Papers by Guillaume Levieux
HAL (Le Centre pour la Communication Scientifique Directe), Nov 2, 2022
This paper describes an automated evaluation of the overall game experience using a synthetic age... more This paper describes an automated evaluation of the overall game experience using a synthetic agent, that we contextualize for First-Person Shooter games. This evaluation method is based on the characterization of the game experience through dynamics of major FPS games. We define dynamics as sequences of events that are meaningful for the player during the game session. As they trigger players' emotional responses, and influence their overall enjoyment and motivation, we classify them according to Motives for Play like curiosity, thrill-seeking, problem-solving, victory, and acquisition, in order to facilitate the evaluation process. Based on that, our evaluation method proposes to select synthetic agent routines that target a distinct game experience while playing a game session, using a selection of game dynamics. As the agent navigates through the level and interacts with opponents, dynamics may occur and, if so, are automatically identified, and then classified as Motives for Play. In the end, this classification can be used to evaluate the game experience and the quality of the level itself during playtesting sessions. It may also be utilized to assist the procedural generation of any level that target a specific game experience. CCS Concepts: • Human-centered computing → User models; • Applied computing → Computer games; Psychology.
HAL (Le Centre pour la Communication Scientifique Directe), Feb 17, 2022
In this paper, we study the link between diculty and player’s motivation in two games developed ... more In this paper, we study the link between diculty and player’s motivation in two games developed by Ubiso®: Rayman®Legends and Tom Clancy’s e Division®. We describe a method to estimate players’ diculty over time and link it’s time varying eect with players retention. Results conrm ow and self-ecacy theory. Also, for the rst hours of playtime, results dier between the two games. We explain that discrepancy with regard to aribution theory : in Rayman Legends, failure can be mainly aributed to the player skills, while in Tom Clancy’s e Division, avatar’s strength plays a fundamental role and can always be relatively quickly improved.
Proceedings of the 12th International Conference on the Foundations of Digital Games, 2017
In this paper, we study the link between di culty and player's motivation in two games developed ... more In this paper, we study the link between di culty and player's motivation in two games developed by Ubiso ®: Rayman®Legends and Tom Clancy's e Division®. We describe a method to estimate players' di culty over time and link it's time varying e ect with players retention. Results con rm ow and self-e cacy theory. Also, for the rst hours of playtime, results di er between the two games. We explain that discrepancy with regard to a ribution theory : in Rayman Legends, failure can be mainly a ributed to the player skills, while in Tom Clancy's e Division, avatar's strength plays a fundamental role and can always be relatively quickly improved.
Human-Computer Interaction - INTERACT 2017, 2017
This paper describes our research investigating the perception of difficulty in video games, defi... more This paper describes our research investigating the perception of difficulty in video games, defined as players' estimation of their chances of failure. We discuss our approach as it relates to psychophysical studies of subjective difficulty and to cognitive psychology research into the overconfidence effect. The starting point for our study was the assumption that the strong motivational pull of video games may lead players to become overconfident, and thereby underestimate their chances of failure. We design and implement a method for an experiment using three games, each representing a different type of difficulty, wherein players bet on their capacity to succeed. Our results confirm the existence of a gap between players' actual and self-evaluated chances of failure. Specifically, players seem to underestimate high levels of difficulty. The results do not show any influence on difficulty underestimation from the players gender, feelings of self-efficacy, risk aversion or gaming habits.
Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, 2019
Difficulty is one of the major motivational pull of video games, and thus many games use Dynamic ... more Difficulty is one of the major motivational pull of video games, and thus many games use Dynamic Difficulty Adjustment (DDA) systems to improve the game experience. This paper describes our research investigating the influence of DDA systems on player's confidence, evaluated using an in-game bet system. Our hypothesis is that DDA systems may lead players to overconfidence, revealed by an overestimation of their success chances when betting. This boost of confidence may be a part of the positive impact of DDA systems on the quality of game experience. We explain our method to evaluate player's confidence and implement it into three games related to logical, motor and sensory difficulties. We describe two experimental conditions where difficulty is either randomly chosen or adapted using a DDA algorithm. Results show how DDA systems can lead players to high level of overconfidence. CCS CONCEPTS • Applied computing → Computer games; Psychology; • Human-centered computing → User studies.
2016 IEEE Conference on Computational Intelligence and Games (CIG), 2016
This paper proposes a method to help understanding the influence of a game design on player reten... more This paper proposes a method to help understanding the influence of a game design on player retention. Using Far Cry R 4 data, we illustrate how playtime measures can be used to identify time periods where players are more likely to stop playing. First, we show that a benchmark can easily be performed for every game available on Steam using publicly available data. Then, we introduce how survival analysis can help to model the influence of game variables on player retention. Game environment and player characteristics change over time and tracking systems already store those changes. But existing model which deals with time varying covariate cannot scale on huge datasets produced by video game monitoring. That is why we propose a model that can both deal with time varying covariates and is well suited for big datasets. As a given game variable can have a changing effect over time, we also include time-varying coefficients in our model. We used this survival analysis model to quantify the effect of Far Cry 4 weapons usage on player retention.
Towards real-time interactive visualization modes of molecular surfaces: examples with udock
2015 IEEE 1st International Workshop on Virtual and Augmented Reality for Molecular Science (VARMS@IEEEVR), 2015
Lecture Notes in Computer Science, 2012
This paper presents a way to dynamically influence the shape and movements of a simulated crowd. ... more This paper presents a way to dynamically influence the shape and movements of a simulated crowd. We propose a tool and system that allows to modify a crowd's dynamics in an intuitive, semantically rich and out of context fashion, while being independent from the global path finding architecture and having a low computational cost. We follow a mixed approach where user-specified navigation fields are combined with steering and global A* pathfinding.
This paper is a first step toward the exploration and fur-ther development of the narrative possi... more This paper is a first step toward the exploration and fur-ther development of the narrative possibilities offered by emergent games. First, we present the five characteristics of emergent narratives that we believe to be fundamental: coherence, agency, possibility space, uncertainty and co-authoring. Then, we investigate the presence of these char-acteristics in various academic experiments. Finally, we de-scribe a novel approach that provides players with high level actions that they can perform in order to manipulate the story in real time.
Faraday Discuss., 2014
Protein–protein interactions play a crucial role in biological processes. Protein docking calcula... more Protein–protein interactions play a crucial role in biological processes. Protein docking calculations' goal is to predict, given two proteins of known structures, the associate conformation of the corresponding complex. Here, we present a new interactive protein docking system, Udock, that makes use of users' cognitive capabilities added up. In Udock, the users tackle simplified representations of protein structures and explore protein–protein interfaces’ conformational space using a gamified interactive docking system with on the fly scoring. We assumed that if given appropriate tools, a naïve user's cognitive capabilities could provide relevant data for (1) the prediction of correct interfaces in binary protein complexes and (2) the identification of the experimental partner in interaction among a set of decoys. To explore this approach experimentally, we conducted a preliminary two week long playtest where the registered users could perform a cross-docking on a datas...
VTX: High-performance molecular structure and dynamics visualization software
HAL (Le Centre pour la Communication Scientifique Directe), Jun 27, 2022
HAL (Le Centre pour la Communication Scientifique Directe), Jul 5, 2022
Training virtual agents to play a game using reinforcement learning (RL) has gained a lot of trac... more Training virtual agents to play a game using reinforcement learning (RL) has gained a lot of traction in recent years. Indeed, RL has delivered agents with superhuman performances on multiple gameplays. Yet, from a human-machine interaction standpoint, raw performance is not the only dimension of a "good" game AI. Exhibiting diversified behaviours is key to generate novelty, one of the core components of player engagement. In the RL framework, teaching agents to discover multiple strategies to achieve the same task is often framed as skill discovery. However, we observe that the current RL literature defines diversity as the exploration of different states, i.e. the incentive of the agent to "see" new observations. In this work, we argue that this definition does not make sense from a gameplay point of view. Instead, diversity should be defined as a distance on observations from an observer, external to the agent. We illustrate how DIAYN/SMERL, state of the art RL algorithms for skill discovery, fail to discover meaningful behaviours in a simple tag game. We propose an easy fix by introducing the notion of diversity spaces, defined as the observations gathered by a third-party external to the agent.
IEEE Transactions on Games
This paper presents Genetic-WFC, a procedural level generation algorithm that mixes genetic optim... more This paper presents Genetic-WFC, a procedural level generation algorithm that mixes genetic optimization with Wave Function Collapse, a local adjacency constraints propagation algorithm. We use a synthetic player to evaluate the novelty, safety and complexity of the generated levels. Novelty is maximized when the synthetic player goes on tiles not visited for a long time, safety is related to how far it can see, and complexity evaluates the variability of the surrounding tiles. WFC extracts constraints from example levels, and allows us to perform the genetic search on levels with few local asset placement errors, while using as little level design rules as possible. We show that we are able to rely on WFC while optimizing the results, first by influencing WFC asset selection and then by re-encoding the chosen modules back to our genotype, in order to optimize crossover. We compare the fitness curves and best maps of our method with other approaches. We then visually explore the kind of levels we are able to generate by sampling different values of safety and complexity, giving a glimpse of the variability that our approach is able to reach.
D.E.E.P. - Dialogue fondé sur les Emotions, l'Expérience et la Personnalité
HAL (Le Centre pour la Communication Scientifique Directe), 2007
Shape similarity computation is the main functionality for shape matching and shape retrieval sys... more Shape similarity computation is the main functionality for shape matching and shape retrieval systems. Existing shape similarity frameworks proceed by parameterizing shapes through the use of global and/or local representations computed in the 3D or 2D space. Up to now, global methods have demonstrated their rapidity, while local approaches offer slower, but more accurate solutions. This paper presents a shape similarity system driven by a global descriptor encoded as a Digital Elevation Model (DEM) associated to the input mesh. The DEM descriptor is obtained through the jointly use of a mesh flattening technique and a 2D panoramic projection. Experimental results on the public dataset TOSCA [BBK08] and a comparison with state-of-the-art methods illustrate the effectiveness of the proposed method in terms of accuracy and efficiency.
Proteins are macromolecules central to biological processes that display a dynamic and complex su... more Proteins are macromolecules central to biological processes that display a dynamic and complex surface. They display multiple conformations differing by local (residue side-chain) or global (loop or domain) structural changes which can impact drastically their global and local shape. Since the structure of proteins is linked to their function and the disruption of their interactions can lead to a disease state, it is of major importance to characterize their shape. In the present work, we report the performance in enrichment of six shape-retrieval methods (3D-FusionNet, GSGW, HAPT, DEM, SIWKS and WKS) on a 2 267 protein structures dataset generated for this protein shape retrieval track of SHREC'18.
Entertainment Computing and Serious Games, 2019
Difficulty is a fundamental factor of enjoyment and motivation in video games. Thus, many video g... more Difficulty is a fundamental factor of enjoyment and motivation in video games. Thus, many video games use Dynamic Difficulty Adjustment systems to provide players with an optimal level of challenge. However, many of these systems are either game specific, limited to a specific range of difficulties, or require much more data than one can track during a short play session. In this paper, we introduce the δ-logit algorithm. It can be used on many game types, allows a developer to set the game's difficulty to any level, with, in our experiment, a player failure error prediction rate lower than 20% in less than two minutes of playtime. In order to roughly estimate the difficulty as quickly as possible, δ-logit drives a single metavariable to adjust the game's difficulty. It starts with a simple +/-δ algorithm to gather a few data points and then uses logistic regression to estimate the players failure probability when the smallest required amount of data has been collected. The goal of this paper is to describe δ-logit and estimate its accuracy and convergence speed with a study on 37 participants playing a tank shooter game.
Human-Computer Interaction - INTERACT 2017, 2017
This paper investigates the perception of difficulty in video games, defined as the players' esti... more This paper investigates the perception of difficulty in video games, defined as the players' estimation of their chances of failure. We discuss our approach with regard to the psychophysical studies of subjective difficulty and to the cognitive psychology research on overconfidence bias. We assume that the strong motivational pull of video games may lead players to be overconfident and underestimate their chances of failure. Our method is tested within three games related to three types of difficulty, where the players have to bet on their capacity to win each challenge. Results confirm the existence of a gap between the players actual and self-evaluated chances of failure. More precisely, players seem to strongly underestimate high levels of difficulty. Results do not show any influence of the players gender, feeling of self-efficacy, risk aversion and gaming habits on the difficulty estimation error.
Designing tangible video games
Proceedings of the 32nd annual ACM conference on Human factors in computing systems - CHI '14, 2014
ABSTRACT In this paper, we present a collaborative game designed for Sifteo Cubes, a new tangible... more ABSTRACT In this paper, we present a collaborative game designed for Sifteo Cubes, a new tangible interface for multiplayer games. We discuss how this game exploits the platform's interface to transfer some of the game mechanics into the non-digital world, and how this approach affects both the player's experience and the design process. We present the technical limitations encountered during game development and analyze video recordings of play sessions with regard to the play strategies developed by the players. Then, we identify two properties that this game shares with many other games on tangible platforms and discuss how these properties influence both the game design process and the player experience. We advocate that these properties provide players with more freedom and relatedness, while helping to create an easy-to-learn and customizable gameplay, despite their own design limitations.
Uploads
Papers by Guillaume Levieux