The impact of positive, negative and neutral stimuli in a virtual reality cognitive-motor rehabilitation task: a pilot study with stroke patients
- Mónica S. Cameirão†1, 2Email author,
- Ana Lúcia Faria†2, 3,
- Teresa Paulino2,
- Júlio Alves1, 2 and
- Sergi Bermúdez i Badia1, 2
© The Author(s). 2016
Received: 7 December 2015
Accepted: 12 July 2016
Published: 9 August 2016
Virtual Reality (VR) based methods for stroke rehabilitation have mainly focused on motor rehabilitation, but there is increasing interest in integrating motor and cognitive training to increase similarity to real-world settings. Unfortunately, more research is needed for the definition of which type of content should be used in the design of these tools. One possibility is the use of emotional stimuli, which are known to enhance attentional processes. According to the Socioemotional Selectivity Theory, as people age, the emotional salience arises for positive and neutral, but not for negative stimuli.
For this study we developed a cognitive-motor VR task involving attention and short-term memory, and we investigated the impact of using emotional images of varying valence. The task consisted of finding a target image, shown for only two seconds, among fourteen neutral distractors, and selecting it through arm movements. After performing the VR task, a recall task took place and the patients had to identify the target images among a valence-matched number of distractors. Ten stroke patients participated in a within-subjects experiment with three conditions based on the valence of the images: positive, negative and neutral. Eye movements were recorded during VR task performance with an eye tracking system.
Our results show decreased attention for negative stimuli in the VR task performance when compared to neutral stimuli. The recall task shows significantly more wrongly identified images (false memories) for negative stimuli than for neutral. Regression and correlation analyses with the Montreal Cognitive Assessment and the Geriatric Depression Scale revealed differential effects of cognitive function and depressive symptomatology in the encoding and recall of positive, negative and neutral images. Further, eye movement data shows reduced search patterns for wrongly selected stimuli containing emotional content.
The results of this study suggest that it is feasible to use emotional content in a VR based cognitive-motor task for attention and memory training after stroke. Stroke survivors showed less attention towards negative information, exhibiting reduced visual search patterns and more false memories. We have also shown that the use of emotional stimuli in a VR task can provide additional information regarding patient’s mood and cognitive status.
KeywordsEmotional stimuli Stroke Virtual Reality Cognitive and motor rehabilitation Valence Eye tracking
According to the World Health Organization, fifteen million people worldwide suffer a stroke each year, leaving 5 million survivors permanently disabled. For those who survive, reducing the impact of post stroke impairment is a major goal . It has been estimated that more than 70 % of individuals experience some degree of cognitive decline in the first few weeks following stroke and that more than one third remain cognitively impaired even 1 year post stroke . These cognitive impairments have a direct influence on patients’ quality of life, being associated with greater rates of institutionalization  and higher health-care costs .
Among the most frequent sequels, post stroke patients commonly present decreased executive functioning, mental slowing, and impairment of goal formulation, initiation, planning, organizing, sequencing, executing, abstracting, and attention , being also at risk of developing dementia [2, 5]. Although several screening tools are available and are generally administered, specific deficits are only detectable with more complete neuropsychological assessments, which are rarely performed . Moreover, most assessments are paper-and-pencil based and are not performed in the context of meaningful real world tasks, thus they may miss important information. Consequently, there is a need to develop and employ more insightful assessment tools. This would allow the prescription of intensive cognitive and motor rehabilitation programs precisely tailored to the needs of patients, hence maximizing gains and transference of those to real-world tasks.
Virtual Reality (VR) as an assessment and rehabilitation tool
Recent research has shown that VR can be used in the assessment of motor and cognitive function by using simulations that relate to real-world skills [7–10]. Current assessment methods lack this aspect and evaluation in real-world context is costly and many times impracticable. Instead, VR has been applied to assess motor function in the context of ADLs  as well as cognitive functions in the context of a virtual city . In fact, VR neuropsychological assessment tools have been also validated against traditional methods [8, 12], holding promise for the future of neuropsychological assessment. Moreover, enriched virtual environments have the potential to optimize rehabilitation by manipulating practice conditions that explicitly engage motivational, cognitive, motor control and sensory feedback-based learning mechanisms . Unfortunately, most VR rehabilitation approaches are generally dedicated either to motor or cognitive rehabilitation aspects. Nevertheless, given the dual motor and cognitive components of ADL, a combined motor and cognitive VR approach could provide training more consistent with real-world settings . In fact, a recent meta-analysis identified a moderate association (r = 0.43) suggestive of interdependency between cognitive and motor recovery in stroke survivors . This has also been observed in a study with stroke survivors [16, 17] that used a VR adaptation of the widely used Toulouse Piéron (TP) cancellation task . Using VR adaptations of standardized assessment instruments is particularly interesting because it allows direct comparison with the paper-and-pencil counterpart.
Eye gaze during action execution in VR
There is strong evidence showing that motor areas are being engaged not only during motor execution but also during the observation of motor actions [19–21]. Eye gaze is closely linked to prediction and motor control, and has been used to study the neural mechanisms underlying the observation and execution of actions [22, 23]. Based on the premise of shared neural mechanisms for execution and observation, researchers have shown that action observation can have a positive impact in the rehabilitation of motor function after stroke [24, 25]. Recent studies on gaze metrics of healthy participants during action execution and observation have provided further evidence that, both execution and observation, partially share underlying neural processes . A study combining VR and eye tracking with stroke survivors and healthy participants has shown that movement metrics in the observation of a VR reaching task are sensitive to motor impairment . A previous study using the same system also identified differences in stroke survivors between action execution and observation, and between paretic and non-paretic arms during observation of motor actions but not during action execution . Given the emergence of novel low cost eye tracking devices, the combination of eye tracking technology and VR has large potential to be used in stroke rehabilitation to inform about underlying mechanisms during VR training.
The role of emotional stimuli in rehabilitation
Despite the wealth of evidence concerning the value of VR for rehabilitation of stroke patients , there is surprisingly little research about the type of content being used (neutral, abstract, emotional, etc.). One particularly interesting case is the use of emotional stimuli with different valences. Affective valence refers to the pleasantness of a given stimulus, with positive and negative valence indicating attractive and aversive stimuli, respectively. Stimuli of neutral valence are commonly perceived as having no or weak valence (positive or negative) .
From the literature with healthy participants we know that emotional stimuli are remembered better and more vividly than non-emotional stimuli . This phenomenon, known as the emotional enhancement of memory, has been replicated across a range of paradigms and stimulus types. These emotional enhancements in memory are, at least, partly due to the increased attention directed toward emotional items at encoding . However, emotional items are often remembered at the expense of their contexts, this is, peripheral features of visual scenes are remembered less when an emotional item is present in the scene than when only non-emotional items are present. The rationale behind this phenomenon is that items with a high affective valence tend to capture attention and to get prioritized in the processing chain . Since more attentional resources are directed towards the emotional components, people seem more likely to encode the emotional components of the scene and less likely to encode neutral contextual information.
The processing of emotional stimuli seems also to be affected by age . In fact, the Socioemotional Selectivity Theory states that there is an age-associated motivational shift towards emotional goals . This theory states that when emotional material is attended to, is weighed more heavily, processed more deeply, and better remembered than non-emotional material. Recent evidence shows that the recall of emotional information is disproportionately positive as people age . These results are consistent with eye tracking research that showed that when a negative and a neutral picture are displayed together, both young and older adults initially glance at the negative picture but young adults look for a longer time at the negative picture .
It has also been found that emotional stimuli can induce more false memories than non-emotional stimuli in healthy individuals, with stimuli of negative valence being more often falsely remembered when compared to stimuli of positive or neutral valence . The apparent paradox, that negative emotions can simultaneously improve and impair memory by its high valence, has been consistently found in memory recall experiments, inclusively in participants with major depression . One hypothesis is that negative valence causes a narrowing of attention such that, spatial and temporal information associated with the emotional item, are better attended to and later remembered, while peripheral information is likely to be forgotten. An example is the weapon focus effect, where there is enhanced memory for a weapon in a scene but reduced memory for details of the background . This focus may lead to selective memory for emotional components . However, some researchers argue that selective memory of emotional content is not strongly related to attention at encoding .
The valence of the pictures will have an effect in VR task performance, with lower performance for negative pictures and higher performance for positive pictures when compared to neutral targets.
The valence of the pictures will have an effect in recall, with negative pictures generating more false memories when compared to positive and neutral images.
As secondary objectives, we 1) study how the cognitive profile of stroke patients modulates performance in tasks using emotional content; and 2) assess the feasibility of the proposed VR rehabilitation paradigm that integrates both cognitive and motor domains for improved transference to real-world settings.
Participants were recruited at the Nélio Mendonça and João Almada Hospitals (Madeira Health Service, Portugal), based on the following inclusion criteria: ischemic stroke; normal or corrected-to-normal vision; capacity to be seated; non-aphasic and with sufficient cognitive ability to understand the task instructions, as assessed by the clinicians. The sample consisted of ten (7 female, 3 male) middle-aged (54.2 ± 9.2 years old) stroke survivors (1 right hemisphere, 8 left hemisphere and 1 cerebellum), at 16.6 ± 19.5 months post stroke, and 8.1 ± 5.8 years of schooling. 6 had some computer literacy. The study was approved by the Madeira Health Service - SESARAM Ethical Committee (approval number 47/2013) and all the participants gave their informed consent.
Participants were seated at a distance of ~55 cm from a 21.5 inch monitor with a total height of ~41 cm positioned on the table, and rested both arms on a tabletop. The task consisted on performing two-dimensional movements on the surface of the table with a single arm. These enabled participants with no force against gravity to perform the task. Most participants used their paretic arm to perform the task; participants without sufficient motor capacity used their unaffected arm (3/10 participants). The arm used for the interaction wore a colored glove that allowed real arm movements to be captured through a camera-based color tracking software (AnTS)  and mapped onto the movements of a virtual arm. The VR environment had a built-in calibration function to compute the active range of motion as described in . This calibration matches the maximum physical range of motion of the participant to the maximum range of motion required in the VR task, normalizing the required motor effort to the skill set of the user. The participant controls a virtual representation of the arm to complete the task.
The VR task was designed as a VR adaptation of the TP task (TP-VR) . This task was extended to also incorporate emotional stimuli. A target image of 21 × 21 cm2 is presented in the center of the screen. Immediately after, in a 3D environment, fifteen cubes with images are displayed in a 3x5 grid structure on top of a table. From these fifteen images, one is the target and fourteen are distractors, randomly selected from a set of 92 neutral valence images or from the set of 8 TP symbols.
Emotional stimuli pictures
The pictures were selected from the International Affective Picture System (IAPS) . This widely used picture set, that has also been used in studies with stroke survivors [44–46] , consists of photographs of people, animals, objects, and scenes that have been originally rated through the 9-points Self-Assessment Manikin (SAM) along the dimensions of affective valence (ranging from unpleasant to pleasant) and arousal (ranging from calm to excited). 182 images were selected for the purpose of this study. The categorization of the images as positive, negative and neutral was based on the original valence and arousal scores provided by the IAPS. Our selection of neutral, negative and positive images had valence scores in the range of 4.5–5.5, 1.66–2.58 and 7.53–8.34, respectively. A Friedman test confirmed that valences were significantly different across conditions (χ2 (2) = 28.0, p < 0.001). Arousal of the images was kept neutral with scores between 4.5 and 5.5.
Study design and protocol
A within-subjects design was used with three experimental conditions corresponding to 3 types of stimuli (positive, negative, and neutral). In addition, the abstract stimuli of the original TP task were also used for comparison with the paper-and-pencil TP task, since this is a well-established attention assessment tool.
Before starting the experiment, participants went through an average of five training trials only with abstract stimuli (TP-VR). The training was intended to provide a clear understanding of the VR task and valence rating, as well as to get used to the natural user interface (AnTS) .
Demographics and clinical profile of the participants
Months post stroke
Stroke side (L/R/C)
MoCA (Max = 30)
TP (Max = 100)
SIS strength (Max = 100)
SIS hand (Max = 100)
SIS communication (Max = 100)
SIS memory (Max = 100)
SIS recovery (Max = 100)
GDS-30 (Max = 30)
N = 10
54.2 ± 9.2
16.6 ± 19.5
17.4 ± 6.4
58.0 ± 33.9
56.3 ± 22.0
41.0 ± 34.9
78.6 ± 25.2
80.4 ± 12.1
61.1 ± 23.9
8.8 ± 6.4
From the VR task performance data, we extracted for each condition the percentage of correct target selection, resulting from the performance of both motor and cognitive components of the task; the mean task completion time per participant; and valence of the images as rated by the participants. From the recall task, we computed the recall performance as the percentage of correctly recalled images (true positives) per valence condition (neutral, positive, negative); the percentage of wrongly recalled images (false positives); and the total number of errors in the recall task (missed images/false negatives plus wrongly identified images/false positives). We tested for differences across conditions for the different dependent variables. To understand how performance in the VR and recall tasks could be explained as a function of the profile of the participants, we ran linear regressions for the results obtained under each condition, using the MoCA and GDS scores as factors in our models. In addition, the Pearson and Spearman coefficients were computed to search for meaningful correlations between variables for parametric and nonparametric data respectively.
Eye tracking data was temporally smoothed with a squared Savitzky-Golay FIR smoothing filter to data frames of length 29, and converted from screen coordinates (X, Y) to degrees. Resting periods and segments with missing or corrupted data were removed from the analysis. According to the velocity (v) profile of the data, eye movements were classified into 1) fixations (v < 5 deg/s); 2) saccadic movements (v > 30 deg/s), and 3) smooth pursuit (5 < v < 30 deg/s) consistent with [52, 53]. For each behavior detected, the number of occurrences and their duration were assessed. In addition, the total eye gaze trajectory length (in pixels) and its dispersion (standard deviation) were extracted. Heat maps to determine the density of the occurrence of the different eye gaze events were generated on a 10 × 10 grid of approximately 4.3° resolution centered at the target location, and smoothed using cubic interpolation.
For each variable, the normality of the distribution was assessed using the one sample Kolmogrov-Smirnov test. Because most distributions deviated from normality, non-parametric statistical tests were used for the analysis. For assessing the overall difference between experimental conditions, a Friedman test was used. For pairwise comparisons, the Wilcoxon’s T matched pairs signed ranks test was used for related measures, with a Bonferroni correction to account for the number of comparisons. One-tailed tests were used when hypotheses were directional. Data were analyzed using Matlab (MathWorks Inc., Natick, MA, USA) and the Statistical Package for the Social Sciences 20 (SPSS.20).
Primary objective - Effect of emotional content on VR task performance and eye gaze behavior
Does emotional content affect VR task performance?
Does emotional content impact eye gaze?
Median and (IQR) of eye tracking metrics for neutral, positive and negative valence
Fixation duration (ms)
Saccade duration (ms)
Smooth pursuit count
Smooth pursuit duration (ms)
Trajectory length (px)
Trajectory Std (px)
Is eye gaze modulated by the correctness of choices?
Concerning the length of eye gaze trajectories, we identified differences for trajectory length (Fr(2) = 8.0, p < 0.05) and trajectory dispersion (standard deviation) (Fr(2) = 6.0, p < 0.05). Trajectories were shorter for positive images (Mdn = 237.4px, IQR = 183.3) when compared to neutral (Mdn = 372.8px, IQR = 63.8) and negative images (Mdn = 303.3px, IQR = 70.8). However, only the negative-neutral medians were significantly different (T = 0.0, p < 0.05/3) (Fig. 7b). Finally, for the dispersion of the trajectory as measured by its standard deviation, although there was less variability for negative images (Mdn = 142.6px, IQR = 58.6) than for positive (Mdn = 185.1px, IQR = 79.6) and neutral images (Mdn = 204.6px, IQR = 23.8), these differences were not considered significant after Bonferroni correction (p > 0.05/3) (Fig. 7c).
Primary objective - Effect of emotional content on memory recall
Concerning the recall task, we did not find significant differences between the overall recall performance across conditions (Fr(2) = 0.0, p > 0.05). Interestingly, when considering only the wrongly identified images (false memories) we found significant differences across conditions (Fr(2) = 7.58, p < 0.05) (Fig. 3b). We hypothesized that there would be more false memories for negative images when compared to positive and neutral images. Planned comparisons revealed that there were significantly more false memories for negative images (Mdn = 6.2 %, IQR = 20.4) than for neutral images (Mdn = 0.0 %, IQR = 1.6) (T = 0.0, p < 0.05/2, one tailed). There was, however, no difference between positive (Mdn = 6.3 %, IQR = 14.1) and negative target images (T = 7.5, p > 0.05/2, one tailed).
Secondary objectives - Effect of the cognitive profile of patients on the performance in the VR and recall tasks
These models reveal a significant negative effect of depressive symptomatology as assessed by the GDS on task performance for images with positive valence, as well as a significant positive effect of general cognitive functioning as assessed by the MoCA on task performance for images with negative valence.
These models show specific effects of both depressive symptomatology and cognitive function for neutral images but not for images with positive or negative valence. Recall performance for neutral images is positively affected by cognitive function but negatively affected by depressive symptomatology. The opposite relationship is observed in the case of recall errors.
Analysis of correlations between scores in cognitive domains and performance in the VR and recall tasks
VR task performance
MoCA - Total
MoCA - Exec Funct
MoCA - Naming
MoCA - Attention
MoCA - Language
MoCA - Reasoning
MoCA - Memory
MoCA - Orientation
Secondary objectives - Feasibility of the proposed VR rehabilitation paradigm
Our VR task assumes that there is a correspondence between VR and paper-and-pencil counterparts, and that the emotional content of images is perceived consistently by healthy individuals and stroke survivors.
Is performance on paper equivalent to VR?
In the original TP paper-and-pencil version, targets are always visible during performance. The paradigm was slightly simplified in the VR version to address the memory component and make it compatible with the presentation of emotional pictures (there is only one target symbol per trial, which is only visible for 2 s). Consequently, a comparison of the TP paper-and-pencil performance (Mdn = 60.0 %, IQR = 65.0) with the TP-VR performance (Mdn = 82.2 %, IQR = 30.4) revealed significantly better performance in VR when compared to traditional TP (T = 5.0, p < 0.05).
Do the patients’ ratings of the images differ from those of the original IAPS ratings for healthy individuals?
Overall, there was no significant difference between the original IAPS ratings (5.02 ± 2.37) and the participants reported ratings (5.01 ± 1.78). Nevertheless, when we analyzed separately the 3 emotional categories, we found significant differences between the ratings for the positive (IAPS = 7.85 ± 0.27, Patients = 6.75 ± 0.86; Mann–Whitney, U = 15.0, p < 0.001) and negative (IAPS = 2.15 ± 0.29, Patients = 3.15 ± 1.10; U = 36.5, p = 0.01), but not for neutral stimuli (IAPS = 5.07 ± 0.26, Patients = 5.14 ± 1.02; U = 73.0, p > 0.05). This indicates a consistent but more moderate rating of the emotional content in the images by our patients than those originally provided by IAPS.
There is growing evidence of the existence of cognitive-motor interference, and that motor and cognitive recovery need to be considered together . Despite motor and cognitive stroke recovery patterns may differ, dual training has been shown to be effective for gait recovery , and there has also been shown a relationship of cognitive abilities such as reasoning and comprehension with functional motor performance, for the particular case of upper limb . Here we presented a novel VR rehabilitation task that integrates both cognitive and motor domains. In this first study, we only addressed the cognitive component of the task by investigating the impact of emotional stimuli in task performance and eye-gaze behavior.
In this study, we assessed the impact of emotional stimuli in a group of ten stroke survivors. We used 3 types of stimuli: positive, negative and neutral. Additionally, we used abstract stimuli from the TP task to compare the VR task with its paper-and-pencil counterpart. A higher performance has been found in the VR version of the task, which can be explained by the use of only one target image (a less demanding cognitive task) and a natural user interface to facilitate the interaction between the patient and the VR environment (adapted motor challenge), eliminating some of the constraints of a paper-and-pencil task. This prevents situations in which the motor deficits, for instance in the dominant arm, may impede proper execution of the paper-and-pencil task.
The emotional ratings of the target images by our patients were consistent with those of the IAPS, but showed less extreme ratings for positive and negative valence. Even though ratings were more moderate, image valence had a measurable effect on task performance. We identified lower VR task performance for negative images than for neutral. This is consistent with the positivity effect described by the Socioemotional Selectivity Theory, which postulates that attention processes are less targeted to negative information in older adults . Hence, given the average age of the general stroke population, this finding leads us to consider that positive and neutral content might be better for attention rehabilitation in this population.
In the case of the recall task, we observed that the lower performance for negative stimuli in VR was not replicated for overall performance during recall. These results are in accordance with the study by Steimetz and Kensinger  which shows that selective memory for emotional information is not strongly related to attention at encoding in healthy individuals. Interestingly, when we analyzed only the wrongly identified images -false positives- we found that the negative distractors led to significantly more mistakes. This finding is also consistent with the literature, with negative valence content causing more false memories , and a narrowing of attention at the cost of decreased memory for less salient details [38, 56]. This effect could also be observed in our study in several cases. For example, when having an image of a starving person as a target and then having two different images of a starving person in the recall task, the participant tended to select both, even if completely different. Overall, our findings lead us to conclude that using both positive and neutral stimuli will provide better results in the training of attention. Yet, if we want to increase task difficulty or memory challenge, we may consider negative stimuli since they are more difficult to remember and lead to more false memories. VR allows an easy customization depending on the objective and patient profile, which is more difficult to accomplish in traditional methods.
Another contribution of this study is the quantification of the relationship of patient profile -through the assessment of cognitive function and depressive symptomatology- and task performance depending on the emotional content of images. Our regression model analysis revealed that VR task performance for negative stimuli can be explained by the MoCA scores, being it a good metric for cognitive status. A positive relationship between cognitive function and performance with negative stimuli is consistent with increased difficulty of the task, as shown by VR task performance data. This finding is in agreement with our correlation analysis, which showed significant correlations of MoCA and multiple of its subdomains with overall VR performance. Of particular interest is the positive correlation of the memory subdomain of MoCA with VR performance for positive images. Further, a regression model showed that depressive symptomatology, as assessed by GDS-30, has a selective negative impact on VR task performance for positive stimuli. Consistent with the regression models, GDS-30 scores negatively correlated with performance for positive images. This finding is also in accordance with extensive literature that identified a general effect of negative mood in cognition in participants with mood disorders and depression [57, 58]. Interestingly, our models for the recall task capture effects of both cognitive function and depressive symptomatology, only for neutral stimuli and not for positive or negative. We also found a positive correlation of the memory subdomain of MoCA with recall performance, as well as a negative correlation with the number of errors for negative images. These data provide further evidence of distinct processing of emotional images in tasks involving attention, such as our VR task, in which the processing of negative stimuli is more strongly affected by overall cognitive function. Recall task performance correlates negatively with GDS-30 for negative images. Consequently, our data suggests that recall of positive stimuli is less modulated by cognitive function and depressive symptomatology, being it easier to recall than neutral or negative stimuli. This finding is also consistent with the shift towards emotional stimuli stated by the Socioemotional Selectivity Theory , and with the decline with age for negative stimuli . Altogether, these findings take us to consider that, besides a rehabilitation purpose, this VR task making use of neutral, negative and positive stimuli may also be valuable in providing information about the patient’s mood and cognitive status.
Fixations and saccades in our VR task are most likely related to a visual search component, whereas smooth pursuit segments may be more related to eye-hand coordination . Compared to previous eye tracking research on action observation and execution in VR with stroke survivors [23, 27], our data shows shorter fixations, smooth pursuits and longer saccades. These changes are probably associated to an increase of visual search patterns due to the existence of multiple distractors in our task. The eye tracking density maps show more disperse gaze patterns for neutral than for emotionally charged stimuli, supporting the premise of an attentional shift towards stimuli with emotional content. More concretely, our data shows differences in eye gaze metrics only for incorrect choices and not for correct ones. We observed lower count of saccades for positive stimuli, and shorter trajectories for negative stimuli. Hence, eye gaze metrics are suggestive of more determined and direct responses and with shorter search patterns for stimuli with positive or negative valence than for neutral. This information complements the available VR task performance data, which indicated more errors for negative stimuli. These data support the idea that errors in the VR task are the result of encoding the emotional content of the images, while neglecting peripheral non-emotional information.
In this study, distinct effects of overall cognitive function and mood were observed for images with neutral, positive and negative valence, for both attention and memory recall. These results contribute towards understanding how emotional content of images can be used on a VR paradigm for tailoring stimuli in cognitive-motor rehabilitation to each patient profile. However, a deeper understanding of the role of the motor component of the task needs to be developed. For this reason, we are currently investigating the effect of this VR task using positive valence in a 1-month randomized controlled longitudinal intervention, with pre and post motor assessment.
ADL, Activities of Daily Living; GDS, Geriatric Depression Scale; IAPS, International Affective Picture System; LB, Line Bisection; MoCA, Montreal Cognitive Assessment; SAM, Self-Assessment Manikin; SIS, Stroke Impact Scale; TP, Toulouse Piéron; VR, Virtual Reality
This work was supported by the European Commission through the RehabNet project - Neuroscience Based Interactive Systems for Motor Rehabilitation - EC (303891 RehabNet FP7-PEOPLE-2011-CIG); by the Fundação para a Ciência e Tecnologia (Portuguese Foundation for Science and Technology) through UID/EEA/50009/2013; and by the Agência Regional para o Desenvolvimento da Investigação, Tecnologia e Inovação (ARDITI) through Madeira 14-20.
Availability of data and material
The central tendency and dispersion data from which the conclusions are drawn are provided in the article. Raw data is available from the corresponding author on reasonable request.
MSC, ALF and SBB defined and designed the research study. TP implemented the algorithms, software and technical specifications underlying the experimental conditions. ALF and TP collected the data. ALF, JA, MSC and SBB analyzed the data. MSC, ALF and SBB interpreted the results. All authors revised and approved the current version of the manuscript.
The authors declare that they have no competing interests.
Consent for publication
Consent for publication of individual data has been obtained from all the participants of the study, including the participants in the images of the paper.
Ethics approval and consent to participate
This study was approved by the Madeira Health Service - SESARAM Ethical Committee (approval number 47/2013) and all the participants gave their informed consent.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Cumming TB, Marshall RS, Lazar RM. Stroke, cognitive deficits, and rehabilitation: still an incomplete picture. Int J Stroke. 2013;8:38–45.View ArticlePubMedGoogle Scholar
- Gottesman RF, Hillis AE. Predictors and assessment of cognitive dysfunction resulting from ischaemic stroke. Lancet Neurol. 2010;9:895–905.View ArticlePubMedPubMed CentralGoogle Scholar
- Pasquini M, Leys D, Rousseaux M, Pasquier F, Hénon H. Influence of cognitive impairment on the institutionalisation rate 3 years after a stroke. J Neurol Neurosurg Psychiatry. 2007;78:56–9.View ArticlePubMedGoogle Scholar
- Claesson L, Linden T, Skoog I, Blomstrand C. Cognitive impairment after stroke - Impact on activities of daily living and costs of care for elderly people - The Goteborg 70 + stroke study. Cerebrovasc Dis. 2005;19:102–9.View ArticlePubMedGoogle Scholar
- Desmond DW, Moroney JT, Sano M, Stern Y. Incidence of dementia after ischemic stroke results of a longitudinal study. Stroke. 2002;33:2254–62.View ArticlePubMedGoogle Scholar
- McClure J, Salter K, Foley N, Mahon H, Teasell R. Adherence to Canadian best practice recommendations for stroke care: vascular cognitive impairment screening and assessment practices in an ontario inpatient stroke rehabilitation facility. Top Stroke Rehabil. 2012;19:141–8.View ArticlePubMedGoogle Scholar
- Vourvopoulos A, Faria AL, Ponnam K, Bermúdez i Badia S. RehabCity: design and validation of a cognitive assessment and rehabilitation tool through gamified simulations of activities of daily living. Funchal: 11th Int. Conf. Adv. Comput. Entertain. Technol; 2014.View ArticleGoogle Scholar
- Buxbaum LJ, Dawson AM, Linsley D. Reliability and validity of the Virtual Reality Lateralized Attention Test in assessing hemispatial neglect in right-hemisphere stroke. Neuropsychology. 2012;26:430–41.View ArticlePubMedGoogle Scholar
- Josman N, Kizony R, Hof E, Goldenberg K, Weiss PL, Klinger E. Using the virtual action planning-supermarket for evaluating executive functions in people with stroke. J Stroke Cerebrovasc Dis. 2014;23:879–87.View ArticlePubMedGoogle Scholar
- Jansari AS, Devlin A, Agnew R, Akesson K, Murphy L, Leadbetter T. Ecological assessment of executive functions: a new virtual reality paradigm. Brain Impair. 2014;15:71–87.View ArticleGoogle Scholar
- Adams RJ, Lichter MD, Krepkovich ET, Ellington A, White M, Diamond PT. Assessing upper extremity motor function in practice of virtual activities of daily living. IEEE Trans Neural Syst Rehabil Eng. 2015;23:287–96.View ArticlePubMedGoogle Scholar
- Armstrong CM, Reger GM, Edwards J, Rizzo AA, Courtney CG, Parsons TD. Validity of the Virtual Reality Stroop Task (VRST) in active duty military. J Clin Exp Neuropsychol. 2013;35:113–23.View ArticlePubMedGoogle Scholar
- Levin MF, Weiss PL, Keshner EA. Emergence of virtual reality as a tool for upper limb rehabilitation: incorporation of motor control and motor learning principles. Phys Ther. 2015;95(3):415–25.View ArticlePubMedGoogle Scholar
- Kizony R, Zeilig G, Weiss PL, Baum-Cohen I, Bahat Y, Kodesh E, et al. Development of a real world simulation to study cognitive, locomotor and metabolic processes in older adults. Sweden: Proc 10th Intl Conf Disabil. Virtual Real. Assoc. Technol; 2013.Google Scholar
- Mullick AA, Subramanian SK, Levin MF. Emerging evidence of the association between cognitive deficits and arm motor recovery after stroke: a meta-analysis. Restor Neurol Neurosci. 2015;33:389–403.View ArticlePubMedPubMed CentralGoogle Scholar
- Faria AL, Vourvopoulos A, Cameirão MS, Fernandes JC, Bermúdez i Badia S. An integrative virtual reality cognitive-motor intervention approach in stroke rehabilitation: a pilot study. Gothenbg: 10th ICDVRAT; 2014. Sept 2–4 2014.Google Scholar
- Vourvopoulos A, Faria AL, Cameirão MS, Bermúdez i Badia S. Quantifying cognitive-motor interference in virtual reality training after stroke: the role of interfaces. Gothenbg: 10th ICDVRAT; 2014. Sept 2–4 2014.Google Scholar
- Piéron H. Metodologia psicotécnica. 1955.Google Scholar
- Buccino G, Binkofski F, Fink GR, Fadiga L, Fogassi L, Gallese V, et al. Action observation activates premotor and parietal areas in a somatotopic manner: an fMRI study. Eur J Neurosci. 2001;13:400–4.PubMedGoogle Scholar
- Dinstein I, Hasson U, Rubin N, Heeger DJ. Brain areas selective for both observed and executed movements. J Neurophysiol. 2007;98:1415–27.View ArticlePubMedPubMed CentralGoogle Scholar
- Filimon F, Rieth CA, Sereno MI, Cottrell GW. Observed, executed, and imagined action representations can be decoded from ventral and dorsal areas. Cereb Cortex. 2015;25:3144–58.View ArticlePubMedGoogle Scholar
- Brouwer AM, Franz VH, Gegenfurtner KR. Differences in fixations between grasping and viewing objects. J Vis. 2009;9:18–8.View ArticlePubMedGoogle Scholar
- Alves J, Vourvopoulos A, Bernardino A, i Badia SB. Eye Gaze Correlates of Motor Impairment in VR Observation of Motor Actions. Methods Inf Med. 2016;55(1):79-83.Google Scholar
- Mulder T. Motor imagery and action observation: cognitive tools for rehabilitation. J Neural Transm. 2007;114:1265–78.View ArticlePubMedPubMed CentralGoogle Scholar
- Ertelt D, Small S, Solodkin A, Dettmers C, McNamara A, Binkofski F, et al. Action observation has a positive impact on rehabilitation of motor deficits after stroke. Neuroimage. 2007;36(Supplement 2):T164–73.View ArticlePubMedGoogle Scholar
- Causer J, McCormick SA, Holmes PS. Congruency of gaze metrics in action, imagery and action observation. Front Hum Neurosci. 2013;7:604.Google Scholar
- Alves J, Vourvopoulos A, Bernardino A, Bermúdez i Badia S. Eye gaze patterns after stroke: correlates of a VR action execution and observation task. In: PervasiveHealth14 - 8th Int. Conf. Pervasive Comput. Technol. Healthc. 2014.Google Scholar
- Laver K, George S, Thomas S, Deutsch JE, Crotty M. Cochrane review: virtual reality for stroke rehabilitation. Eur J Phys Rehabil Med. 2012;48:523–30.PubMedGoogle Scholar
- Kensinger EA. Remembering emotional experiences: the contribution of valence and arousal. Rev Neurosci. 2004;15:241–52.View ArticlePubMedGoogle Scholar
- Talmi D, Anderson AK, Riggs L, Caplan JB, Moscovitch M. Immediate memory consequences of the effect of emotion on attention to pictures. Learn Mem. 2008;15:172–82.View ArticlePubMedPubMed CentralGoogle Scholar
- Schimmack U. Attentional interference effects of emotional pictures: threat, negativity, or arousal? Derryberry D, editor. Emotion. 2005;5:55–66.View ArticlePubMedGoogle Scholar
- Isaacowitz DM, Wadlinger HA, Goren D, Wilson HR. Selective preference in visual fixation away from negative images in old age? An eye-tracking study. Psychol Aging. 2006;21:40–8.View ArticlePubMedGoogle Scholar
- Carstensen LL, Fung HH, Charles ST. Socioemotional selectivity theory and the regulation of emotion in the second half of life. Motiv Emot. 2003;27:103–23.View ArticleGoogle Scholar
- Charles ST, Mather M, Carstensen LL. Aging and emotional memory: the forgettable nature of negative images for older adults. J Exp Psychol Gen. 2003;132:310–24.View ArticlePubMedGoogle Scholar
- Rösler A, Ulrich C, Billino J, Sterzer P, Weidauer S, Bernhardt T, et al. Effects of arousing emotional scenes on the distribution of visuospatial attention: changes with aging and early subcortical vascular dementia. J Neurol Sci. 2005;229–230:109–16.View ArticlePubMedGoogle Scholar
- Dehon H, Larøi F, Van der Linden M. Affective valence influences participant’s susceptibility to false memories and illusory recollection. Emotion. 2010;10:627–39.View ArticlePubMedGoogle Scholar
- Grassi-Oliveira R, Gomes CF de A, Stein LM. False recognition in women with a history of childhood emotional neglect and diagnose of recurrent major depression. Conscious Cogn. 2011;20:1127–34.View ArticlePubMedGoogle Scholar
- Loftus EF, Loftus GR, Messo J. Some facts about “weapon focus.”. Law Hum Behav. 1987;11:55–62.View ArticleGoogle Scholar
- Kensinger EA. Remembering the details: effects of emotion. Emot Rev. 2009;1:99–113.View ArticlePubMedPubMed CentralGoogle Scholar
- Steinmetz KRM, Kensinger EA. The emotion-induced memory trade-off: more than an effect of overt attention? Mem Cognit. 2013;41:69–81.View ArticlePubMedGoogle Scholar
- Mathews Z, i Badia SB, Verschure P. A novel brain-based approach for multi-modal multi-target tracking in a mixed reality space. In: Proc. 4th Intuit. Int. Conf. Workshop Virtual Real. 2007.Google Scholar
- Vourvopoulos A, Faria AL, Cameirão MS, Bermúdez i Badia S. RehabNet: a distributed architecture for motor and cognitive neuro-rehabilitation. understanding the human brain through virtual environment interaction. In: IEEE 15th Int. Conf. E-Health Netw. Appl. Serv. Heal. 2013.Google Scholar
- Lang P, Bradley M, Cuthbert B. International affective picture system (IAPS): affective ratings of pictures and instruction manual. Gainesville: University of Florida; 2008.Google Scholar
- Grabowska A, Marchewka A, Seniów J, Polanowska K, Jednoróg K, Królicki L, et al. Emotionally negative stimuli can overcome attentional deficits in patients with visuo-spatial hemineglect. Neuropsychologia. 2011;49:3327–37.View ArticlePubMedGoogle Scholar
- Oren N, Soroker N, Deouell LY. Immediate effects of exposure to positive and negative emotional stimuli on visual search characteristics in patients with unilateral neglect. Neuropsychologia. 2013;51:2729–39.View ArticlePubMedGoogle Scholar
- Buratto LG, Zimmermann N, Ferré P, Joanette Y, Fonseca RP, Stein LM. False memories to emotional stimuli are not equally affected in right- and left-brain-damaged stroke patients. Brain Cogn. 2014;90:181–94.View ArticlePubMedGoogle Scholar
- Freitas, Simões MR, Alves L, Santana I. Montreal Cognitive Assessment (MoCA): normative study for the Portuguese population. J Clin Exp Neuropsychol. 2011;33:989–96.View ArticlePubMedGoogle Scholar
- Hartman-Maeir A, Katz N. Validity of the Behavioral Inattention Test (BIT): Relationships With Functional Tasks. Am J Occup Ther. 1995;49:507–16.View ArticlePubMedGoogle Scholar
- Toulouse E, Piéron H, Pando AC. T-P: Toulouse-Piéron: (prueba perceptiva y de atención): manual. Tea; 2004.Google Scholar
- Duncan PW, Bode RK, Min Lai S, Perera S. Rasch analysis of a new stroke-specific outcome scale: the Stroke Impact Scale. Arch Phys Med Rehabil. 2003;84:950–63.View ArticlePubMedGoogle Scholar
- Barreto J, Leuschner A, Santos F, Sobral M. Geriatric depression scale. Lisbon Port: Tests Scales Dement. Group Study Brain Aging Dement; 2008.Google Scholar
- Holmqvist K, Nyström M, Andersson R, Dewhurst R, Jarodzka H, Weijer van de J. Eye Tracking: A comprehensive guide to methods and measures. Oxford: OUP; 2011.Google Scholar
- Ettinger U, Kumari V, Crawford TJ, Davis RE, Sharma T, Corr PJ. Reliability of smooth pursuit, fixation, and saccadic eye movements. Psychophysiology. 2003;40:620–8.View ArticlePubMedGoogle Scholar
- Pritchard C, Mayers A, Baldwin D. Changing patterns of neurological mortality in the 10 major developed countries--1979-2010. Public Health. 2013;127:357–68.View ArticlePubMedGoogle Scholar
- Fong KN, Chan CC, Au DK. Relationship of motor and cognitive abilities to functional performance in stroke rehabilitation. Brain Inj. 2001;15:443–53.View ArticlePubMedGoogle Scholar
- Bennion KA, Ford JH, Murray BD, Kensinger EA. Oversimplification in the study of emotional memory. J Int Neuropsychol Soc. 2013;19:953–61.View ArticlePubMedPubMed CentralGoogle Scholar
- Marvel CL, Paradiso S. Cognitive and neurological impairment in mood disorders. Psychiatr Clin North Am. 2004;27:19. viii.View ArticlePubMedPubMed CentralGoogle Scholar
- Chepenik LG, Cornew LA, Farah MJ. The influence of sad mood on cognition. Emotion. 2007;7:802–11.View ArticlePubMedGoogle Scholar
- Gauthier GM, Vercher J-L, Ivaldi FM, Marchetti E. Oculo-manual tracking of visual targets: control learning, coordination control and coordination model. Exp Brain Res. 1988;73:127–37.View ArticlePubMedGoogle Scholar