Highlights
- •
Topographical troubles are frequent among stroke patients with unilateral neglect.
- •
Virtual reality tools allow for evaluating the spatial navigation of individuals with neglect in a safe and informative manner.
- •
Individuals with and without neglect were helped by lateralized auditory cues during a virtual spatial navigation task.
- •
The help provided to individuals with neglect by such cues was potentiated by one prism adaptation session.
Abstract
Background
Unilateral neglect is common among right-hemispheric stroke individuals and also concerns the auditory modality. Prism adaptation can improve auditory extinction during a dichotic listening task, but its effect during an ecological task has not been studied.
Objective
The main objective was to evaluate whether lateralized cueing before and after prism adaptation improved virtual spatial navigation of stroke individuals with visual and auditory unilateral neglect. Secondary objectives were to assess spatial memory and obtain a better understanding of the mechanism of the cueing treatment by using an eye-tracker.
Methods
We included 22 stroke individuals with left visual and auditory neglect, 14 individuals without neglect, and 12 healthy controls. After a familiarization task, participants underwent 3 evaluation sessions. Participants were first passively shown a path that they had then to actively reproduce by using a joystick. A path with lateralized beeping sounds indicating direction and a path without any sounds were followed in a randomized order. After prism adaptation, the participants followed a third path with lateralized beeping sounds. The time of navigation and number of trajectory mistakes were recorded. After navigation, spatial memory was assessed. Additionally, an eye-tracker was used during the navigation period.
Results
The navigational performance of participants with neglect was significantly better with than without auditory cues, especially after prism adaptation. With auditory cues, participants without neglect reached the navigational performance of healthy controls. The spatial memory of individuals with neglect was significantly lower with auditory cues. Eye-tracking analyses showed that participants with neglect made more saccades and looked longer at the right-square angles in the absence of auditory cues.
Conclusions
This study demonstrates the positive effect of auditory cues in virtual spatial navigation of individuals with visual and auditory neglect and the potentiation of the help of cues after prism adaptation.
1
Introduction
Unilateral neglect is frequent after a right-hemispheric stroke and can be defined as a loss of the ability to react to or treat sensory stimuli presented on the contralesional side of a hemispheric brain lesion, associated with reduced perceptual, motor, attentional and/or representational ability of the individual inside this space . The visual canal is usually used as a compensation and stimulation modality for the stimulated hemi-space . Yet, neglect also concerns other sensory modalities, such as audition, and people with neglect have been shown to have deficits of detection, identification, and localization of contralesional sounds . Such deficits have been reported when using free-field stimulation , dichotic presentation with an earphone , a pointing-to-sounds technique , and auditory midline perception . Individuals with neglect also show non-lateralized auditory deficits , which suggests that such auditory deficits are due to disturbed multisensory spatial integration processes . Brain structures, such as the parietal cortex, superior temporal lobe , and frontal and pre-motor cortex , may be involved in multisensory spatial integration processes. Prism adaptation is a promising bottom-up rehabilitation technique for unilateral neglect , and its efficacy on visual , tactile , and auditory symptoms of neglect has been demonstrated. Jacquin-Courtois et al. demonstrated that prism adaptation improved auditory extinction evaluated by a dichotic listening task.
Several studies have evaluated the impact of cues on spatial neglect and showed that individuals with neglect were slow, or even unable, to accurately react to stimuli in the contralesional space when competing distracters were present . This phenomenon is explained by perceptual extinction in the presence of double stimulation. The effect of cues on attention in individuals with spatial neglect is not yet well known. Robertson (1995; 2001) claimed that people with neglect presented a non-lateralized attentional deficit and showed that phasically increasing the person’s alertness decreased the spatial bias in perceptual awareness. Others assumed that the attentional deficit in people with neglect was lateralized . Schürmann et al. tested individuals using visual and auditory cues in a Posner task and suggested that 3 components impaired their performance in the Posner paradigm: a non-spatial supramodal deficit, a deficit of the contralesional hemi-field, and an extinction-type impairment. The effect of cues on neglect could be secondary to non-lateralized enhancement of sustained attention or displacement of the focus of attention. Eye-trackers have been used to study the underlying attentional mechanisms of spatial neglect .
Contrary to paper-and-pencil tasks, the use of a virtual environment allows for the evaluation of spatial cognition on a large scale, such as that in the real environment . The assessment of spatial navigation in a virtual environment allows for the collection of immediate feedback on performance, using various forms and modalities to administer the cues in a controlled and standardized way to improve the person’s errorless performance and develop potential aids for spatial navigation. Several virtual reality tools have been used to evaluate spatial cognition in people with unilateral neglect .
Here, our main objective was to evaluate whether lateralized cueing before and after a prism adaptation session improved virtual spatial navigation of individuals with right hemispheric stroke and visual and auditory unilateral neglect (objective 1). The secondary objectives were to (1) evaluate whether lateralized cueing before and after a prism adaptation session improved the spatial memory of these individuals (objective 2) and (2) evaluate the mechanism of treatment by lateralized auditory cues by using an eye-tracking device (objective 3).
2
Methods
In this study, the condition without auditory cues was called “without auditory cues”; the condition with auditory cues before prism adaptation “with auditory cues”; and that with auditory cues after prism adaptation “after prism adaptation”.
2.1
Study design
This study was an exploratory, prospective, controlled, and randomized trial, promoted by the University of Bordeaux. Informed written consent was obtained from all participants before testing, and the study was approved by the Ethics Committee Sud-Ouest et Outre-Mer III in November 2016. We included 48 individuals: 22 with right hemispheric stroke with unilateral visual and auditory neglect, 14 with right hemispheric stroke without visual or auditory neglect, and 12 healthy controls.
2.2
Participants
At the time of testing, the 36 individuals with a first focal right hemispheric stroke were in- or out-patients at one of the neurorehabilitation centers of Bordeaux (University Hospital of Bordeaux, La Tour de Gassies, Arcachon Hospital). All had stroke and fulfilled the following inclusion criteria: presence of a unilateral focal right hemispheric stroke documented by MRI and/or CT; no previous history of neurological or psychiatric illness; and preserved visuo-motor coordination with the right hand; 22 (group 1) had left unilateral visual neglect, assessed by classical neuropsychological tests (Behavioral Inattention Test battery) , line bisection task , copy-drawing task , Bell’s test , and Catherine Bergego questionnaire (completed by the individual with an occupational therapist) . The presence of visual neglect was defined by a deficit in at least 2 of these tests. The presence of left-ear extinction was assessed by using a dichotic listening task developed by the Neurorehabilitation Unit of Vaudois University Hospital. The other 14 stroke participants (group 2) had no unilateral neglect (neither visual nor auditory). The 12 healthy controls (group 3) had no history of brain damage. All individuals were right-handed. Stroke participants and healthy controls spoke French, were able to read and understand short instructions, and did not present any of the following exclusion criteria: loss of visual acuity that prevents the use of a computer, average or severe deafness (40 dB of hearing loss in one ear measured by pure tone audiometry) or asymmetric loss (>10 dB difference in the mean threshold between the right and left ear); major behavioral problems or comprehension deficits; dementia, according to the criteria of the Diagnostic and Statistical Manual of Mental Disorders, 4 th edition ; or legal incapacity, epileptic seizure during the 6 months preceding the testing, except for the 2 weeks after the onset of brain damage, pregnancy or breast-feeding, or under age 18.
2.3
Experimental procedures
2.3.1
Evaluation in a virtual district
The virtual environment was a 3-D reproduction of a district that represented a medium-sized North American town (created in 2015 by Florian Larrue, engineer) ( Fig. 1 ). Instructions were given orally before the beginning of the training and evaluation sessions. First, the investigator demonstrated and explained the correct use of the joystick. The software began the familiarization task in a different part of the virtual district than that used for the paths. No paths used for the evaluations could be crossed during the familiarization process. The participants were told to navigate until they felt completely comfortable with the software and joystick, without any time limit, to ensure that the observed differences between participants were not due to a lack of dexterity or a comprehension deficit.
Participants performed a series of 3 evaluation sessions of 15 minutes each. A path, including 6 intersections with one landmark for each, was passively shown to the participants with a 23” laptop (learning stage). Then, they had to actively follow the same path by using the joystick (reproduction stage). Participants were asked to observe (for 180 s), then navigate 2 different paths, one with and the other without auditory cues. The order of the 2 paths was randomized. Auditory cues consisted of beeping sounds that indicated the direction to take at each intersection: when the participant had to turn right or left, beeping sounds started 5 m (16.40 feet) before the intersection in the right or left ear, respectively, through an earphone; when the participant had to continue straight-ahead, beeping sounds started 5 m before the intersection in both ears. Finally, participants had to follow a third path with auditory cues after a prism adaptation session.
2.3.2
Prism adaptation: (as detailed by Jacquin-Courtois et al. )
We performed the 3 steps associated with the prism adaptation procedure: the pre-exposure baseline measurement of pointing (consisting of 10 pointing movements), the exposure to prismatic displacement to elicit sensorimotor adaptation (consisting of 100 pointing movements), and the post-exposure after-effect measurement (consisting of one pointing movement). The total duration of this exposure was approximately 10 minutes. All participants reached an after-effect of at least 3°, which reflected proper adaptation .
2.4
Main and secondary judgment criteria
2.4.1
Navigation assessment
During the reproduction path, the number of trajectory mistakes (/6) and the time of navigation were recorded (main judgement criteria, objective 1). The time of navigation consisted of the time from the moment the participant started to move to the end-point at which the computer stopped the simulation.
2.4.2
Spatial memory assessment (secondary judgment criteria, objective 2)
For the landmarks, the participants were asked to name the 6 landmarks they had seen along the path (free recall of landmarks) and to recognize them (recognition of landmarks) among a pictorial list that included 3 morphologically (e.g., a ladder of another shape and color than the one present in the environment), semantically (e.g., a scooter instead of a bicycle), or unrelated intruders for each landmark of the environment. For the route, participants had to repeat the landmarks in the chronological order of their occurrence during the task (landmark ordering) and also recall which direction they took at each intersection (direction choice at each intersection), by using screenshots of the intersections. For the survey, the participants had to choose the exact outline of the path they followed among 4 suggested pathways (path outline choice) and to point in the direction of the starting point from an image of the end point.
2.4.2
Eye-tracking analyses (objective 3)
We used a Tobii eye-tracker (Tobii Pro TX300), which recorded the movements of eyes of participants during the familiarization period and evaluation phase. The period of interest started 5 m before the intersection and ended 5 m after. During that period, we divided the screen between the right and the left side of the path. The judgment criteria were the total number of eye saccades and the duration of eye-fixation toward the right and left-square angles during the active and passive paths. First, we analyzed the results concerning the 6 intersections followed by the results concerning the 2 left, 2 right, and 2 straight-ahead intersections.
2.5
Statistical analyses
Statistical analysis involved use of SPSS v22. According to Mauchly’s Test of Sphericity, our dataset violated the assumption of sphericity, in particular, for one of the main judgment criteria (i.e., time of navigation): χ 2 (2) = 7.6, P < 0.05. Thus, we used a non-parametric approach. We used Kruskal–Wallis tests for between-group analyses. For significant results, pairwise comparison with Wilcoxon Signed–Rank test was used for within-group analyses, adjusted by Bonferroni correction given the large number of tests. Hence, P values were corrected for all Kruskal–Wallis and Wilcoxon Signed–Rank tests by using the formula px = k*ps , where px denotes the experiment-wise type-1 error rate, ps a single test type-1 error rate, and k the number of pairwise comparisons made. Moreover, for ANOVA and Kruskal–Wallis tests, partial eta squares (η 2 ) that reflect the effect size were calculated by using SPSS and the formula proposed by Tomczak and Tomczak . For Wilcoxon signed–rank tests, effect sizes were calculated by hand by using the Z values <SPAN role=presentation tabIndex=0 id=MathJax-Element-1-Frame class=MathJax style="POSITION: relative" data-mathml='r=Z/N’>?=?/?‾‾√r=Z/N
r = Z / N
as proposed by Tomczak and Tomczak . P < 0.05 was considered statistically significant.
3
Results
For the 22 individuals with right hemispheric stroke with unilateral visual and auditory neglect, the mean (SD) age was 65.8 (8.8) years (17 men); for the 14 with right hemispheric stroke without visual or auditory neglect, the mean (SD) age was 63.9 (15.5) years (9 men); and for the 12 healthy controls, the mean (SD) age was 67.6 (10.0) years (7 men) ( Table 1 ).
| Demographic status | With neglect | Without neglect | Healthy controls | ANOVA results |
|---|---|---|---|---|
| Age (year), mean (SD) | 65.8 (8.8) | 63.9 (15.5) | 67.6 (10.0) | F(2,45) = 0.3, P = 0.7, partial η 2 = 0.02 |
| Sex (M/W) | 17/5 | 9/5 | 7/5 | Chi 2 = 3.5, P = 0.2 |
| Level of education (1; 2; 3) | 1; 6; 15 | 3; 8; 3 | 0; 3; 9 | F(2,45) = 6.8, P = 0.003, partial η 2 = 0.2 |
| Etiology of stroke (i/h) | 13/9 | 8/2 | NA | F(1,34) = 2.9, P = 0.10, partial η 2 = 0.08 |
| Delay post-stroke (days) | 662.4 | 274.3 | NA | F(1,34) = 3.3, P = 0.08, partial η 2 = 0.09 |
| Santa Barbara Sense of Direction Scale (/7), mean (SD) | 3.3 (1.0) | 3.6 (1.4) | 3.9 (1.3) | F(2,45) = 1.1, P = 0.4, partial η 2 = 0.04 |
Results of the neuropsychological assessments and those of spatial navigation and memory of the 3 groups are in Table 2 and Fig. 2 , respectively.
| Neuropsychological evaluations | With neglect | Without neglect | Healthy controls | ANOVA results |
|---|---|---|---|---|
| Visual neglect | Group a | Group b | Group c | |
| The line bisection task | ||||
| Line 1 (distance in cm) | 9.5 (0.7) * , c | 9.7 (1.1) | 9.6 (0.4) * , a | F(2,45) = 0.8, P = 0.5, partial η 2 = 0.03 |
| Line 2 (distance in cm) | 9.7 (0.5) | 9.7 (0.9) | 9.4 (0.7) | F(2,45) = 2.1, P = 0.1, partial η 2 = 0.09 |
| Line 3 (distance in cm) | 11.0 (2.0) | 9.9 (1.4) | 9.6 (0.6) | F(2,45) = 3.9, P = 0.03, partial η 2 = 0.2 |
| Copy-Drawing Task (/1) | 0.9 (0.6) | 0.9 (± 0.4) | 1.0 (0.0) | F(2,45) = 1.6, P = 0.2, partial η 2 = 0.07 |
| Figure and shape copying (/3) | 1.8 (0.9) ** , b , *** , c | 2.6 (0.9) ** , a | 3.0 (0.0) *** , a | F(2,45) = 11.6, P = 0.0001, partial η 2 = 0.3 |
| Letter Cancellation (/40) | 30.7 (8.1) ** , c | 35.8 (4.0) | 38.2 (2.2) | F(2,45) = 6.7, P = 0.003, partial η 2 = 0.2 |
| Star cancellation (/54) | 40.1 (10.6) * , b , *** , c | 47.9 (9.6) * , a | 52.9 (1.4) *** , a | F(2,45) = 8.6, P = 0.0007, partial η 2 = 0.3 |
| Albert’s test (/36) | 34.6 (2.5) | 35.7 (0.8) | 36.0 (0.0) | F(2,45) = 3.5, P = 0.04, partial η 2 = 0.1 |
| Representational drawing (/3) | 2.4 (0.9) | 2.9 (0.3) | 3.0 (0.0) | F(2,45) = 3.9, P = 0.03, partial η 2 = 0.2 |
| Bell’s test | ||||
| Total duration (s) | 134.4 (64.7) | 139.8 (70.8) | 85.9 (30.1) | F(2,45) = 2.9, P = 0.06, partial η 2 = 0.1 |
| Left omissions (/15) | 5.6 (4.4) * , b , ** , c | 2.7 (2.3) * , a | 1.2 (1.2) ** , a | F(2,45) = 7.4, P = 0.002, partial η 2 = 0.3 |
| Total number of omissions (/35) | 10.3 (7.5) * , b , *** , c | 4.8 (4.3) * , a | 2.0 (1.6) *** , a | F(2,45) = 9.3, P = 0.0004, partial η 2 = 0.3 |
| Catherine Bergego Scale | ||||
| Self-assessment (/30) | 7.2 (7.2) * , b , *** , c | 2.2 (3.4) * , a | 0.0 (0.0) *** , a | F(2,45) = 8.3, P = 0.0009, partial η 2 = 0.3 |
| Assessment by others (/30) | 14.7 (10.0) *** , b , *** , c | 0.2 (0.8) *** , a | 0.0 (0.0) *** , a | F(2,45) = 26.8, P = 2.2 10 −8 , partial η 2 = 0.5 |
| Auditory neglect | ||||
| Number of words found in the right ear (/30) | 25.8 (5.6) | 26.0 (5.8) | 22.7 (7.5) | F(2,45) = 1.2, P = 0.3, partial η 2 = 0.05 |
| Number of words found in the left ear (/30) | 5.6 (6.5) *** , b , *** , c | 26.6 (5.7) *** , a | 22.2 (9.9) *** , a | F(2,45) = 35.3, P = 1.2 10 −9 , partial η 2 = 0.6 |
| Asymetric cue | 69.9 (30.7) *** , b , *** , c | − 1.3 (5.3) *** , a | 20.0 (28.3) *** , a | F(2,45) = 33.7, P = 2.2 10 −9 , partial η 2 = 0.6 |
Related posts:
The Annals of Physical and Rehabilitation Medicine through the 2010s: A generalist journal of rehabilitation with a French touch
Bi-cephalic transcranial direct current stimulation combined with functional electrical stimulation for upper-limb stroke rehabilitation: A double-blind randomized controlled trial
New factors that affect quality of life in patients with aphasia
High sedentary behaviour and low physical activity levels at 12 months after cardiac rehabilitation: A prospective cohort study
Dosage for cost-effective exercise-based falls prevention programs for older people: A systematic review of economic evaluations
Could non-invasive brain stimulation help treat dysarthria? A single-case study
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
