Many videogames require players to perform under pressure; however, not all players respond equivalently to pressure: why are some players more likely to tilt (lose control during play) or choke (perform poorly relative to their ability) whereas others seem to thrive under pressure? Given the importance of both emotion regulation in tilting and optimal arousal in achieving optimal performance, we propose that individual differences in ability to down-regulate negative affect under stress--known as failure-related action-state orientation (fASO)--could explain propensity to choke under pressure. We conducted an online between-subjects experiment (N=144) in which we measured baseline performance in Super Hexagon (day 1), then exposed participants to a stress induction (i.e., PASAT-C) or had them play a low-intensity bubble-popping game before playing again (day 2). Under stress, players higher in fASO performed better relative to their baseline in terms of average time alive and stalled progress; whereas, without stress, players lower in fASO performed better on both measures. Traits reflective of proposed explanations for choking (i.e., reinvestment, attentional control) did not influence performance under pressure. The ability to down-regulate negative affect and overcome setbacks is a useful theoretical lens to explore why some players choke under pressure, whereas others thrive.

Developing skill and overcoming in-game challenges is of great interest to both players and game designers. Players can improve through repetition, but sometimes practice does not lead to improvement and progress stalls. It would be useful if designers could help players make progress without compromising their long-term skill development. We carried out a study to investigate how two techniques—checkpoints and breaks—affect in-game progress and player skill. Checkpoints allow multiple attempts at a challenge without having to repeat earlier sections; this aids progress, but could potentially hinder skill development. Second, breaks in gameplay have been shown to accelerate skill development, but their effectiveness is unknown when the breaks are integrated into the game's design. Our study evaluated the effects of game-integrated breaks and checkpoints on players' in-game progress (when the techniques were present) as well as two test sessions (with all techniques removed). Our results showed that both checkpoints and breaks aid progress (combining both had the largest effect) and that neither technique reduced performance in the transfer task, suggesting that skill development was not hindered. Our work provides evidence that checkpoints and breaks are valuable techniques that can assist both player progress and skill.

Socializing is an important part of why people choose to play games and is at the core of many game mechanics. Anxiety and fear about social interactions can lead to withdrawal from socializing in the physical world, yet players with social anxiety preferentially choose MMORPGs — a highly social genre — raising questions of whether social anxiety expresses differently during in-game interactions. In the present study (N=181), we explore whether and how social anxiety translates into MMORPGs. By developing a measure of in-game social anxiety, we find that although fear and apprehension of socializing in the physical and game worlds are related, they differently affect preferences, behaviours, and experiences. Social anxiety in the physical world drives reasons for playing, whereas in-game anxiety affects behaviours, reducing participation in activities related to socializing and difficult in-game challenges. Our findings can inform the design of social games and the links between social anxiety and social gaming.

Online social interactions in multiplayer games can be supportive and positive or toxic and harmful; however, few methods can easily assess interpersonal interaction quality in games. We use behavioural traces to predict affiliation between dyadic strangers, facilitated through their social interactions in an online gaming setting. We collected audio, video, in-game, and self-report data from 23 dyads, extracted 75 features, trained Random Forest and Support Vector Machine models, and evaluated their performance predicting binary (high/low) as well as continuous affiliation toward a partner. The models can predict both binary and continuous affiliation with up to 79.1% accuracy (F1) and 20.1% explained variance (R2) on unseen data, with features based on verbal communication demonstrating the highest potential. Our findings can inform the design of multiplayer games and game communities, and guide the development of systems for matchmaking and mitigating toxic behaviour in online games.

Digital games require players to learn various skills, which is often accomplished through play itself. In multiplayer games, novices can feel overwhelmed if competing against better players, and can fail to improve, which may lead to unsatisfying play and missed social play opportunities. To help novices learn the requisite skills, we first determined how experts accomplish an important task in multiplayer FPS games -- locating their opponent. After determining that an understanding of audio cues and how to leverage them was critical, we designed and evaluated two systems for introducing this skill of locating opponents through audio cues -- a training system, and a modified game interface. We found that both systems improved accuracy and confidence, but that the training system led to more audio cues being recognized. Our work may help people of disparate skill play together, by scaffolding novices to learn and use a strategy commonly employed by experts.