Abstract
This column attempts to describe the characteristics of current cyberpsychology research in Europe. In particular, CyberEurope aims at describing the leading research groups and projects running on the other side of the Ocean.
Bodily Self and Episodic Memory
The bodily self, which is the fundamental aspect of consciousness, may influence which is a higher cognitive aspects of self-representation, including the autobiographical self. 1 The sense of self is centered on the combination of multisensory, exteroceptive, and interoceptive bodily processing (e.g., touch, vestibular, visual, etc.) in the brain.1–3 A feature of the “I” experience is bodily self-consciousness (BSC)—conscious experiences such as self-location and body ownership, along with the first-person perspective. It is possible to study BSC in virtual reality (VR) with avatars by manipulating body ownership, agency (i.e., motor control over the avatar), and self-location (including visuospatial perspective).4,5
Episodic memory (EM) is a part of long-term memory that concerns the encoding, storage, and recollection of specific events. Often EMs are personally meaningful and related to our sense of self. 6 Research has shown that body ownership manipulation in VR during encoding enhanced the following EM retrieval measures: memory accuracy, immediate reliving, delayed emotional intensity, vividness, and belief in memory accuracy. 7 In another VR study, self-location was manipulated during encoding, and the researchers found that encoding events from a third-person perspective led to a deficit in EM retrieval. 6
Different aspects of the bodily self, including visuospatial perspective, self-location, and body ownership, are critical factors for successful retrieval/recognition and vivid and detailed episodic memories. 1 However, current research has examined visuospatial perspective taking and body ownership separately, and further research is needed to deepen the interaction between the BSC components.
VRBodyMem: A Research Project
Based on the above rationale, the VRBodyMem project will develop a framework that manipulates not only visuospatial perspective taking (self-location) during both encoding and retrieval, but also body ownership through synchronous and asynchronous visuotactile stimulation. The project has been developed as part of the Master “Cyberspace, Behavior and E-Therapy (CYBER),” supported by the Universidade Lusófona, the Universitat de Barcelona, and the Université Paris Cité, and one of the partner universities and laboratories, the Catholic University of Milan and the Applied Technology for Neuro-Psychology Laboratory at the IRCCS Istituto Auxologico Italiano. The results will be useful for understanding clinical conditions in which either/both encoding and retrieval are disembodied and for proposing innovative training solutions.
Methodology and measures
The project will use a within-subjects design to manipulate virtual embodiment via an avatar during encoding and retrieval of a recognition task. Stimuli will be selected from a standardized database. Participants will sit in a comfortable chair and will be assigned to either the first and third person encoding perspective; after this, a synchronous or asynchronous visuotactile stimulation will be delivered to foster embodiment and disembodiment in the avatar. The same procedure will be done for the recognition phase.
The study will use the remember–know (RK) 8 procedure. After the encoding phase, participants will select whether the stimuli are “new” or “old,” where “new” corresponds to the new stimuli added to the recognition phase and “old” corresponds to the stimuli also shown during the encoding session. In addition, if participants choose “old” during recognition, they will be asked to choose whether they “remember” or “know” the stimuli. Here, “remember” means “I vividly remember” and “know” means “I have a sense of familiarity with the stimuli.” This study will have a 2 × 2 × 2 × 2 design, which means that there will be 16 possible conditions. According to the hypotheses of the experiment, we will choose the most appropriate conditions. For example, we could compare embodiment in the avatar during encoding and recognition with a disembodied (third-person perspective with body ownership over the avatar) condition during encoding and recognition.
After each memory phase in VR, participants will be asked to complete questionnaires to assess their sense of presence, for which the study will use the Presence and Cybersickness Questionnaire (e.g., ITC-SOPI), and a Virtual Embodiment Questionnaire to assess embodiment in the avatar after synchronous or asynchronous stimulation. In addition, to document the effect of the conditions, reaction times for old/new and RK responses will be stored. Moreover, hits, misses, false positives, and false negatives will be stored, and we will compute d-prime (i.e., a measure of recognition sensitivity to old stimuli compared to newly added stimuli in the recognition portion).
Virtual environment
The virtual environment developed for the project (using Unity 2020.3.1f1 version) is a square room with an area of 5 × 5 virtual units (virtual meters). The avatar (male or female) sits on a chair facing a virtual screen. The participant perceives the virtual screen in front of him or her, viewing the stimuli from a first- or third-person perspective (using a Meta Quest 2 headset). The stimulus is a set of images loaded from a folder that is randomly presented to participants at the time of encoding and retrieval for random display on a virtual screen. The stimulus lists will be different in each encoding–recognition pair (e.g., for a two within-subjects design, two different stimulus lists will be used).
During the synchronous and asynchronous phase, which precedes the encoding or recognition phase, the participant will see the Oculus controller touching the virtual body, and at the same time, he/she will perceive a stick gently touching the real body, which may or may not be synchronized with the Oculus controller moving in the virtual world. An opaque curtain will hide the avatar's face from the third-person perspective, leaving a shadow of someone's face behind it to improve embodiment in the avatar. Avatars (male, female) wear black clothes and have no prominent details.
The interactive part of the program is divided into keyboard and mouse interaction controlled by the experimenter and retrieval interaction by the subject. The experimenter can activate encoding or retrieve the stimulus presentation by pressing a key on the keyboard, entering the subject's ID number, and showing the instructions to the participant. During the retrieval phase, the participant uses a controller to answer “if a stimulus is “new” or “old” or “remember” or “know”.
Conclusion
Building on previous literature, the project expects that disembodied conditions (e.g., third-person perspective with embodiment in the virtual avatar through synchronous stimulation) at encoding and recognition will lead to reduced memory performance compared to embodied conditions (first-person perspective with embodiment in the virtual avatar through synchronous stimulation). Furthermore, the developed framework allows the effects of mismatch conditions (e.g., disembodied at encoding and embodied at recognition, or vice versa) on memory to be studied.
As underlined by Daniele Di Lernia, researcher at the Catholic University of Milan, “The proposed methodology and the developed VR software may also help to advance the study of EM to better understand psychiatric—e.g., depersonalization—and neurological—e.g., amnesia—conditions in which the BSC may be distorted.”