Research ReportNo sequence dependent modulation of the Simon effect in Parkinson's disease
Introduction
Patients with basal ganglia (BG) dysfunction invariably exhibit impaired control over a range of motor and non-motor behaviours, with deficits primarily revealed where performance relies on internal or voluntary control processes [9], [28], [29]. With Parkinson's disease (PD), profound difficulty initiating and facilitating wanted behaviour has conventionally been attributed to impaired BG inhibitory output and subsequent underactivation along thalamo-cortical circuitry [45], a consequence of the deterioration of the dopaminergic nigrostriatal system of the BG [41], [43].
More recent models of BG function propose that this group of subcortical nuclei focus neural resources via context-dependent facilitation/amplification or inhibition of neural activity [40], [44], influencing the selection and suppression of competing inputs, enabling preferential processing of relevant inputs for response selection [47], [61]. Thus, the resolution of response conflict is assisted by the active inhibition of inappropriate inputs/responses, with disease universally compromising a wide variety of motor, cognitive and emotional functions [18], [71], [72].
Not surprisingly, there is also growing evidence of impairment in the control of more automatic processes in PD, with ill-focused cortical activity susceptible to interference by automatic response tendencies. Deficient inhibitory control has been proposed by a number of authors [26], [56], [61], with reports of patients more rapidly disengaging attention [7], [22], [23], [55], [74], experiencing difficulty in inhibiting a subliminally primed response [61], and demonstrating impaired response inhibition in stop-signal reaction time tasks [26]. Oculomotor behaviour is similarly compromised, consistent with the interconnectedness of BG–thalamocortical circuitry [34], [44], with more erroneous responses to cue stimuli in antisaccade paradigms [2], [8], [19], and reduced latencies for reflexive saccade tasks [36], [59]. Our own investigations with this patient group have revealed both impaired response suppression to irrelevant visual stimuli and hyper-reflexive orienting (manuscript submitted for publication 2005).
Inhibitory modulation of automatic response tendencies is a key concept in various accounts of stimulus–response (S–R) effects, including the Simon effect (SE) [63]. If the relative spatial positions of stimulus and response correspond, participants typically respond faster than when they do not, even though stimulus location is defined as irrelevant for the task. A spatial code is seemingly derived from an irrelevant target, either facilitating or interfering with subsequent processing. Cognitive dual route models typically propose two parallel routes of response selection [21], [25], [38], [76], [77], with a stimulus-driven or direct route whereby response is activated by spatially corresponding target position and a slower, indirect route which selects responses on the basis of relevant stimulus features (e.g. colour or form). With corresponding S–R positions, convergence of both routes is thought to facilitate response decision, and with non-correspondence, it is assumed that responses conflict leads to prolonged reaction times.
It is increasingly clear, however, that stimulus-position-triggered activation is not unconditionally relayed by the direct route, but is dependent upon the intentions of the performer. Task instructions, context of responses and set effects all have systematic effects on the SE, with response activation seemingly contingent upon working memory representations of S–R mappings [1]. Thus, response conflict is subject to contextual modulations. In the SE task, response times are dependent upon the immediately preceding event, the SE typically large when stimulus and response are spatially compatible in the preceding trial, and reduced or absent when the preceding trial features S–R non-correspondence or spatial incompatibility [57], [64], [65], [68].
Various accounts of trial-by-trial control of the direct route activation are proposed, and include Sturmer et al.'s ancillary monitoring mechanism. This mechanism purportedly registers the correspondence or non-correspondence between a correct response and that activated by stimulus position, with conflicts in response preparation thought to be resolved by preventing stimulus position from activating a response (selectively suppressing direct response-activation) following an S–R incompatible trial, thus, a SE is only found following corresponding trials [64]. Conversely, Praamstra and Plat have proposed that the direct response-activation route may be normally inhibited, but is temporarily released from inhibition after a trial with spatially compatible S–R locations.
An alternative explanation of this effect is proposed by feature integration theory [31], which retains the idea of parallel voluntary and automatic response routes, without the requirement of trial-to-trial modulation of the contribution of the direct response-activated route. Instead, co-occurring stimulus and response features are thought to be integrated into transient memory episodes, referred to as event files. Where perceptual input or response demands overlap in a subsequent trial, activation of either item of the previous event file is thought to co-activate other features of the event file. If these features mismatch current stimulus conditions or response demands, coding problems arise. The feature code must be ‘unbound’ from the preceding event file and integrated into the next. Therefore, response times reflect not only the positive or negative effects of spatial correspondence, but the unbinding effects of non-corresponding event files.
Whether conceptualised as working memory representations of appropriate stimulus response associations, or gating of automatic response routes according to some predetermined criteria, the control of S–R translation is likely to incorporate both (pre-) frontal regions, including the cingulate gyrus and dorsolateral pre-frontal cortex, and the posterior parietal cortex (PPC). A context-dependent feedback signal may be send backwards to the PPC from more anterior structures to inhibit automatic capture of attention by visual events, and subsequent motor activation.
However, the focusing of often coarsely coded cortical representations (i.e. memories, intentions, sensory features) into behaviourally significant responses by cortical and brainstem motor regions, is also dependent upon regulation via BG–thalamocortical circuitry, the selection and suppression of competing responses a final gating mechanism in the resolution of conflict. Although investigations that have adopted choice reaction-time paradigms with PD patients have reported normal compatibility effects [10], [11], [14], Praamstra and Plat report increased sensitivity to the spatial position of reaction stimuli, with respect to both increased error rate, and incomplete suppression of the Simon effect following S–R incompatible trials. This effect was accompanied by the higher amplitude of movement-related lateralised potentials in PD, with reduced BG output considered responsible for failing to keep early visuomotor activity from leaking activity to response-related neurons, and as a consequence, failed suppression of automatic response activation [56]. Notably, the former studies, with the exception of Praamstra and Plat, did not address sequence modulation of the effect.
Our study sought to investigate the SE phenomenon in patients with BG dysfunction, within the context of oculomotor behaviour, considered a relevant goal-directed action, especially given the functional and closely-coupled anatomical relationship between visuospatial attention and the oculomotor system [3], [16], [50], [51], [53]. Significantly, a number of imaging studies investigating covert shifts of attention, in the absence of actual eye movement, have demonstrated activation in the same network of structures involved in volitional eye movements, including frontal, parietal and cingulate cortices, without evidence of significant regional segregation [3], [16], [50], [51], [53]. Specifically, we sought to ascertain whether inhibitory dysfunction could be generalised beyond more conventional oculomotor behaviours, and beyond the context of more arbitrary responses like manual push-button or space-bar responses [62]. Given that the BG are necessary for the motor system to sustain competing response tendencies, it was anticipated that in PD, results would reveal a failure to inhibit automatic response tendencies, and that sequential analyses would reveal anomalous modulation.
Section snippets
Participants
Ten patients with mild to moderate idiopathic PD participated voluntarily in this study. All were clinically diagnosed with PD by a neurologist (O.W.), with motor disabilities responsive to anti-Parkinsonian medication. Ages varied between 46 and 70 years (M = 60.20 years, SD = 8.97 years), with clinical manifestations of the disease evident from 1 to 17 years. Disease severity was evaluated using the motor sub-scale of the Unified Parkinson's Disease Rating Scale. Clinical data for this group
Results
For all correctly executed trials, a significant Group effect, F(1, 18) = 8.75, P < 0.01, demonstrated that saccadic latencies were significantly shorter for PD patients (568 ms) than control subjects (723), overall. A main effect of Present trial compatibility, F(1, 18) = 18.17, P < 0.001, revealed that for both groups, compatibility of stimulus location and response side influenced latencies (or response times) according to the pattern typically found in Simon tasks. Responses following S–R
Discussion
The present study sought to evaluate response tendencies and inhibitory modulation of automatic response processes in PD. Specifically, this investigation utilised a spatial Simon task with S–R compatibility determined on the basis of cue shape. Given the closely-coupled functional relationship between attention and eye movement, and the overlap of neural networks activated by each [3], [16], [50], [51], [53], we adopted a saccadic eye movement paradigm for this investigation. Unlike control
Acknowledgments
We gratefully acknowledge the help and cooperation of all subjects who so willingly participated in this study, as well as Professor Robert Iansek from the Movement Disorders Laboratory at the Kingston Centre for his assistance with recruitment and assessment of patients. This research was supported, in part, by a Jean Gilmore bursary, awarded to J. Fielding by the AFWU-SA Inc. Trust Fund.
References (77)
- et al.
A parametric fMRI study of overt and covert shifts of visuospatial attention
NeuroImage
(2001) - et al.
Re-orientation of attention in Parkinson's disease: an extension to the vibrotactile modality
Neuropsychologia
(1993) - et al.
Deficits in saccadic eye-movement control in Parkinson's disease
Neuropsychologia
(2005) - et al.
A common network of functional areas for attention and eye movements
Neuron
(1998) - et al.
Cognition and the inhibitory control of saccades in schizophrenia and Parkinson's disease
Prog. Brain Res.
(2002) - et al.
Mini-Mental State. A practical method for grading the cognitive state of patients for the clinician
J. Psychiatr. Res.
(1975) - et al.
Lateral prefrontal damage affects processing selection but not attention switching
Cogn. Brain Res.
(2002) - et al.
Verbal memory in non-demented patients with idiopathic Parkinson's disease
Neuropsychologia
(1999) - et al.
Cognitive functions and corticostriatal circuits: insights from Huntington's disease
Trends Cogn. Sci.
(1998) The pathophysiology of movement disorders
Neurol. Clin.
(1984)