The role of individual differences in attentional blink phenomenon and real-time-strategy game proficiency

Abstract

The impact of action videogame playing on cognitive functioning is the subject of debate among scientists, with many studies showing superior performance of players relative to non-players on a number of cognitive tasks. Moreover, the exact role of individual differences in the observed effects is still largely unknown. In our Event-Related Potential (ERP) study we investigated whether training in a Real Time Strategy (RTS) video game StarCraft II can influence the ability to deploy visual attention measured by the Attentional Blink (AB) task. We also asked whether individual differences in a psychophysiological response in the AB task predict the effectiveness of the video game training. Forty-three participants (non-players) were recruited to the experiment. Participants were randomly assigned to either experimental (Variable environment) or active control (Fixed environment) group, which differed in the type of training received. Training consisted of 30 h of playing the StarCraft II game. Participants took part in two EEG sessions (pre- and post-training) during which they performed the AB task. Our results indicate that both groups improved their performance in the AB task in the post-training session. What is more, in the experimental group the strength of the amplitude of the P300 ERP component (which is related to a conscious visual perception) in the pre training session appeared to be predictive of the level of achievement in the game. In the case of the active control group in-game behaviour appeared to be predictive of a training-related improvement in the AB task. Our results suggest that differences in the neurophysiological response might be treated as a marker of future success in video game acquisition, especially in a more demanding game environment.

Keywords: Attentional blink; Cognitive training; Event-related potentials (ERPs); Individual differences; Real-time strategy; Video games.

Figure 1

(a) Study design: two measurement sessions were carried out during the study (pre-training and post-training). Training included 30 h of playing in the real-time strategy game (StarCraft II), spread over 4 weeks. Training varied depending on the group. (b) Example of two trials of Attentional blink task. The first trial consists of 19 letters. T1 is presented as a green letter G, and the “X” (T2) appears in Lag 2. The second trial consists of 16 letters, where T1 is presented as a green A letter, and “X” (T2) appears in Lag 7. Each trial started with a fixation cross after which a string of (16 or 19) letters was presented. Each letter was presented for 100 ms and then participants were asked about the type of T1 and the presence of T2. Each trial ended with a blank screen presented for 1500 ms after response to the second question. (c) While all of the participants played as a Terran faction during training, the opponent's race and strategy varied according to the training group type. Participants from the Variable group could match three factions, from each could use one of five strategies. The faction and the strategy were randomly selected before each match for the Variable group. In the case of a Fixed group, participants always played against the Terran faction, which used an economic strategy.

Figure 2

(a) Mean number of matches played on each level during training for Variable and Fixed groups. (b) Mean overall number of matches played during training for Variable and Fixed groups. (c) Mean overall time spent in training for Variable and Fixed groups. Asterisks indicate statistical significance: • p = .07, ∗ p < .05, ∗∗ p < .01, ∗∗∗ p < .001.

Figure 3

(a) Comparison of accuracy from attentional blink task between sessions at every Lag condition and in No T2 appearance condition for each training group. We observed very similar changes for lag conditions in both groups but all effects were stronger in a Variable group (b) than in Fixed group (c). Asterisks indicate statistical significance (• p = .09, ∗ p < .05, ∗∗ p < .01, ∗∗∗ p < .001).

Figure 4

(a) Averaged brain activity recorded on the Cz electrode during attentional blink task, with 0 point being T2 presentation moment. Waves represent each group and measurement session separately, and waves were averaged over all lags. The green color specifies time when significant differences between Variable and Fixed groups were observed (in the 2nd session, so it was a difference between dark blue line and dark red which yield the significant effect). (b) Differential waves (2nd session minus 1st session) for each training group, with 0 point being T2 presentation moment. We can clearly observe opposite effects in amplitude change after training. (c) Localization of electrodes with indication from which electrodes we took signal for our analyses. Note that in graph 4a we are showing signals from a single electrode (Cz).

Figure 5

(a) The theoretical moderation model with the P300 component's mean amplitude obtained during pre-training measurement as a predictor, Group as a moderator variable, and the mean number of matches played on the two most difficult levels (which were included in the analyses) as an independent variable. (b) The theoretical moderation models corresponding to the specific levels, which were included in the model presented above. a and b are the path and interaction coefficients (unstandardized regression weights with standard errors in parentheses). Asterisks indicate significant regression paths (• p < .07, ∗ p < .05, ∗∗ p < .01, ∗∗∗ p < .001). (c) Relationship between P300 component's mean amplitude recorded during pre-training measurement and the mean number of matches played on each of the two most difficult levels. ∗ symbol placed on the legend, corresponds to a significant effect (Variable group: Pearson's coefficient = .445, p = .043; Fixed group: Pearson's coefficient = -.242, p = .278).

Figure 6

(a) The theoretical moderation model with the mean number of matches played on the two easiest levels as a predictor, Group as a moderation variable, and difference of accuracy in attentional blink task (session 2 accuracy rate – session 1 accuracy rate) as an independent variable. (b) The theoretical moderation models corresponding to the specific levels, which were included in the model presented above. a and b are the path and interaction coefficients (unstandardized regression weights with standard errors in parentheses). Asterisks indicate significant regression paths (• p < .07, ∗ p < .05, ∗∗ p < .01, ∗∗∗ p < .001). (c) Relationship between the number of marches played on the two easiest levels and task accuracy difference. ∗ symbols placed on the legends correspond to significant conditions. Two versions of correlations were conducted: with the outlier presented on the graph (Fixed group: Pearson's coefficient = -.445, p = .038; Variable group: Pearson's coefficient = .304, p = .181) and without the outlier (Fixed group: Pearson's coefficient = -.445, p = .038; Variable group: Pearson's coefficient = .174, p = .464).