=Paper= {{Paper |id=Vol-3425/paper5 |storemode=property |title= |pdfUrl=https://ceur-ws.org/Vol-3425/paper5.pdf |volume=Vol-3425 |authors=Victoria Brugada-Ramentol,Maiya Morgan,Hossein Jalali,Amir Bozorgzadeh |dblpUrl=https://dblp.org/rec/conf/mum/Brugada-Ramentol22 }} ==== https://ceur-ws.org/Vol-3425/paper5.pdf
Age-related Performance Decline Detected by a Virtual
Reality Multidomain Cognitive Training Solution (Enhance
VR)
Victòria Brugada-Ramentol1 , Maiya Morgan1 , Hossein Jalali1 and Amir Bozorgzadeh1
1
    Virtuleap, Portugal


                                             Abstract
                                             Cognitive training systems aim to improve specific domains or global cognition by engaging users in cognitively demanding
                                             tasks. While screen-based applications can improve performance in the trained cognitive abilities, they often result in poor
                                             transferability to activities of daily living. These systems exclude the user’s body and motor skills. Immersive Virtual Reality
                                             (IVR) systems, in contrast, present the user with body-related information, such as proprioceptive and visuomotor information,
                                             allowing for an immersive and embodied experience of the environment. This feature renders VR a very appealing tool
                                             for cognitive training and neurorehabilitation applications. Enhance VR is an IVR-based cognitive training and monitoring
                                             application that offers short daily cognitive workouts. The games are designed to train and monitor specific cognitive domains
                                             (memory, task flexibility, information processing, orientation, attention, problem-solving, and motor control). Enhance VR
                                             enables the collection of a large and varied behavioral dataset. Our analysis showed that age is a determinant of performance,
                                             as measured by the Enhance VR scoring system, with score games decreasing with age. Overall, suggesting that the Enhance
                                             VR app can potentially capture informative datasets that predict individual cognitive variables.

                                             Keywords
                                             Cognitive training, Cognitive monitoring, Immersive virtual reality, Age-related decline



1. Cognitive Training                                                                                                 ever, screen-based CCT solutions have failed to show
                                                                                                                      transfer to untrained cognitive categories and tasks and
Cognitive decline, the gradual deterioration of cognitive                                                             ADLs [3, 8]. Furthermore, computer-based interventions
abilities and functioning, progresses in parallel with age                                                            showed improvements in cognition and non-cognitive
and becomes more pronounced in populations at risk of                                                                 measures (e.g., mood), but failed to improve ADLs in
neurodegenerative diseases. The deterioration of cogni-                                                               people with dementia [9]. A 12-week program showed
tive abilities has a direct influence on the execution of                                                             improved capacities in early-stage Alzheimer’s disease
activities of daily living (ADLs) and negatively affects                                                              (AD), without positive effects on ADLs [10].
autonomy and well-being. A combination of a healthy                                                                      The lack of ecological validity in screen-based CCT
diet, moderate exercise, and cognitive stimulation have                                                               could, in part, explain why these systems do not result in
been proposed as strategies to attempt to slow down the                                                               a transfer of benefits to ADLs and general cognition. VR
progression of cognitive decline [1].                                                                                 systems, on the other hand, yield ecologically valid envi-
   Computerized cognitive training (CCT) has the po-                                                                  ronment scenarios with precise control over the experi-
tential to become a non-pharmacological intervention                                                                  mental variables [11, 12]. VR has shown to be potentially
to maintain the cognitive functioning of cognitively im-                                                              relevant for cognitive training in the elderly population,
paired individuals. CCT has been shown to reduce cog-                                                                 as a result of many different features, such as the flexibil-
nitive decline in healthy aging [2], but with no clear ef-                                                            ity of the environments and the possibility to gather rich
fectiveness in delaying difficulties in instrumental ADLs.                                                            data, as a result of increased immersion [13]. Ultimately,
CCT programs have shown an improvement in trained                                                                     improving cognitive functioning in healthy and at-risk
cognitive abilities (i.e., memory, reasoning, or speed-of-                                                            older adults and promoting the transfer of benefits.
processing) [3], which was maintained for 5 [4] or even
10 years [5]. A meta-analysis found a small and signifi-
cant effect on cognitive function in healthy older adults                                                             2. Virtual Reality Environments as
[6]. Overall, CCT has shown promising, but heteroge-                                                                     a Cognitive Training and
neous results in global cognitive functioning [7]. How-
                                                                                                                         Monitoring Tool
Workshop on Virtual Reality for Health and Wellbeing, 21st
International Conference on Mobile and Ubiquitous Multimedia,                                                         VR scenarios offer many advantages over screen-based
November 27th-30th, 2022, Lisbon                                                                                      cognitive training methodologies. Traditional method-
$ victoria@virtuleap.com (V. Brugada-Ramentol)                                                                        ologies have to trade external validity (i.e., how the task
                                       © 2023 Copyright for this paper by its authors. Use permitted under Creative
                                       Commons License Attribution 4.0 International (CC BY 4.0).                     accurately represents the measure in real-life scenarios)
    CEUR
    Workshop
    Proceedings
                  http://ceur-ws.org
                  ISSN 1613-0073
                                       CEUR Workshop Proceedings (CEUR-WS.org)
and the internal validity of conducting the tasks in a          experience is a gamified and engaging representation of
controlled laboratory setting with controlled variables         the neuropsychological validated tests. The Enhance VR
[14]. VR, on the other hand, offers realistic scenarios         is compatible with a wide range of commercially available
with a high degree of control over desirable experimental       headsets.
variables [11], such as the stimuli, and the presence of           The Enhance VR app consists of a library of games
distractors, among others. Additional advantages of VR          that engages seven main cognitive categories. As of Oc-
systems include the possibility of creating situations that     tober 2022, the library offers 15 games divided across
are adapted to the individual’s needs or that are impossi-      all of these skills: Pizza Builder, Whack-a-mole, Shuffled
ble to recreate in real-life scenarios and collecting precise   (Attention), React (Flexibility), Assembly and Harmonize
measurements of physical movements while sourcing               (Information processing), Memory Wall, Maestro, Magic
large amounts of behavioral data [14].                          Deck (Memory), Stacker and Odd Egg (Problem-solving),
   Moreover, immersive VR (IVR) systems that are dis-           Slinger and Balance (Motor Skills), and Orbital and Hide
played through a head-mounted display with embedded             and Seek (Spatial Orientation) ([21], Table 1).
head-tracking, which updates the environment according             The design of each individual Enhance VR game can be
to the participant’s movement, allow for a naturalistic         compared to other workflows proposed for the design of
interaction with the environment. In these scenarios, the       cognitive assessment exercises [22]. Whenever the need
behavior of the environment matches the expected phys-          for a specific cognitive category is identified, different
ical and motor consequences of the participant’s action.        classical paradigms are selected and evaluated to maxi-
This is also referred to as the plausibility illusion. Fur-     mize the training abilities in an IVR environment. Once
thermore, IVR environments engage the sensorimotor              the candidate paradigm has been identified, the extensive
system enhancing the illusion of embodiment over a vir-         literature review identifies task requirements. Namely,
tual body or body part [15, 16]. By providing immersive         (1) the mechanics of the paradigm, (2) the actions that are
experiences of the scenarios, virtual environments pro-         required from the participant, (3) the category and type
vide the feeling that the virtual world is the real world       of stimuli, (4) the outcome measures that evaluate the
and provide the feeling of actually “being” in the vir-         specific cognitive ability, and (5) identify the parameters
tual environment (i.e., placement illusion). The feeling        that can be modified to increase the difficulty of the task.
of presence in a virtual environment has been shown to          These parameters will define the design of the gamified
increase motivation for learning and attention to the task      virtual reality exercise, developed by a team of game
[17].                                                           designers, developers, artists, and quality assessment.
   Sensorimotor contingencies are, thus, important for             The difficulty of the Enhance VR app is dependent
immersion. VR systems are particularly interesting for          and controlled by adjusting the relevant parameters for
this matter, as they allow for the integration of proprio-      each exercise. The Enhance VR app tests the baseline
ceptive, visual, and motor information [18]. Ultimately,        performance for each game by presenting a first longer
VR scenarios provide a high degree of realism that elicit       session (hereafter, benchmark session). From there on,
naturalistic behaviors from participants [19]. The natural-     every session starts at the level where the user left off.
istic interaction with the environment provides a higher
degree of ecological validity than current screen-based
or pen-and-paper solutions. As a result, VR systems be-         4. Enhance VR Enables the
come interesting tools for the training and rehabilitation         Collection of Behavioral Data
of cognitive functions in distinct populations [20].
                                                                   for the Detection of Cognitive
                                                                   Performance
3. Developing a Cognitive
   Training and Monitoring                               Individual progression is tracked by the Enhance VR
                                                         Performance Index (EPI), calculated as an aggregate of
   Solution in VR                                        weighted performance across all cognitive categories in
                                                         addition to scores for each game, the main cognitive
The Enhance VR app (https://virtuleap.com/download/)
                                                         categories, and subcategories scores. The app collects
is designed to be played regularly in the form of short
                                                         data on self-reported mood and sleeping hours at the start
workouts. Each workout is composed of three randomly
                                                         of each workout. Upon registration, users are required
chosen games. The mechanics of each game is motivated
                                                         to input demographic data, such as age. The Enhance VR
by the mechanics of validated neuropsychological princi-
                                                         app is compliant with GDPR and HIPAA regulations.
ples. Every game stems from the collaboration between
                                                            IVR systems enable the collection of large and varied
scientists and game designers to ensure that the mechan-
                                                         behavioral datasets [13]. As the participants engage with
ics of each test are maintained. At the same time, the
                                                         the Enhance VR games, the system collects not only the
performance of the participants but also a large number
of game-related variables. Some of these variables in-
clude motor outputs (e.g., movement coordinates or hand
preferences), and reaction times, among others. Thus,
the Enhance VR app collects behavioral data obtained
by game-specific events. These variables would provide
a sensitive description of the individual cognitive sta-
tus and allow for a self-to-self comparison of cognitive
changes. Thus, we propose that the behavioral data from
Enhance VR could help identify age-related cognitive
decline.

4.1. Enhance VR Scores are Affected by
     Age
The Enhance VR app is available commercially in VR
stores, such as HTC Vive Port and Oculus store. Data
collected from these users is anonymized. Users with non-
plausible age ranges (>100 years old) or a combination
of age and education (e.g. those who reported being 12
years old with a Master’s degree) were removed from the       Figure 1: Median benchmark game scores organized by age
analysis. Age groups above 80 were excluded from the          groups.
analysis due to insufficient data points for all the games.
The analysis was performed on a sample of 21,230 users.
Furthermore, only data from the benchmark sessions
were selected.                                                ful co-design between scientific and development teams.
   We observe a decreasing trend in the cognitive scores      Age-related cognitive decline can also be related to an
measured by the Enhance VR app with increased age             increasing prevalence of systemic disorders that are also
groups. The highest median value was obtained by the          associated with cognitive impairment.
groups 11-20 and 21-30. The lowest median score was              The advantages of IVR are not only restricted to cogni-
observed in the 71-80 group (Figure 1).                       tive training but also to provide immersive scenarios for
                                                              cognitive assessment [24]. As a result of the naturalistic
                                                              interaction and realistic scenarios, VR cognitive assess-
5. Discussion                                                 ment offers the opportunity to accurately measure the
                                                              individual’s cognitive performance. Digital biomarkers
IVR scenarios offer an ecologically valid environment         are objective, quantifiable physiological and behavioral
that can potentially improve global cognitive functions       data that are collected by digital devices. Previous stud-
and have the potential to transfer benefits to untrained      ies have found a positive correlation between behavioral
cognitive abilities. The Enhance VR app takes advantage       data obtained from gameplay patterns from specialized
of the immersive and ecological validity of VR environ-       and off-the-shelf games such as mouse clicks and cogni-
ments to provide a structured and controlled setting for      tive function [25, 26]. Thus, these data could be used to
cognitive training. The Enhance VR app offers mentally        monitor biological or pathogenic processes and assess
challenging exercises that aim to train specific cogni-       responses to pharmacological interventions [27].
tive skills presented in a structured manner and extracts        The Enhance VR collects longitudinal behavioral infor-
behavioral measures of cognitive fitness [23].                mation that, when combined with supervised machine
   Increasing age has been previously related to a slow       learning algorithms and clinically relevant data, could
decline in cognitive abilities. Our preliminary analysis      provide novel non-invasive digital biomarkers of cogni-
shows that the Enhance VR app scoring system reflects         tive status. Ultimately, the ability to identify patterns
an age-related trend of performance decline. Thus, sug-       of impaired cognitive and motor performance at a pre-
gesting that the Enhance VR app has the potential to          clinical stage could act as the earliest manifestations of
describe the individual profile. The next steps include the   cognitive impairment and would be key in the develop-
analysis of behavioral data. Evaluation of the Enhance        ment of prevention and treatment strategies. Integrating
VR app will require confirmation of the face and content      these digital biomarkers with current diagnostic methods
validity of the tests since the inclusion of a strong motor   could provide an earlier, more sensitive, and specific de-
component may add confounding variables despite care-         tection of cognitive changes, as well as monitor disease
progression and act as proxies for the efficacy of phar-           reality tools: What we know so far, Brain Sciences
macological interventions in clinical trials for disease-          11 (2021) 528.
modifying drugs, as well as side effects.                      [8] A. M. Owen, A. Hampshire, J. A. Grahn, R. Stenton,
   The potential application of Enhance VR is not exclu-           S. Dajani, A. S. Burns, R. J. Howard, C. G. Ballard,
sive to age-related diseases cognitive impairment is a             Putting brain training to the test, Nature 465 (2010)
hallmark of several diseases and conditions, including             775–778.
ADHD, traumatic brain injury, multiple sclerosis, long         [9] J. A. García-Casal, A. Loizeau, E. Csipke, M. Franco-
COVID, and even a common side effect of chemotherapy.              Martín, M. V. Perea-Bartolomé, M. Orrell,
The large diversity of areas covered by our games en-              Computer-based cognitive interventions for people
ables the creation of curated playlists for these different        living with dementia: a systematic literature
clinical indications, maximizing our reach and impact.             review and meta-analysis, Aging & mental health
                                                                   21 (2017) 454–467.
                                                              [10] B. S. A. Polat, O. Karadas, Evaluation of a computer-
References                                                         based cognitive training program for early-stage
                                                                   alzheimer’s disease, Ann Clin Analyt Med 13 (2022)
 [1] T. Ngandu, J. Lehtisalo, A. Solomon, E. Levälahti,
                                                                   175–179.
     S. Ahtiluoto, R. Antikainen, L. Bäckman, T. Hänni-
                                                              [11] C. J. Bohil, B. Alicea, F. A. Biocca, Virtual reality in
     nen, A. Jula, T. Laatikainen, et al., A 2 year mul-
                                                                   neuroscience research and therapy, Nature reviews
     tidomain intervention of diet, exercise, cognitive
                                                                   neuroscience 12 (2011) 752–762.
     training, and vascular risk monitoring versus con-
                                                              [12] T. D. Parsons, Virtual reality for enhanced ecologi-
     trol to prevent cognitive decline in at-risk elderly
                                                                   cal validity and experimental control in the clinical,
     people (finger): a randomised controlled trial, The
                                                                   affective and social neurosciences, Frontiers in hu-
     Lancet 385 (2015) 2255–2263.
                                                                   man neuroscience 9 (2015) 660.
 [2] C. Hertzog, A. F. Kramer, R. S. Wilson, U. Linden-
                                                              [13] A. C. M. Bauer, G. Andringa, The potential of im-
     berger, Enrichment effects on adult cognitive devel-
                                                                   mersive virtual reality for cognitive training in el-
     opment: can the functional capacity of older adults
                                                                   derly, Gerontology 66 (2020) 614–623.
     be preserved and enhanced?, Psychological science
                                                              [14] A. J. Martingano, S. Persky, Virtual reality expands
     in the public interest 9 (2008) 1–65.
                                                                   the toolkit for conducting health psychology re-
 [3] K. Ball, D. B. Berch, K. F. Helmers, J. B. Jobe, M. D.
                                                                   search, Social and personality psychology compass
     Leveck, M. Marsiske, J. N. Morris, G. W. Rebok, D. M.
                                                                   15 (2021) e12606.
     Smith, S. L. Tennstedt, et al., Effects of cognitive
                                                              [15] V. Brugada-Ramentol, I. Clemens, G. G. de Polavieja,
     training interventions with older adults: a random-
                                                                   Active control as evidence in favor of sense of own-
     ized controlled trial, Jama 288 (2002) 2271–2281.
                                                                   ership in the moving virtual hand illusion, Con-
 [4] S. L. Willis, S. L. Tennstedt, M. Marsiske, K. Ball,
                                                                   sciousness and cognition 71 (2019) 123–135.
     J. Elias, K. M. Koepke, J. N. Morris, G. W. Rebok,
                                                              [16] M. V. Sanchez-Vives, B. Spanlang, A. Frisoli,
     F. W. Unverzagt, A. M. Stoddard, et al., Long-term
                                                                   M. Bergamasco, M. Slater, Virtual hand illusion
     effects of cognitive training on everyday functional
                                                                   induced by visuomotor correlations, PloS one 5
     outcomes in older adults, Jama 296 (2006) 2805–
                                                                   (2010) e10381.
     2814.
                                                              [17] J. N. Bailenson, N. Yee, J. Blascovich, A. C. Beall,
 [5] G. W. Rebok, K. Ball, L. T. Guey, R. N. Jones, H.-
                                                                   N. Lundblad, M. Jin, The use of immersive virtual
     Y. Kim, J. W. King, M. Marsiske, J. N. Morris, S. L.
                                                                   reality in the learning sciences: Digital transforma-
     Tennstedt, F. W. Unverzagt, et al., Ten-year effects
                                                                   tions of teachers, students, and social context, The
     of the advanced cognitive training for independent
                                                                   Journal of the Learning Sciences 17 (2008) 102–141.
     and vital elderly cognitive training trial on cogni-
                                                              [18] M. V. Sanchez-Vives, M. Slater, From presence to
     tion and everyday functioning in older adults, Jour-
                                                                   consciousness through virtual reality, Nature Re-
     nal of the American Geriatrics Society 62 (2014)
                                                                   views Neuroscience 6 (2005) 332–339.
     16–24.
                                                              [19] G. Riva, B. K. Wiederhold, F. Mantovani, Neuro-
 [6] A. Lampit, H. Hallock, M. Valenzuela, Computer-
                                                                   science of virtual reality: from virtual exposure to
     ized cognitive training in cognitively healthy older
                                                                   embodied medicine, Cyberpsychology, behavior,
     adults: a systematic review and meta-analysis of
                                                                   and social networking 22 (2019) 82–96.
     effect modifiers, PLoS medicine 11 (2014) e1001756.
                                                              [20] D. Perez-Marcos, Virtual reality experiences, em-
 [7] S. Lasaponara, F. Marson, F. Doricchi, M. Cavallo,
                                                                   bodiment, videogames and their dimensions in neu-
     A scoping review of cognitive training in neurode-
                                                                   rorehabilitation, Journal of neuroengineering and
     generative diseases via computerized and virtual
                                                                   rehabilitation 15 (2018) 1–8.
                                                              [21] V. Brugada-Ramentol, A. Bozorgzadeh, H. Jalali,
     Enhance vr: A multisensory approach to cognitive
     training and monitoring, Frontiers in Digital Health
     4 (2022).
[22] S. Krohn, J. Tromp, E. M. Quinque, J. Belger,
     F. Klotzsche, S. Rekers, P. Chojecki, J. de Mooij,
     M. Akbal, C. McCall, et al., Multidimensional eval-
     uation of virtual reality paradigms in clinical neu-
     ropsychology: Application of the vr-check frame-
     work, Journal of Medical Internet Research 22 (2020)
     e16724.
[23] G. Sala, N. D. Aksayli, K. S. Tatlidil, T. Tatsumi,
     Y. Gondo, F. Gobet, Near and far transfer in cogni-
     tive training: A second-order meta-analysis, Col-
     labra: Psychology 5 (2019).
[24] P. Kourtesis, S. Collina, L. A. Doumas, S. E. MacPher-
     son, Validation of the virtual reality everyday as-
     sessment lab (vr-eal): An immersive virtual reality
     neuropsychological battery with enhanced ecologi-
     cal validity, Journal of the International Neuropsy-
     chological Society 27 (2021) 181–196.
[25] P. Dagum, Digital biomarkers of cognitive function,
     NPJ digital medicine 1 (2018) 10.
[26] R. L. Mandryk, M. V. Birk, The potential of
     game-based digital biomarkers for modeling mental
     health, JMIR mental health 6 (2019) e13485.
[27] S. Vasudevan, A. Saha, M. E. Tarver, B. Patel, Digital
     biomarkers: Convergence of digital health technolo-
     gies and biomarkers, NPJ digital medicine 5 (2022)
     36.