Dissociated long lasting improvements of straight-ahead pointing and line bisection tasks in two hemineglect patients

Abstract

In this experiment, we evaluated over a longer time period the previously demonstrated effects of a short prism adaptation on hemispatial neglect. We followed two patients (PE and SA), during a period of 5 days (1 day before and 4 days after the prism adaptation procedure), repeatedly measuring their performances on a straight-ahead pointing task and a line bisection task. We also assessed the comparative temporal evolution of the rightward biased egocentric reference frame (as measured by the straight-ahead demonstration) and a classical neuropsychological symptom of neglect, namely the rightward bias observed on line bisection. Firstly, the results showed that prismatic effect could be maintained for 4 days, on the two tasks (separately straight-ahead for PE, line bisection for SA). This long-term effect implies a very profound action of prism adaptation based on active processes and opens large possibilities for clinical applications. Secondly, no correlation was found between the evolution of the performances on the two tasks, neither for patient PE, nor for patient SA (within-subject double-dissociation). Moreover, a double-dissociation between subjects was demonstrated on long-term effects. A new conception has thus to be found to explain the various symptoms manifested in neglect and more investigations have to be performed in order to establish to what extent they can be considered independent. It can be concluded that elucidating the mechanism through which prism adaptation affects neglect could lead to a better understanding of the neglect syndrome.

Introduction

The syndrome of unilateral neglect is frequently observed in right-handed patients following right hemisphere brain damage. This condition can be defined as a loss of visual awareness for contralesional stimuli despite residual implicit processing [10], [24]. Typically, the patient fails to report, respond to or orient toward left-sided stimuli [15], [18], or tends to underestimate it [28]. Neglect symptoms have been described to range from sensori–motor levels [17], [26] to mental representations of space [4], [31]. In addition, neglect is responsible for poor functional recovery and is resistant to treatment [12].

Orienting in space requires an integration of retinal, eye-position and head-position signals, together with vestibular information. These signals are classically considered to be further transformed to build a unitary egocentric reference, aligned with the sagittal body axis (review in [21]). This hypothesis draws on the notion that spatially-directed behaviour from perception to action has to be coded in a supramodal coordinate system. One of the classical manifestations of hemispatial neglect is an alteration of the egocentric reference, which can be tested by requiring the subject to point straight-ahead in the dark [20], [44]. Patients often produce subjective mid-line demonstrations deviated to the right [7], [16].

Along this line, a reference-shift hypothesis of neglect has been proposed, according to which all the orienting biases observed in neglect patients [1], [13], [17], [26], [27] are due to “an illusory rotation of the egocentric reference, somewhat as if the subject felt being constantly rotated toward the lesion side” [19]. Postulating that the rightward shift of the egocentric reference constitutes the anchor of the neglect syndrom [23], this hypothesis predicts that a restoration of the centred straight-ahead demonstration should produce an improvement in performance on other neglect tests. Support for this hypothesis was found in the temporary improvement of neglect patients after different types of physiological manipulation [5], [22], [30], [31], [32], [37], [41] that compensate for the pathological shift of the egocentric reference, and hence afford an improvement of the patient's behaviour in the left hemispace [23].

More recently, however, Chokron and Bartolomeo [6], Farnè et al. [11] and Bartolomeo and Chokron [2] studied two, 23 and 18 neglect patients respectively and challenged this reference shift theory of hemispatial neglect by showing that not all patients exhibited a rightward shift of the egocentric reference. In those three studies, a total of 43 patients were examined, and only 27 of them exhibited a deviation of the manual straight-ahead to the right. In addition, Farnè et al. [11] showed that the same proportion of right brain-damaged patients without neglect exhibited a similar trend.

In a previous investigation, we studied 16 neglect patients with the aim of testing the effects of the adaptation to an optical deviation towards the ipsilesional side on their spatial deficit through two experiments [34]. In the first experiment, the effect of prism adaptation was tested on manual straight-ahead demonstration with eyes closed. After adaptation, all patients exposed to the optical shift of the visual field (n=8) showed an improvement on this task. In the second experiment, the effect of prism adaptation on clinical manifestations of neglect was tested on six patients, as assessed by classical neuropsychological tests. The battery included the line bisection task, line cancellation task, drawing from memory, copying and reading. Upon removal of the prisms, each of the six patients (four with hemianopia and two without hemianopia) showed an improvement in performance on each of these five tests, showing that prism adaptation improves symptoms of spatial neglect [34]. A control group of six neglect patients performing the pointing procedure with neutral goggles exhibited no significant improvement. By contrast with the other physiological manipulations [5], [22], [30], [31], [32], [37], [41], this improvement was maintained 2 h after the prism exposure [34].

The first question raised by these former results involved the potential duration of the benefits induced by prism adaptation. We therefore followed the performances of two neglect patients over a testing period of 1 week on two continuous and sensitive tests adapted to repeated measures: straight-ahead demonstration and line bisection.

This longitudinal study allowed us to assess a secondary question about the reference shift hypothesis of neglect. In our previous study [34], the three neglect patients who took part in the two experiments were improved on both the straight-ahead demonstration (Exp. 1) and the classical neuropsychological tests (Exp. 2). This conjunction may support the view that the core of neglect is indeed a deviation of the internally represented midsagittal plane of the body [23]. If this view is correct, a covariation should be expected for the two types of performance. Alternatively, if the deviation of the egocentric reference and the visual neglect symptoms rely on separate mechanisms, it should be possible to selectively alter one of them without affecting the other (simple dissociation) and vice versa (double-dissociation). If no correlation is observed between the two tested variables, then at least two different and independent mechanisms have to be advocated, each being responsible for a particular neglect symptom. Hence, we also compared the effects of prism adaptation on straight-ahead demonstrations (evaluating the egocentric reference) and on line bisection (one of the most sensitive tests of neglect) in these two neglect patients. A lack of correlation between the egocentric reference and a cancellation task has been obtained recently with a group analysis [11]. It should be emphasised here that the lack of an inter-individual correlation (as obtained by [11] or [2]) does not rule out the possibility for an intra-individual link between the two variables explored. Only longitudinal studies can distinguish changes over time within individuals from differences among people in their baseline levels (cohort effect); Cross-sectional studies cannot [9]. Repeated observations of individuals enable a direct study of change in parameters A and B. These changes can be used to assess the link between A and B in a more sensitive manner. The advantages of using prism adaptation is that it allowed us to compare the temporal evolution of the two parameters within-subject (as well as between subjects), assuming that prism adaptation will trigger alterations in at least one of the tested parameters. If the two parameters were to show the same evolution over time, it would be compatible with the hypothesis that the prismatic effect on cognitive symptoms is mediated by an alteration of the egocentric reference.