Selective enhancement of attentional networks in college table tennis athletes: a preliminary investigation

Participants

A total of 65 individuals categorized as athletes or non-athletes participated in the study. They were recruited through advertisements posted in the campus of Shanghai University of Sport. The athlete group was composed of 31 table tennis players (mean age = 21.9, ranging from 19 to 25, 11 females) whom satisfied all of the following criteria: (1) had 5 or more years of professional training experience; (2) qualified as the National Player at Second Grade or above; (3) trained more than three times a week in the last two years; and (4) trained for two or more hours each time. The non-athlete group was composed of 35 students (mean age = 21.9, ranging from 19 to 25, 14 females) majoring in psychology or kinesiology. The non-athlete group matched the athlete group in age and education, but they had no experience of playing table tennis, nor any experience of athlete training. The non-athlete group had a moderate physical activity level which was measured by the Taiwan version of the International Physical Activity Questionnaire (IPAQ) (Liou et al., 2008). All the participants were right-handed and had normal or corrected to normal visual acuity. No individuals reported having a history of neurological or psychiatric disorder. Written informed consent was obtained from each participant prior to the study. All participants received a payment of approximately $10 for taking part in the experiment. Table 1 shows the main characteristics of the subjects. This study was approved by the Ethics Committee of the Shanghai University of Sport (No. 2015014).

Attention network test

Attention network test (ANT) was designed to assess the function of the three different attention networks (Fan et al., 2002). A fixation cross was presented in the center of a computer screen at the onset of each trial. After a random interval of 400–1,600 ms, cues would present in one of the four possible conditions: no cue, center cue (the fixation cross was replaced by an asterisk), double cue (two asterisks were respectively displayed above and below the fixation cross), or spatial cue (an asterisk were displayed either above or below the fixation cross). The cues remained visible for 100 ms. The presentation of asterisks provided temporal information about the appearance of target stimuli. The asterisk in the spatial cue condition provided additional information about the location of target stimuli. The spatial cues were always valid. The fixation cross was displayed alone for 400 ms after the disappearance of cue. Then a target stimulus was presented above or below the fixation cross according to the indication of the previous cue. The target stimulus consisted of five horizontally arranged arrows or lines. Participants were required to press the corresponding key to indicate the direction of the central target arrow. The other four arrows or lines served as flankers in the task with three possible conditions: congruent condition (arrows pointed in the same direction as the central arrow), incongruent condition (arrows pointed in the opposite direction of the central arrow), or neutral condition (lines with no direction information). The target stimulus remained on the screen until the participant responded or for 1,700 ms if no answer was given.

The participants were instructed to concentrate on the fixation cross throughout the task. A numeric keyboard was placed in front of the participant and the participant was required to lightly put his left hand index finger on key “1” and right hand index finger on key “3.” Once target stimuli were presented, participants were instructed to respond as fast and accurately as possible by pressing the key “1” for left directed central target arrow and pressing the key “3” when the direction was right.

Four blocks were included in this test. Each block contained 48 trials based on the combination of four cues conditions (no cue, center cue, double cue, and spatial cue), three flankers’ conditions (congruent, incongruent, and neutral), two directions (left or right directed target arrow) and two locations (target displayed above or below the fixation cross). Each trial was presented only once in a block. The stimuli were presented and the data were recorded using Psychtoolbox (Brainard, 1997) (see Fig. 1).

The three components of attentional network were computed as follows: no cue RTs versus double cue RTs for alerting, central cue RTs versus spatial cue RTs for orienting and congruent flankers RTs versus incongruent flankers RTs for executive network.

Procedure

As a requirement of the advertisements, all the participants had to contact the researchers by telephone first. A survey about the demographic data of participants was conducted during the call. Athlete participants were further asked about their training experiences. Participants who met the criteria (see ‘Participants’) were invited to our laboratory on another day to participate in the experiment. They were instructed to abstain from alcohol for 24 h and from caffeine-containing substances for 12 h before the experiment.

After arriving at the laboratory, participants were asked to sign an informed consent form and were assessed by the Taiwan version of the International Physical Activity Questionnaire (IPAQ). Then the purpose of the study and the instruction of ANT were introduced to them in written form. After participants reported understanding the instructions, they performed the ANT task individually in a dimly lit and quiet room. At first, they had to perform a practice block with 24 random trials. If their response accuracy reached 80%, they could perform their next four experimental blocks of 48 trials in each; otherwise, they would perform another practice block until their accuracy reached 80%. Participants were allowed to rest between each block, and they could start the next block by pressing any keys once they felt adequately rested. Completing the whole task required about 17 min, including both practice and experimental blocks.

Design and statistical analysis

A mixed factors design was adopted in the study. The athlete and non-athlete group was a between-subjects variable, the cue type (no cue, central cue, double cue and spatial cue) and flankers type (neutral, congruent, incongruent) were within-subject variables. The dependent variables were response times (RTs) and accuracy rates. They were analyzed with a 2 (group) × 4 (cue type) × 3 (flanker type) mixed-design ANOVA.

A t-test between athlete and non-athlete groups was carried out in order to explore the effect of athlete experience on each component of attentional network.

Participant characteristics

No significant differences were observed in age (F_(1,63) = 0.00, p = 0.98), height (F_(1,63) = 3.29, p = 0.07), weight (F_(1,63) = 2.92, p = 0.09), BMI (F_(1,63) = 0.64, p = 0.43), average reaction time (F_(1,63) = 1.45, p = 0.23) and accuracy rate (F_(1,63) = 0.07, p = 0.79), and as expected, a significant difference was observed in physical activity level (overall score on IPAQ) (F_(1,63) = 4.29, p < 0.05) of the two groups (see Table 1).

Mean RTs

For the RTs analysis, an outlier correction was done by excluding the trials which were 3 standard deviations from the mean for each flanker condition (congruent, incongruent and neutral) individually. The method of outlier correction was suggested by one reviewer. The wrong trials were also excluded, and the proportion of excluded data was 1.1%. Results showed a significant main effect of cue type (F_(3,189) = 138.82, p < 0.01, ${\eta }_p^{\mathrm{2}}=0.70$), the RTs were the longest in the no cue condition, and the shortest in the spatial cue condition. A significant main effect also observed in flanker type (F_(2,126) = 318.31, p < 0.01, ${\eta }_p^{\mathrm{2}}=0.84$). The RTs were longer in the incongruent condition than in the congruent or neutral condition. Furthermore, there were significant interactions between flanker type and cue type (F_(6,378) = 7.90, p < 0.01, ${\eta }_p^{\mathrm{2}}=0.11$), group and flanker type (F_(2,126) = 4.68, p < 0.01, ${\eta }_p^{\mathrm{2}}=0.7$). The interaction contrasts for flanker type and cue type revealed significant differences between the congruent and incongruent conditions, incongruent conditions and neutral conditions under all cue conditions, no significant differences were observed between congruent and neutral conditions under all cue conditions. The interaction contrast for the group and flanker type revealed significant differences between the groups under incongruent condition, no significant differences were observed between the groups under congruent and neutral conditions. There were no significant main effect of group (F_(1,63) = 1.45, p = 0.23, ${\eta }_p^{\mathrm{2}}=0.02$), group × cue type (F_(3,189) = 0.63, p = 0.60, ${\eta }_p^{\mathrm{2}}=0.01$) or group × cue type × flanker type (F_(6,378) = 0.41, p = 0.87, ${\eta }_p^{\mathrm{2}}=0.00$) interaction. The description data of the mean RTs and standard deviations of athlete and non-athlete group according to the cue and flanker type are shown in Table 2.

Accuracy

For the accuracy analysis, significant main effects of cue type (F_(3,189) = 7.89, p < 0.01, ${\eta }_p^{\mathrm{2}}=0.11$), and flanker type (F_(2,126) = 39.9, p < 0.01, ${\eta }_p^{\mathrm{2}}=0.39$) were revealed. Furthermore, there were significant interactions between flanker type and cue type (F_(6,378) = 4.10, p < 0.01, ${\eta }_p^{\mathrm{2}}=0.6$). Interaction contrast revealed significant differences between the congruent and incongruent conditions, incongruent conditions and neutral conditions under all cue conditions, no significant differences were observed between congruent and neutral conditions under all cue conditions. There were no significant main effect of group (F_(1,63) = 0.03, p = 0.87, ${\eta }_p^{\mathrm{2}}=0.00$), group and flanker type (F_(2,126) = 1.59, p = 0.21, ${\eta }_p^{\mathrm{2}}=0.3$), group × cue type (F_(3,189) = 0.88, p = 0.45, ${\eta }_p^{\mathrm{2}}=0.01$) or group × cue type × flanker type (F_(6,378) = 1.59, p = 0.15, ${\eta }_p^{\mathrm{2}}=0.2$) interaction. The descriptive data of the mean accuracy and standard deviations of athlete and non-athlete group according to the cue and flanker type are shown in Table 3.

Differences of athletes and non-athletes on the 3 components of attentional network

Independent samples t-tests were carried out for each component of the attentional system (alerting, orienting and executive networks). Results showed a significant difference between athlete and non-athlete group on executive network (t₍₆₃₎ = 2.36, p = 0.02), while no differences were observed on alerting (t₍₆₃₎ =  − 0.05, p = 0.96) or orientation (t₍₆₃₎ =  − 1.32, p = 0.19) networks (see Fig. 2).