Behavioral And Functional Magnetic Resonance Imaging Study On Internet Gaming Addiction In Adolescents

Background Internet gaming or online gaming has become an important part of everyday life in adolescents in our country, but internet game addiction has also brought a lot of health problems both physiologically and psychologically with them. More and more concerns were taken for it from psychologists, educators and even the whole society. Internet game addiction, as a subtype of internet addiction or internet addiction disorder, has shown negative effects on net gaming players or addicts evidenced by many researches, and psychological and cognitive disorders were included besides the physical problems. On the other side, some researchers have also found that net gaming may be beneficial to the players in their neural network of attention orientation. But it is still disputed whether the enhancement of the networks' function belongs to exogenous or endogenous attention, and the neural mechanism underlying this improvement remains unclear. In present study, the behavioral differences between gaming addicts and control subjects were evaluated by using a revised paradigm, and functional MRI techniques were also employed to explore the neural mechanism related to the behavioral performance in the gaming addicts group.PartⅠ:Behavioral Study of Internet Gaming Addiction in AdolescentsObjective By using a paradigm revised from both Posner's classical visual-spatial cue paradigm and ANT paradigm of Fan et al, the aims of our present study were to explore differences of the functional enhancement in attention orientation networks between gaming addicts and the normal subjects through testing and comparing their behavioral performance, and then to analyze the facilitative mechanism of the behavioral improvement resulting from gaming experience.Materials and Methods Subjects in present study were divided into two groups, 17 gaming addicts were in the addiction group and 20 normal subjects in the control group. The paradigm used in the study was written with E-Prime (version 2.0) software (Psychology Software Tools, INC. www.pstnet.com). The task was compiled and run on an IBM PC, with a 17 inch LCD monitor(resolution 1024×768; refresh rate 85Hz). And the results of subjects'responses (reaction times, RTs) were recorded automatically by E-prime software. Our paradigm employed different pictures together with their cutting (physical) images including gaming pictures as cues, revised from both Posner's classical visual-spatial cue paradigm and ANT paradigm of Fan et al. The paradigm was designed in event-related manner, in which 5 runs were included in a task, and there were 60 trials in each run. Each trial contain 3 components:central visual fixation"+", pictorial cues and target stimulus of the task. Data analysis was conducted with SAS for Windows 9.13 and SPSS13.0. Two groups (addiction group and control group), two conditions (valid cue and invalid cue) and three levels (game picture, neutral face picture and neutral view picture) of cues were included in the design of the test. Methods of statistical analysis were MANOVA, one-way ANOVA, paired-sample and independent-sample t-test. The following analyses were conducted:①Analysis of influence on mean RTs under valid and invalid cue conditions;②Analysis of influence on mean RTs on three levels of cue pictures;③Analysis of trends in the change of mean RTs.Results:1. Gaze cueing effect (GCE) on subjects in control and addiction groups:①For the same subject, there was statistically significant difference of mean RTs under valid and invalid cue conditions (F(1,105)=130.88, P=0.000);②Conducted with paired-sample t-test, there were statistically significant differences of mean RTs under valid and invalid cue conditions within the same group. Control group:Game (t=-8.601, P=0.000); Face (t=-4.033, P=0.001); View (t=-5.659, P=0.000). Addiction group:Game (t=-5.558, P=0.000); Face (t=-2.432, P=0.027); View (t=-3.663, P=0.002).③Conducted with independent-sample t-test, there were statistically significant differences of mean RTs under valid and invalid cue conditions between two groups (F(1,219)=8.166, P=0.005).2. Compare of mean RTs on three cue levels between control and addiction groups.①There were no statistically significant differences of mean RTs among three cue levels for the same subject (F(2,105)=0.33, P=0.7199);②There were statistically significant differences of mean RTs on three cue levels between the two groups. (F(2,218)=15.12, P=0.0000). On each cue level, the range of different-values of RTs between the two groups was from 100.89ms to111.92ms, and the mean value was 108.02ms;③There were statistically significant differences of mean RTs between the two groups on each cue level of pictures. (F(1,105)=24.16, P<0.0001).3. Trends of mean RTs change for subjects in control and addiction groups:①There were statistically significant differences of mean RTs on three different cue levels under the two cue conditions (F(1,105)=0.03, P=0.9739), no interactions were found between group division and cue level, trends of mean RTs were shown as game>view>face;②Conducted with MANOVA, there were no statistically significant differences on trends of mean RTs change between the two groups from valid cue to invalid cue (F(2,105)=0.55, P=0.5769)；③Conducted with MANOVA, there were no statistically significant differences on trends of mean RTs change from valid cue to invalid cue among the three cue levels for subjects in two groups(F(2,105)=3.03, P=0.0525);④Conducted with one-way ANOVA on control group and addiction group separately, we found that there were statistically significant differences on trends of mean RTs change among the three cue levels for addiction group (F (2,48)=3.62, P=0.0345). However, there were no statistically significant differences on trends of mean RTs change among the three cue levels for the control group(F (2,57)=0.63,P=0.5380).Conclusions:Both endogenous and exogenous attention orientation were included in GCE which representing attention orientation; Facilitation of attention based on gaming experience has significantly improved the attention processing, functions of both endogenous and exogenous attention orientation were enhanced, and there is a possibility of endogenous-weighted enhancement between the two systems.PartⅡ:Functional Magnetic Resonance Imaging of Internet Gaming Addiction in AdolescentsObjective:Based on behavioral results of PartⅠstudy, the aim of our study in this part is to further explore the facilitation neural mechanism of attention orientation by using fMRI technique with our revised paradigm.Materials and Methods:Subjects in our present study were divided into two groups:the control group and the addiction group, with 15 subjects in each group. Event-related paradigm design was employed in fMRI study, and the same test task to partⅠstudy and a whole-body MRI system of GE1.5T (Twin-speed infinity with ExciteⅡ) scanner with head orthogonal coil were adopted. Data preprocessing, statistical analysis and results demonstration were performed using SPM2 (statistics parameter mapping) package. Preprocessing of the fMRI data include time rectification within the slice, motion correction, spacial standardization and data smoothing. Analysis within group:Conducted with independent-sample t-test in basic mode of SPM2, the g-f and g-v mean activation images were acquired in the control and addiction group respectively. Two sample t-test was used to get discrepancy images between two groups under g-f and g-v conditions. Results:①Activated brain areas under g-f condition in the control group inlcude:bilateral dorsal lateral prefrontal cortex (DLPFC), bilateral medial part of prefrontal cortex (MPFC), bilateral posterior parietal cortex (PPC), bilateral temporal lobes, left caudate nucleus, bilateral occipital lobes, and bilateral cerebellum.②Activated brain areas under g-v condition in the control group include bilateral DLPFC, bilateral ventral lateral prefrontal cortex (VLPFC), bilateral MPFC; bilateral PPC, bilateral temporal lobes, and bilateral cerebellum.③Activated brain areas under g-f condition in the addiction group:bilateral DLPFC, bilateral VLPFC, bilteral MPFC, left PPC, and bilateral temporal lobe.④Activated brain areas under g-v condition in the addiction group:bilateral DLPFC, bilateral VLPFC, left PPC, right postcental gyrus, bilateral temporal lobes, and right middle cingulate gyrus.⑤compared with the addiction group, bilateral PPC, left inferior temporal gyrus, and right middle cingulate gyrus showed increased activation under g-f condition in the control group.⑥compared with the addiction group, right DLPFC, left precentral gyrus, bilateral temporal lobe, and bilateral precuneus presented with increased activation under g-v condition in the control group.⑦compared with control group, right DLPFC, left MPFC, left postcental gyrus, bilateral temporal lobe, right hippocampus, and right basal ganglia showed increased activation in the addiction group under g-f condition.⑧compared with control group, bilateral DLPFC, bilateral MPFC, bilateral PPC, left inferior temporal gyrus, bilateral occipital lobes, left cingulated gyrus, right basal ganglia, bilateral thalamus, and left cerebellum appeared increased activation in the addiction group under g-v condition.Conclusions:Results of this fMRI study had shown brain areas relevant to both endogenous and exogenous attention networks activated both in control and addiction groups, whereas the predominant activated areas were relevant to endogenous attention network. Furthermore, compared with the control group, the most obviously activated brain area in addiction group was the right LDLPFC, an important node of endogenous attention network. The function of endogenous attention orientation network might be enhanced with player's net gaming experience, and the enhancement might be the neural mechanism of the subjects'behavioral performance in addiction group.