Fear is a multifaceted emotion, challenging to define and assess accurately. Biologically, it is an innate response to threats, whereas psychologically, it is shaped by individual experiences and societal influences, sometimes evolving into specific phobias. One such phobia, arachnophobia (the fear of spiders), is particularly widespread and can vary widely in intensity among individuals. This paper presents a prototypal system that classifies the severity of arachnophobia using Machine Learning (ML) algorithms within a Virtual Reality (VR) game-based environment. The proposed system utilizes a two-stage clustering approach to analyze the behavioral data collected during VR exposure. Additionally, participants' self-reported fear levels are measured using the Spider Phobia Questionnaire (SPQ), to provide a comprehensive assessment. Preliminary results suggest that this method could be effectively used to classify arachnophobia intensity level, thus offering potential applications in both clinical settings and video games.

We present a study on stress detection in virtual reality using The Last Secret, a stealth game we developed that integrates advanced multimodal stress detection techniques and in-game behavioral metrics. Our game responds to the player's emotional state and stress levels, dynamically adapting its gameplay in real-time. Players need to manage their stress to navigate the game effectively: maintaining calmness keeps them in an optimal flow state, whereas excessive stress can push them into a frustrating state, making progression significantly more difficult. We present experimental results showing the impact of stress-modulated gameplay on player engagement and performance, revealing findings aligned with the flow theory.

In the realm of video games, achieving a state of flow is paramount for player engagement, striking a delicate balance between boredom and anxiety. Under such circumstances, monitoring player's stress level plays a crucial role. Here, by addressing state-space dynamics, we model stress unfolding over time, allowing for both its continuous and discrete assessment. This study builds upon our previous research, advancing stress assessment techniques to enhance tailored and immersive gaming experiences through Virtual Reality in horror games. We provide detailed insights into our collected data, which includes physiological measurements, motion behavioural data, and participants' self-reported stress levels. Additionally, we conduct an in-depth analysis on individual participants to delve deeper into the dynamics of stress experienced by each player. Results achieved give evidence that the measurement of motion behavioural data, exclusively collected from the headset, well compares to that based on electrodermal activity (EDA), more classically related to stress assessment. Data are made publicly available athttps://zenodo.org/records/15025199.

A critical challenge in the creation of Non-Playable Characters (NPCs) lies in the alignment between personality traits and emotions with the accurate portrayal of facial expressions. A mismatch between these two character’s aspects or the absence of effective synchronization between them could damage the believability of the game, with subsequent negative impacts on the player’s level of immersion and experience. This research introduces FACial EXpressions (FAC.EX), a software module crafted to enhance video game development by endowing NPCs with nuanced facial and internal responses during dialogues. Insights collected from feedback provided by 20 volunteers, including gamers and developers, affirm the perspective effectiveness of the module in simulating human-like facial expressions and emotions. The obtained positive results lay the foundation for potential future extensions aimed at meeting the dynamic requirements of the evolving game industry.

In this work, we explored the advantages of dynamic game balancing using players’ affective state introducing StrEx, a plugin designed to modulate the stress level induced by a video game in an unobtrusive way. The proposed plugin is also intended to validate results obtained by [1]. Our system collects motion behavioral data and updates a model of the player’s stress level to guide the transitions of a Finite State Machine which regulates the stress level induced by the game through the generation of game content. We designed and developed a virtual reality horror-survival game, with the aim of testing the system functioning. Our approach shows promising potential to create more immersive and engaging video games exploiting affective-based adaptation.