This meta-analysis synthesizes current literature concerning the effects of active video games (AVGs) on children/adolescents' health-related outcomes. A total of 512 published studies on AVGs were located, and 35 articles were included based on the following criteria: (i) data-based research articles published in English between 1985 and 2015; (ii) studied some types of AVGs and related outcomes among children/adolescents and (iii) had at least one comparison within each study. Data were extracted to conduct comparisons for outcome measures in three separate categories: AVGs and sedentary behaviours, AVGs and laboratory-based exercise, and AVGs and field-based physical activity. Effect size for each entry was calculated with the Comprehensive Meta-Analysis software in 2015. Mean effect size (Hedge's g) and standard deviation were calculated for each comparison. Compared with sedentary behaviours, AVGs had a large effect on health outcomes. The effect sizes for physiological outcomes were marginal when comparing AVGs with laboratory-based exercises. The comparison between AVGs and field-based physical activity had null to moderate effect sizes. AVGs could yield equivalent health benefits to children/adolescents as laboratory-based exercise or field-based physical activity. Therefore, AVGs can be a good alternative for sedentary behaviour and addition to traditional physical activity and sports in children/adolescents.

This paper investigates two strategies for improving coreference resolution: (1) training separate models that specialize in particular types of mentions (e.g., pronouns versus proper nouns) and (2) using a ranking loss function rather than a classification function. In addition to being conceptually simple, these modifications of the standard single-model, classification-based approach also deliver significant performance improvements. Specifically, we show that on the ACE corpus both strategies produce f-score gains of more than 3% across the three coreference evaluation metrics (MUC, B 3 , and CEAF).

BackgroundAlthough emerging research is demonstrating the potential health impact of exergaming, investigations have primarily been conducted in laboratory settings among small samples with short-term interventions. Information on the effectiveness of exergaming in underserved children's objective physical activity (PA) in population-based settings is also scarce. Moreover, most empirical studies have only included 1 type of exergame in the intervention. Therefore, this study's purpose was to investigate the long-term impact of a multigame exergaming intervention among underserved children integrated within school curricula. Specifically, this study examined the effect of exergaming on children's accelerometer-determined sedentary behavior (SB), light PA, moderate-to-vigorous PA (MVPA), and energy expenditure (EE) over 2 years as compared with regular physical education (PE) classes.MethodsA total of 261 second- and third-grade children (134 girls, 127 boys; mean age 8.27 years) were recruited from 2 Texas elementary schools. Children's pre-test 3-day SB, light PA, MVPA, and EE at school were assessed in the fall of 2012. Participants were assigned to 1 of 2 groups: (1) exergaming/PE group (125 min weekly of exergaming-based PA program) and (2) comparison group (125 min weekly of PE). PA (SB, light PA, and MVPA) and EE outcome variables were assessed again in 2013 (post-test) and 2014 (follow-up).ResultsSignificant time effects were observed for SB (F(1, 162) = 25.0, p < 0.01, η2 = 0.14), light PA (F(1, 162) = 9.6, p < 0.01, η2 = 0.06), and MVPA (F(1, 162) = 6.2, p = 0.01, η2 = 0.04) but not for EE (F(1, 162) = 0.63, p > 0.05, η2 = 0.004). Subsequent pairwise comparisons revealed significant increases from pre- to post-test for light PA (p < 0.01), MVPA (p < 0.01), and EE (p = 0.02) with no changes in SB (p > 0.05). Conversely, significant decreases occurred in light PA (p < 0.01) from post-test to follow-up with no differences seen in MVPA (p = 0.08) and EE (p = 0.06) over the same time period. A significant increase was seen, however, for SB from post-test to follow-up.ConclusionExergaming PE can have the same positive effect on children's light PA, MVPA, and EE as regular PE. More research is necessary to discern how to promote long-term PA participation after conclusion of the intervention.

BackgroundPlayers may not acquire adequate levels of moderate-to-vigorous physical activity (MVPA) when playing commercial video games. This study's goal was to evaluate the effects of an exercise bike video game played by using a mobile application-based exergame that was designed exclusively to promote participants' MVPA, with additional attention paid to this game's ability to promote greater situational interest.MethodsAn experimental design was used with 163 students (aged 20.31 ± 1.30, 18–26 years, 61.3% male), all of whom were randomly allocated into an experimental group and a control group. Physical activity (PA) levels were assessed with ActiGraph GT3X+ (ActiGraph Inc., Fort Walton Beach, FL, USA) accelerometers. The situational interest scale was used to evaluate students' situational interest in both groups. Multivariate analysis of variance was conducted to examine the differences between sedentary behavior, PA levels, and situational interest between groups. Regression analyses were also used, with the purpose being to evaluate the strength of the relationship between PA and situational interest.ResultsResults revealed that the experimental group had higher degrees of sedentary behavior, light PA, total interest, instant enjoyment, exploration intention, attention demand, novelty, and challenge, whereas the control group received higher scores for MVPA (control 95.01% vs. experimental group 89.94%). Regression analysis indicated that instant enjoyment (β = 0.49, p < 0.01), exploration intention (β = 0.18, p < 0.05), and attention demand (β = 0.17, p < 0.05) were positive predictors for total interest, explaining 43% of its variance.ConclusionA newly designed mobile application-based exergame played via an exercise bike may enhance situational interest and provide a decent level of PA for players.

This paper presents a detailed comparative framework for assessing the usefulness of unsupervised word representations for identifying so-called implicit discourse relations. Specifically, we compare standard one-hot word pair representations against low-dimensional ones based on Brown clusters and word embeddings. We also consider various word vector combination schemes for deriving discourse segment representations from word vectors, and compare representations based either on all words or limited to head words. Our main finding is that denser representations systematically outperform sparser ones and give state-of-the-art performance or above without the need for additional hand-crafted features.