Elsevier

Neuropsychologia

Volume 49, Issue 1, January 2011, Pages 34-42
Neuropsychologia

Specific impairment of visual spatial covert attention mechanisms in Parkinson's disease

https://doi.org/10.1016/j.neuropsychologia.2010.11.002Get rights and content

Abstract

Visual deficits in early and high level processing nodes have been documented in Parkinson's disease (PD). Non-motor high level visual integration deficits in PD seem to have a cortical basis independently of a low level retinal contribution. It is however an open question whether sensory and visual attention deficits can be separated in PD. Here, we have explicitly separated visual and attentional disease related patterns of performance, by using bias free staircase procedures measuring psychophysical contrast sensitivity across visual space under covert attention conditions with distinct types of cues (valid, neutral and invalid). This further enabled the analysis of patterns of dorsal–ventral (up–down) and physiological inter-hemispheric asymmetries. We have found that under these carefully controlled covert attention conditions PD subjects show impaired psychophysical performance enhancement by valid attentional cues. Interestingly, PD patients also show paradoxically increased visual homogeneity of spatial performance profiles, suggesting flattening of high level modulation of spatial attention. Finally we have found impaired higher level attentional modulation of contrast sensitivity in the visual periphery, where mechanisms of covert attention are at higher demands.

These findings demonstrate a specific loss of attentional mechanisms in PD and a pathological redistribution of spatial mechanisms of covert attention.

Research highlights

▶ Specific loss of attentional mechanisms in Parkinson's disease, independently of sensory deficits. ▶ Pathological redistribution of spatial mechanisms of covert attention in Parkinson disease. ▶ Loss of right hemispheric dominance of visual attention in Parkinson's disease.

Introduction

There is substantial evidence for non-motor manifestations in Parkinson's disease (Archibald et al., 2009, Bodis-Wollner, 1990, Bodis-Wollner, 2003, Bodis-Wollner et al., 1987, Bodis-Wollner and Tzelepi, 1998, Bodis-Wollner and Yahr, 1978, Castelo-Branco et al., 2009, Mosimann et al., 2004, Silva et al., 2005, Uc et al., 2005, Uc et al., 2007, Van Asselen and Castelo-Branco, 2009). Patient studies clearly separating low level sensory and visual spatial attention deficits within the same task are however still lacking. Recent covert attention studies in normal subjects have allowed for the separation of low level sensory processing from the performance enhancement effects of spatial attention (Carrasco, 2006, Pestilli and Carrasco, 2005). These studies provide a novel methodological opportunity to isolate and distinguish such sources of impairment in patients.

The neural basis of visual deficits in PD (Archibald et al., 2009, Bodis-Wollner, 2003, Silva et al., 2005) and the elucidation of which particular processing mechanisms are impaired (e.g., low level contrast detection and/or high level visual attention mechanisms) are crucial issues that can only be addressed if one measures at the same time contrast sensitivity (CS – reciprocal of Threshold measures) with and without manipulation of spatial bias of attention. We have previously demonstrated the advantages of simultaneously studying low and high level visual functions as a way to disentangle the neural origins of visual sensory, perceptual and cognitive deficits (Castelo-Branco et al., 2006, Castelo-Branco et al., 2007, Castelo-Branco et al., 2009, Kozak and Castelo-Branco, 2009). Also, we have previously developed visual CS tasks under conditions that control the spatial distribution of attention (Silva et al., 2008). Adding a cue leads to an asymmetric redistribution of spatial attention. Experimental control is further enhanced by running randomly interleaved psychophysical staircases in space and time, which helps exploring low and high level mechanisms underlying anisotropies in spatial vision (Silva et al., 2008, Silva et al., 2010).

Visual orienting is related to mechanisms of spatial attention in parietal cortex and is often associated with right hemispheric dominance (Davidson and Hugdahl, 2004, Ivry and Robertson, 1998). The main mechanisms of automatic “exogenous” orienting of spatial attention can be related to inhibition of return (IOR) and automatic orienting. Visual spatial attention can be covertly dissociated from the direction of gaze in a voluntarily driven way, via a mechanism known as “endogenous” attention. This mechanism is in contrast with the above mentioned automatic, stimulus-driven orienting termed “exogenous” attention (Posner & Cohen, 1984). Cueing paradigms are frequently used to study endogenous and exogenous orienting of attention (Posner & Cohen, 1984). An important distinguishing factor between the two is the difference in their time-courses. Whereas the effects of endogenous attention require a few hundred milliseconds to fully develop and can be maintained with effort, exogenous attention peaks within 100–120 ms and diminishes rapidly thereafter (Cheal and Lyon, 1991, Nakayama and Mackeben, 1989).

To explicitly tackle the question of how attention modulates visual performance, we have focused on paradigms where cueing is known to be facilitatory and not to induce inhibition of return, which refers to the slowing of a response to a target stimulus presented in the same location as a previous stimulus (Klein, 2000). Accordingly, at relatively short (e.g., 150-ms) cue–target stimulus onset asynchronies (SOAs), attentional orienting to targets at cued vs. uncued locations is facilitated, whereas at relatively long SOAs (e.g., beyond 300 ms), it is inhibited (Klein, 2000). Reductions in IOR have been argued to reflect impaired inhibitory processes in PD (Poliakoff et al., 2003, but see Grande et al., 2006). Here we were not focused on IOR, which reflects higher level late attentional processing, but rather on cueing with short cue-target SOA to measure early attentional facilitatory/inhibitory effects on visual performance of valid vs. invalid cues (Pestilli & Carrasco, 2005).

Putative deficits in visual orienting are also relevant in terms of visual performance asymmetries. Sources of performance asymmetries have been documented at different levels of the visual system (nasotemporal at the level of the retina, up–down at level of the retina, lateral geniculate nucleus and early visual cortex and left–right spatial cortical hemispheric asymmetries). Although some of these asymmetries may cancel out, as is the case of monocular nasotemporal asymmetries, in general a physiological/behavioral consequence can be identified (Silva et al., 2010). This is clearly the case concerning binocular left–right asymmetries (Silva et al., 2008). The relation of these functional asymmetries to cell density across retinotopic representations at different levels of the visual system have been addressed in our previous work (Silva et al., 2008, Silva et al., 2010). In brief, sources of spatial asymmetries in performance within the visual field (VF) can be ascribed to lower levels of visual processing, such as occipital cortex or even the retina (Carrasco et al., 2004, Maia-Lopes et al., 2008, Silva et al., 2008). A pattern of up/down visual field asymmetry has been shown to be also present at the level of striate/extrastriate cortices (Maunsell & Van Essen, 1987).

Probing such anatomic and physiological substrates of attentional and sensory performance may help provide tools to dissect the different visual processing steps that are affected in PD, including visual attention (Kingstone et al., 2002).

In this paper we have followed the seminal work of Carrasco and colleagues that were to first to show a way to separate sensory perception from attentional enhancement of such performance. We have now extended this innovative strategy to patient work on the non-motor cognitive processes that are often impaired in Parkinson disease (Van Asselen et al., 2009, Van Asselen and Castelo-Branco, 2009). It has indeed been hypothesized that attentional processes are more active in PD patients (Briand, Hening, Poizner, & Sereno, 2001), raising the question whether this is due to increased facilitation or reduced inhibition.

There is widespread evidence on the relation between spatial hemispheric dominance and the overlap between the attentional and spatial orienting network (Davidson and Hugdahl, 2004, Ivry and Robertson, 1998). Orienting of spatial attention can be related to anisotropic patterns of psychophysical performance (Nakayama and Mackeben, 1989, Silva et al., 2008) but it is important to recognize that cortical contribution to asymmetric visual performance independently of attentional biases has also been recently considered (Carrasco et al., 2004, Carrasco et al., 2001, Fuller et al., 2008). This is the case in terms of dorso/ventral (up/down) asymmetries in letter identification (Mackeben, 1999), visual acuity (Altpeter, Mackeben, & Trauzettel-Klosinski, 2000) and attentional conjunctive visual search tasks (He, Cavanagh, & Intrilligator, 1996). Furthermore, performance on orientation discrimination tasks depends on the up/down target location (Carrasco et al., 2001). The relevance of the functional superiority of the inferior field in primates and humans is also documented by the over-representation of the lower visual field in area MT (Maunsell & Van Essen, 1987) and in area V6A (Galletti et al., 1999).

Most of the current evidence for functional anisotropies does nevertheless relate to the right hemispheric dominance of spatial attention and to the beneficial effect of covert spatial attention and valid cues in normal subjects (for a review see Carrasco, 2006). Contrast sensitivity represents a basic visual performance dimension where substantial evidence for the role of focused spatial attention on performance improvement has been achieved. Covert attention may be understood as a neural process that enhances the signal (and thereby contrast sensitivity) from a particular part of the sensory scenario. Indeed, in normal subjects, transient covert attention increases contrast sensitivity at the target location with an informative spatial cue (Pestilli & Carrasco, 2005). Transient covert attention implies visual scanning in the absence of eye movements (prior to planning a saccade or not) and may have both benefits and costs. Accordingly, it may enhance contrast sensitivity at cued locations and impair contrast sensitivity at uncued (neutral) or invalid locations.

Here, we have dissected sensory (using baseline contrast sensitivity), early (superior/inferior) visual patterns of performance and parietal visual attention networks (by assessing left/right performance and validity effects under covert attention conditions) in early stage PD.

Performance was assessed using an achromatic contrast sensitivity task that probed a parvocellular-biased spatiotemporal frequency channel using stationary sinusoidal gratings of intermediate spatial frequency (ISF, Silva et al., 2005, Silva et al., 2008). These patterns were previously shown to also yield hemifield patterns of cortical physiological asymmetry in normal subjects, thereby proving to be adequate for the present study (Silva et al., 2008). Right/left asymmetries are a direct reflection of the hemispheric dominance of the right hemisphere in visuospatial attention, and their study thereby provides, in addition to the effect of peripheral valid, invalid and neutral cues, further clues to the study of attentional deficits in Parkinson's disease. These concepts relate visual transient attention (and orienting) mechanisms with the notion of limited resources and thereby provide an interesting additional paradigm to probe deficits in PD.

In sum, we expected PD patients to show distinct spatial patterns of performance as compared to controls and in particular to show impairment on early facilitation of valid cues. Testing at two distinct eccentricities helped probing the prediction whether more peripheral locations (where attentional demands are higher) show a validity effect as compared to central ones, where attentional demands are lower. We have mainly tested exogenous attention and facilitation, by using a 100 ms cue-target onset asynchrony.

Section snippets

Participants selection

35 PD patients (18 males, 17 females) were initially included for this study. The control group comprised 24 demographically matched subjects. Seven patients were excluded due to neuropthalmological exclusion criteria (see below). Given that the main goal of the study was focused on covert attention mechanisms we have further excluded PD patients that were not able to maintain fixation and/or to inhibit reflexive saccades in the presence of exogenous cues (n = 11). We discuss below this

Methodological exclusion of patients unable to keep covert attention

A substantial proportion of our PD patients showed a surprisingly high tendency to move their eyes towards attentional cues. Our data indicated that pre-selected PD patients have significantly impaired ability to inhibit reflexive saccades when an additional visual pre-cue is added, as compared to controls (Chi-square test, p < 0.01). This difference justified our strict exclusion criteria.

A signature of the loss of inhibitory control typically seen in our PD patients can be observed in Fig. 2.

Baseline sensory performance in PD

Concerning central Zone 1 (Fig. 3), we have found a dissociation in performance between controls and PD patients: the pooled contrast sensitivity in Zone 1 was significantly different between groups (p = 0.01, ANOVA). Given that the study of main effects “collapses” variables across “post hoc” levels we have also tested whether this difference held true across regions (Fig. 3). Interestingly, this difference was still significant even when the analysis was confined to spatial subregions (left, p = 

Discussion

In this work we were able to identify a specific attentional deficit in PD that could be isolated by explicitly controlling for baseline performance in contrast sensitivity tasks. Our paradigm represents an advantage over traditional paradigms because it does not merely rely on reaction time or percent correct measures which are prone to bias (Fuller et al., 2008). We focused on true psychophysical sensitivity and staircases provide in this respect a great advantage when comparing performance

Acknowledgements

This research was funded by grants from the Portuguese Science and Technology Foundation (FCT): PTDC_SAU_NEU_68483_2006, PTDC_PSI_67381_2006 and PIC_IC_82986_2007, as well as by the National Brain Imaging Network of Portugal (BIN).

References (45)

  • M. Mendes et al.

    Visual magnocellular and structure from motion perceptual deficits in neurodevelopmental model of dorsal stream function

    Brain Research Cognitive Brain Research

    (2005, December)
  • F. Pestilli et al.

    Attention enhances contrast sensitivity at cued and impairs it at uncued locations

    Vision Research

    (2005)
  • M.F. Silva et al.

    Retinal and cortical patterns of spatial anisotropy in contrast sensitivity tasks

    Vision Research

    (2008, January)
  • N.K. Archibald et al.

    The retina in Parkinson's disease

    Brain

    (2009)
  • K.M. Bennett et al.

    Covert visuospatial attentional mechanisms in Parkinson's disease

    Brain

    (1995, February)
  • I. Bodis-Wollner et al.

    Visual dysfunction in Parkinson's disease. Loss in spatiotemporal contrast sensitivity

    Brain

    (1987)
  • I. Bodis-Wollner et al.

    Measurements of visual evoked potentials in Parkinson's disease

    Brain

    (1978)
  • D.B. Boles

    A large-sample study of sex differences in functional cerebral lateralization

    Journal of Clinical and Experimental Neuropsychology

    (2005)
  • P. Caffarra et al.

    Orienting of visual attention in Alzheimer's disease: Its implication in favor of the interhemispheric balance

    Neuropsychiatry and Neuropsychology Behavioral Neurology

    (1997)
  • M. Carrasco et al.

    Characterizing visual performance fields: Effects of transient covert attention, spatial frequency, eccentricity, task and set size

    Spatial Vision

    (2001)
  • M. Castelo-Branco et al.

    Visual phenotype in Williams-Beuren syndrome challenges magnocellular theories explaining human neurodevelopmental visual cortical disorders

    The Journal of Clinical Investigation

    (2007, December)
  • M. Cheal et al.

    Central and peripheral precuing of forced-choice discrimination

    Quarterly Journal of Experimental Psychology A: Human Experimental Psychology

    (1991)
  • Cited by (18)

    • Sex, food and threat: Startling changes after subthalamic stimulation in parkinson's disease

      2013, Brain Stimulation
      Citation Excerpt :

      The fact that larger ABR magnitudes were not reported in a previous study on affective modulation of the ABR in off-medicated PD patients [31] may be due to a substantially shorter disease duration than in our patients (mean 5.5 (SD = 4) vs. 14.4 (SD = 3) years). Indeed, attentional deficits have been documented in PD patients, and there is evidence for their progression with disease duration [32,33]. Affective modulation of the ABR becomes evident during viewing of affect-weighted pictures [30], with ABR facilitated by aversive and inhibited by appetitive picture contents reflecting the activation of the aversive and appetitive systems [16].

    • Attention and visual dysfunction in Parkinson's disease

      2012, Parkinsonism and Related Disorders
      Citation Excerpt :

      This overlap between consciousness, working memory and selective attention has been highlighted by others [63], and the suggestion that dopamine plays a crucial goal in gating the information that enters consciousness is intriguing. It is clear that PD patients have problems with ‘top-down’ attention [79], and it can be inferred that altered regulation of dopamine function causes a loss or inflexibility of task relevant top-down biasing signals. This potentially results either in important visual information failing to gain access to WM, or that the information currently occupying the ‘workspace’ fails to be protected from interference, which in turn brings about deranged subsequent biasing signals.

    • Selective attention deficits in early and moderate stage Parkinson's disease

      2012, Neuroscience Letters
      Citation Excerpt :

      These findings are consistent with the possibility that reflexive attentional processes in PD patients may be more active [2]. Sampaio et al. found that PD subjects showed impaired psychophysical performance enhancement by valid attentional cues [35]. Moreover, hyper-reflexivity has also been observed in studies of saccadic eye movements by Kingstone et al. [20].

    • Motor asymmetry and estimation of body-scaled aperture width in Parkinson's disease

      2011, Neuropsychologia
      Citation Excerpt :

      Consistent with this hypothesis, patients are more susceptible to visual distracters in peripheral vision (Deijen, Stoffers, Berendse, Wolters, & Theeuwes, 2006; Machado, Devine, & Wyatt, 2009). A general impairment in visual attention is shown by the failure in PD to benefit from valid spatial cues, or to follow instructions not to look directly at stimuli presented in the periphery (Sampaio et al., 2011). A second type of explanation for mobility problems is that changes to spatial vision and/or proprioception affects navigation in cluttered environments in PD (Lee, Harris, Atkinson, & Fowler, 2001a; Rahman et al., 2008).

    • iSTIMULI: Prescriptive Stimulus Design for Eye Movement Analysis of Patients with Parkinson’s Disease

      2023, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    View all citing articles on Scopus
    View full text