Wednesday, February 17, 2021

Video game play is positively correlated with well-being

Video game play is positively correlated with well-being. Niklas Johannes, Matti Vuorre and Andrew K. Przybylski. Royal Society Open Science, February 17 2021. https://doi.org/10.1098/rsos.202049

Abstract: People have never played more video games, and many stakeholders are worried that this activity might be bad for players. So far, research has not had adequate data to test whether these worries are justified and if policymakers should act to regulate video game play time. We attempt to provide much-needed evidence with adequate data. Whereas previous research had to rely on self-reported play behaviour, we collaborated with two games companies, Electronic Arts and Nintendo of America, to obtain players' actual play behaviour. We surveyed players of Plantsvs.Zombies: Battle for Neighborville and Animal Crossing: New Horizons for their well-being, motivations and need satisfaction during play, and merged their responses with telemetry data (i.e. logged game play). Contrary to many fears that excessive play time will lead to addiction and poor mental health, we found a small positive relation between game play and affective well-being. Need satisfaction and motivations during play did not interact with play time but were instead independently related to well-being. Our results advance the field in two important ways. First, we show that collaborations with industry partners can be done to high academic standards in an ethical and transparent fashion. Second, we deliver much-needed evidence to policymakers on the link between play and mental health.

4. Discussion

How is video game play related to the mental health of players? This question is at the heart of the debate on how policymakers will act to promote or to restrict games’ place in our lives [7]. Research investigating that question has almost exclusively relied on self-reports of play behaviour, which are known to be inaccurate (e.g. [8]). Consequently, we lack evidence on the relation between play time and mental health that is needed to inform policy decisions. To obtain reliable and accurate play data, researchers must collaborate with industry partners. Here, we aimed to address these shortcomings in measurement and report a collaboration with two games companies, Electronic Arts and Nintendo of America, combining objective measures of game behaviour (i.e. telemetry) with self-reports (i.e. survey) for two games: Plantsvs.Zombies: Battle for Neighborville and Animal Crossing: New Horizons. We also explored whether the relation between play time and well-being varies with players' need satisfaction and motivations. We found a small positive relation between play time and well-being for both games. We did not find evidence that this relation was moderated by need satisfactions and motivations, but that need satisfaction and motivations were related to well-being in their own right. Overall, our findings suggest that regulating video games, on the basis of time, might not bring the benefits many might expect, though the correlational nature of the data limits that conclusion.

Our goal was to investigate the relation between play time, as a measure of actual play behaviour, and subjective well-being. We found that relying on objective measures is necessary to assess play time: although there was overlap between the amount of time participants estimated to have played and their actual play time as logged by the game companies, that relation was far from perfect. On average, players overestimated their play time by 0.5 to 1.6 hours. The size of that relation and the general trend to overestimate such technology use are in line with the literature, which shows similar trends for internet use [24] and smartphone use [8,23]. Therefore, when researchers rely on self-reports of play behaviour to test relations with mental health, measurement error and potential bias will necessarily lead to inaccurate estimates of true relationships. Previous work has shown that using self-reports instead of objective measures of technology use can both inflate [45,46] or deflate effects [44]. In our study, associations between objective play time and well-being were larger than those between self-reported play time and well-being. Had we relied on self-reports only, we could have missed a potentially meaningful association.

Players who objectively played more in the past two weeks also reported to experience higher well-being. This association aligns well with literature that emphasizes the benefits of video games as a leisure activity that contributes to people's mental health [42]. Because our study was cross-sectional, there might also be a self-selection effect: People who feel good might be more inclined to pick up their controller. Such a view aligns well with research that shows reciprocal relations between media use and well-being [64,65]. Equally plausible, there might be factors that affect both game play time and well-being [66,67]. For example, people with high incomes are likely to be healthier and more likely to be able to afford a console/PC and the game.

Even if we were to assume that play time directly predicts well-being, it remains an open question whether that effect is large enough to matter for people's subjective experience. From a clinical perspective, it is probably the effect is too small to be relevant for clinical treatments. Our effect size estimates were below the smallest effect size of interest for media effects research that Ferguson [68] proposes. For health outcomes, Norman and colleagues [69] argue that we need to observe a large effect size of around half a standard deviation for participants to feel an improvement. In the AC:NH model, 10 h of game play were associated with a 0.06 standard deviation increase in well-being. Therefore, a half standard deviation change would require approximately 80 h of play over the two weeks (translating to about 6 h per day). However, Anvari and Lakens demonstrated that people might subjectively perceive differences of about a third of a standard deviation on a measure of well-being similar to ours [70], suggesting that approximately three and a half hours of play might be associated with subjectively felt changes in well-being. Nevertheless, it is unclear whether typical increases in play go hand in hand with perceivable changes in well-being. However, even small relations might accumulate to larger effects over time, and finding boundary conditions, such as time frames under which effects are meaningful, is a necessary next step for research [71]. Moreover, we only studied one facet of positive mental health, namely affective well-being. Future research will need to consider other facets, such as negative mental health.

Although our data do not allow causal claims, they do speak to the broader conversation surrounding the idea of video game addiction (e.g. [15]). The discussion about video games has focused on fears about a large part of players becoming addicted [14,21]. Given their widespread popularity, many policymakers are concerned about negative effects of play time on well-being [7]. Our results challenge that view. The relation between play time and well-being was positive in two large samples. Therefore, our study speaks against an immediate need to regulate video games as a preventive measure to limit video game addiction. If anything, our results suggest that play can be an activity that relates positively to people's mental health—and regulating games could withhold those benefits from players.

We also explored the role of people's perceptions in the relation between play time and well-being. Previous work has shown that gamers' experience probably influences how playing affects mental health [51,52]. We explored such a possible moderation through the lens of self-determination theory [50]: We investigated whether changes in need satisfaction, enjoyment and motivation during play changed the association between play time and well-being. We found no evidence for moderation. Neither need satisfaction, nor enjoyment, nor extrinsic motivation significantly interacted with play time in predicting well-being. However, conditional on play time, satisfaction of the autonomy and relatedness need, as well as enjoyment were positively associated with well-being. Extrinsic motivation, by contrast, was negatively associated with well-being. These associations line up with research demonstrating that experiencing need satisfaction and enjoyment during play can be a contributing factor to user well-being, whereas an extrinsic motivation for playing probably does the opposite (e.g. [56]).

Although we cannot rule out that these player experiences had a moderating role, the estimates of the effect size suggest that any moderation is likely to be too small to be practically meaningful. In other words, our results do not suggest that player experience modulates the relation between play time and well-being, but rather contributes to it independently. For example, players who experience a high degree of relatedness during play will probably experience higher well-being, but a high degree of relatedness is unlikely to strengthen the relation between play time and well-being. Future research, focused on granular in-game behaviours such as competition, collaboration and advancement will be able to speak more meaningfully to the psychological affordances of these virtual contexts.

Conditional on those needs and motivations, play time was not significantly related to well-being anymore. We are cautious not to put too much stock in this pattern. A predictor becoming not significant when controlling for other predictors can have many reasons. Need satisfaction and motivations might mediate the relation between play time and well-being; conditioning on the mediator could mask the effect of the predictor [67]. Alternatively, if play time and player experiences are themselves related, including them all as predictors would result in some relations being overshadowed by others. We need empirical theory-driven research grounded in clear causal models and longitudinal data to dissect these patterns.

4.1. Limitations

We are mindful to emphasize that we cannot claim that play time causally affects well-being. The goal of this study was to explore whether and how objective game behaviour relates to mental health. We were successful in capturing a snapshot of that relation and gaining initial insight into the relations between video games and mental health. But policymakers and public stakeholders require evidence which can speak to the trajectory of play and its effect over time on well-being. Video games are not a static medium; both how we play and discuss them is in constant flux [72]. To build on the work we present here, there is an urgent need for collaborations with games companies to obtain longitudinal data that allow investigating all the facets of human play and its effects on well-being over time.

Longitudinal work would also address the question of how generalizable our findings are. We collected data during a pandemic. It is possible the positive association between play time and well-being we observed only holds during a time when people are naturally playing more and have less opportunity to follow other hobbies. Selecting two titles out of a wide range of games puts further limitations on how generalizable our results are. Especially Animal Crossing: New Horizons is considered a casual game with little competition. Therefore, although those two titles were drawn from different genres, we cannot generalize to players across all types of games [73]. The results might be different for more competitive games. Different games have different affordances [74] and, therefore, likely different associations with well-being. To be able to make recommendations to policymakers on making decisions across the diverse range of video games, we urge video game companies to share game play data from more titles from different genres and of different audiences. Making such large-scale data available would enable researchers to match game play with existing cohort studies. Linking these two data sources would enable generalizable, causal tests of the effect of video games on mental health.

Another limiting factor on the confidence in our results is the low response rate observed in both of our surveys. It is possible that various selection effects might have led to unrepresentative estimates of well-being, game play, or their relationship. Increasing response rates, while at the same time ensuring samples' representativeness, remains a challenge for future studies in this field.

Our results are also on a broad level—possibly explaining the small effect sizes we observed. When exploring effects of technology use on well-being, researchers can operate on several levels. As Meier & Reinecke [75] explain, we can choose to test effects on the device level (e.g. time playing on a console, regardless of game), the application level (e.g. time playing a specific game), or the feature level (e.g. using gestures in a multiplayer game). Here, we operated on the application level, which subsumes all possible effects on the feature level. In other words, when measuring time with a game, some features of the game will have positive effects; others will have negative effects. Measuring on the application level will thus only give us a view of ‘net' video game effects. Assessing game behaviour on a more granular level will be necessary to gain more comprehensive insights and make specific recommendations to policymakers. For that to happen, games companies will need to have transparent, accessible APIs and access points for researchers to investigate in-game behaviour and its effects on people's mental health. Such in-game behaviours also carry much promise for studying the therapeutic effects of games, for example, as markers of symptom strength in disorders [76]. In rare cases, researchers were able to make use of such APIs [47,49], but the majority of games data are still not accessible. For PvZ, EA provided a variety of in-game behaviours that we did not analyse here. We invite readers to explore those data on the OSF project of this manuscript.

We relied on objective measures of video game behaviour. These measures are superior to self-reported behaviour because they directly capture the variable of interest. However, capturing game sessions on the side of the video game companies comes with its own measurement error. Video game companies cannot perfectly measure each game session. For example, in our data processing, some game sessions had duplicate start and end times (for PvZ) or inaccurate start and end times, but accurate session durations (for AC:NH). Measurement error in logging technology use is a common issue (e.g. [12,77]), and researchers collaborating with industry partners need to understand how these partners collect telemetry. The field needs to embrace these challenges in measurement rather than defaulting to self-reports.

Last, this study was exploratory and we made decisions about data processing and analysis without specifying them a priori [78]. Such researcher degrees of freedom can yield different results, especially in the field of technology use and well-being [65,79]. In our process, we were as transparent as possible to enable others to examine and build upon our work [31]. To move beyond this initial exploration of objective game behaviour and well-being to a more confirmatory approach, researchers should follow current best practices: they should preregister their research before collecting data in collaboration with industry partners [80,81], before accessing secondary data sources [82], and consider the registered report format [83,84]. Following these steps will result in a more reliable knowledge base for policymakers.

This leads me to speculate on the consequences of adding a lot of women to formerly male domains

Academic corruption 2: Emasculated culture. Arnold Kling, February 16, 2021. http://www.arnoldkling.com/blog/feminized-culture/

Excerpts:

[...]

This leads me to speculate on the consequences of adding a lot of women to formerly male domains. Over the past several decades, a number of important institutions that were formerly almost exclusively male now include many women: academia, journalism, politics, and management positions in organizations. These institutions increasingly are discarding the values that sustained them when the female presence was less.

1. The older culture saw differential rewards as just when based on performance. The newer culture sees differential rewards as unjust.

2. The older culture sought people who demonstrate the most competence. The newer culture seeks to nurture those who are at a disadvantage.

3. The older culture admires those who seek to stand out. The newer culture disdains such people.

4. The older culture uses proportional punishment that is predictable based on known rules. The newer culture suddenly turns against a target and permanently banishes the alleged violator, based on the latest moral fashions.

5. The older culture valued open debate. The newer culture seeks to curtail speech it regards as dangerous.

6. The older culture saw liberty as essential to a good society. The newer culture sees conformity as essential to a good society.

7. The older culture was oriented toward achievement. The newer culture is oriented toward safety. Hence, we cannot complete major construction projects, like bridges, as efficiently as we used to.

I think that in each case, the older culture was consistent with male tendencies (what Benenson calls “warriors”); the newer culture is consistent with female tendencies (what she calls “worriers”). Keep in mind that men can have worrier personalities and women can have warrior personalities, but those are not the norm.

Overall, we have made institutions harder for warriors to navigate. College no longer helps men to make the transition to adulthood. It keeps them sheltered and controlled, and after graduation they end up living with their parents.

Why did opening up opportunities for women lead to this outcome? One can imagine other outcomes. Perhaps women would have assimilated into the male culture, adopting some male tendencies in the process. Perhaps women and men would have retained their different behavioral tendencies but agreed to accommodate one another.

Instead, both men and women seem to have agreed that a purge of male tendencies is in order. Some women scorn male values as tools of oppression, and most men would rather accommodate this view than voice disagreement.

[...]

Findings suggest that individuals are unable to accurately identify AI-generated artwork and they are likely to associate representational art to humans and abstract art to machines

The Role of AI Attribution Knowledge in the Evaluation of Artwork. Harsha Gangadharbatla. Empirical Studies of the Arts, February 16, 2021. https://doi.org/10.1177/0276237421994697

Abstract: Artwork is increasingly being created by machines through algorithms with little or no input from humans. Yet, very little is known about people’s attitudes and evaluations of artwork generated by machines. The current study investigates (a) whether individuals are able to accurately differentiate human-made artwork from AI-generated artwork and (b) the role of attribution knowledge (i.e., information about who created the content) in their evaluation and reception of artwork. Data was collected using an Amazon Turk sample from two survey experiments designed on Qualtrics. Findings suggest that individuals are unable to accurately identify AI-generated artwork and they are likely to associate representational art to humans and abstract art to machines. There is also an interaction effect between attribution knowledge and the type of artwork (representational vs. abstract) on purchase intentions and evaluations of artworks.

Keywords: AI artwork, creativity, evaluation of artwork, experiment, artificial intelligence, machine learning


Among professionals at entrepreneurship events, those who had started their own business were more likely to be Toxoplasmosis gondii positive

Toxoplasmosis: Recent Advances in Understanding the Link Between Infection and Host Behavior. Stefanie K. Johnson1 and Pieter T.J. Johnson. Annual Review of Animal Biosciences, Vol. 9:249-264 (February 2021). https://doi.org/10.1146/annurev-animal-081720-111125

Abstract: Humans, wildlife, and domestic animals are intimately linked through shared infections. Many parasites and pathogens use multiple host species, either opportunistically or sequentially, such that managing disease risk frequently requires a broader understanding of the ecological community. The coccidian protozoan Toxoplasma gondii infects more than one hundred species of vertebrates, ranging from bats to beluga whales. In humans, acute toxoplasmosis can have serious health consequences for immunocompromised individuals. Even amongst asymptomatic patients, however, toxoplasmosis has been linked to a range of behavioral alterations and conditions, such as changes in risk tolerance, neuroticism, mental illness, suicide, and accident proneness. Whether such links are causal or simply correlational has been the subject of intense study and debate; from an evolutionary standpoint, selection may favor parasite-induced alterations in host behavior that increase the likelihood a host is consumed by the definitive host—in this case a domestic or wild felid. Here, we examine current evidence for parasite-induced manipulations of host behavior, in both humans and other animals. We critically evaluate proposed mechanisms through which infection might influence host behavior, which range from inflammation in the brain to changes in hormones or neurotransmitters. Considering estimates that T. gondii may infect up to one-third of the global human population, we conclude by examining the implications of these changes for human behavior, individual fitness, and emergent cultural properties.

Keywords: Toxoplasma gondii , toxoplasmosis, parasite, behavior manipulation