Brandel JP, et al. (1997) Accuracy of the clinical diagnosis of corticobasal degeneration: a clinicopathologic study. Neurology 48:119–25.
Litvan I, Grimes DA, Lang AE, Jankovic J, McKee A, Verny M, et al. (1999) Clinical features differentiating patients with postmortem confirmed progressive supranuclear palsy and corticobasal degeneration. J Neurol 246:1–5.[CrossRef]
Los SA. (1996) On the origin of mixing costs: exploring information processing in pure and mixed blocks of trials. Acta Psychologica 94:145–88.
Lueck CJ, Tanyeri S, Crawford TJ, Henderson L, Kennard C. (1990) Antisaccades and remembered saccades in Parkinson's disease. J Neurol Neurosurg Psychiatry 53:284–8.[Abstract]
Manoach DS, Lindgren KA, Cherkasova MV, Goff DC, Halpern EF, Intriligator J, et al. (2002) Schizophrenic subjects show deficient inhibition but intact task switching on saccadic tasks. Biol Psychiatry 51:816–26.[CrossRef][ISI][Medline]
Meiran N, Chorev Z, Sapir A. (2000) Component processes in task switching. Cognit Psychol 41:211–53.[CrossRef][ISI][Medline]
Mosimann UP, Muri RM, Burn DJ, Felblinger J, O'Brien JT, McKeith IG. (2005) Saccadic eye movement changes in Parkinson's disease dementia and dementia with Lewy bodies. Brain 128:1267–76.
Munoz DP and Everling S. (2004) Look away: the anti-saccade task and the voluntary control of eye movement. Nat Rev Neurosci 5:218–28.[CrossRef][ISI][Medline]
Nagahama Y, Fukuyama H, Turjanski N, Kennedy A, Yamauchi H, Ouchi Y, et al. (1997) Cerebral glucose metabolism in corticobasal degeneration: comparison with progressive supranuclear palsy and normal controls. Mov Disord 12:691–6.[CrossRef][ISI][Medline]
Pierrot-Deseilligny C, Rivaud S, Pillon B, Fournier E, Agid Y. (1989) Lateral visually-guided saccades in progressive supranuclear palsy. Brain 112:471–87.
Pierrot-Deseilligny C, Rivaud S, Gaymard B, Agid Y. (1991) Cortical control of reflexive visually-guided saccades. Brain 114:1473–85.
Ploner CJ, Gaymard BM, Rivaud-Péchoux S, Pierrot-Deseilligny C. (2005) The prefrontal substrate of reflexive saccade inhibition in humans. Biol Psychiatry 57:1159–65.[CrossRef][ISI][Medline]
Rascol O, Clanet M, Montastruc JL, Simonetta M, Soulier-Esteve MJ, Doyon B, et al. (1989) Abnormal ocular movements in Parkinson's disease. Evidence for involvement of dopaminergic systems. Brain 112:1193–214.
Reuter B, Philipp AM, Koch I, Kathmann N. (2006) Effects of switching between leftward and rightward pro- and antisaccades. Biol Psychol 72:88–95.[CrossRef][ISI][Medline]
Rivaud S, Muri RM, Gaymard B, Vermersch AI, Pierrot-Deseilligny C. (1994) Eye movement disorders after frontal eye field lesions in humans. Exp Brain Res 102:110–20.[ISI][Medline]
Rivaud-Péchoux S, Vermersch AI, Gaymard B, Ploner CJ, Bejjani BP, Damier P, et al. (2000) Improvement of memory guided saccades in parkinsonian patients by high frequency subthalamic nucleus stimulation. J Neurol Neurosurg Psychiatry 68:381–4.
Rogers RD and Monsell S. (1995) Costs of a predictable switch between simple cognitive tasks. J Exp Psychol 124:207–31.
Rushworth MF, Paus T, Sipila PK. (2001) Attention systems and the organization of the human parietal cortex. J Neurosci 21:5262–71.
Rushworth MF, Hadland KA, Paus T, Sipila PK. (2002) Role of the human medial frontal cortex in task switching: a combined fMRI and TMS study. J Neurophysiol 87:2577–92.
Schlag-Rey M, Amador N, Sanchez H, Schlag J. (1997) Antisaccade performance predicted by neuronal activity in the supplementary eye field. Nature 390:398–401.[CrossRef][Medline]
Sohn MH, Ursu S, Anderson JR, Stenger VA, Carter CS. (2000) The role of prefrontal cortex and posterior parietal cortex in task switching. Proc Natl Acad Sci USA 97:13448–53.
Vidailhet M, Rivaud S, Gouider-Khouja N, Pillon B, Bonnet AM, Gaymard B, et al. (1994) Eye movements in parkinsonian syndromes. Ann Neurol 35:420–6.[CrossRef][ISI][Medline]
Wagner AD, Bunge SA, Badre D. (2004) Cognitive control semantic memory and priming: contributions of prefrontal cortex. In Gazzaniga MS (Ed.). The new cognitive neuroscience(Massachusetts Institute of Technology Press, Cambridge, MA) pp. 709–25.
Wylie G and Allport A. (2000) Task switching and the measurement of "switch costs". Psychol Res 63:212–33.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Yoshida and M. Tanaka Enhanced Modulation of Neuronal Activity during Antisaccades in the Primate Globus Pallidus Cereb Cortex, May 13, 2008; (2008) bhn069v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Garbutt, A. Matlin, J. Hellmuth, A. K. Schenk, J. K. Johnson, H. Rosen, D. Dean, J. Kramer, J. Neuhaus, B. L. Miller, et al. Oculomotor function in frontotemporal lobar degeneration, related disorders and Alzheimer's disease Brain, May 1, 2008; 131(5): 1268 - 1281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Hubner, A. Sprenger, C. Klein, J. Hagenah, H. Rambold, C. Zuhlke, D. Kompf, A. Rolfs, H. Kimmig, and C. Helmchen Eye movement abnormalities in spinocerebellar ataxia type 17 (SCA17) Neurology, September 11, 2007; 69(11): 1160 - 1168. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||