Acta Psychologica 169 (2016) 11–19

Contents lists available at ScienceDirect

Acta Psychologica

j ourna l homepage: www.e lsev ie r .com/ locate /actpsy

Video game players show higher performance but no difference in speedof attention shifts

David J. Mack a,b,⁎, Helene Wiesmann a, Uwe J. Ilg a

a Hertie Institute for Clinical Brain Research, Department of Cognitive Neurology, University of Tübingen, Tübingen, Germanyb Department of Neurology, UniversityHospital Zurich, Switzerland

⁎ Corresponding author at: Department of NeurologyIntegrated Systems Laboratory, ETH Zurich, Gloriastrasse 3

E-mail address: david-jule.mack@iis.ee.ethz.ch (D.J. M

http://dx.doi.org/10.1016/j.actpsy.2016.05.0010001-6918/© 2016 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 22 June 2015Received in revised form 27 April 2016Accepted 2 May 2016Available online xxxx

Video games have become both a widespread leisure activity and a substantial field of research. In a variety oftasks, video game players (VGPs) perform better than non-video game players (NVGPs). This difference ismost likely explained by an alteration of the basicmechanismsunderlying visuospatial attention.More specifical-ly, the present study hypothesizes that VGPs are able to shift attention faster than NVGPs. Such alterations in at-tention cannot be disentangled from changes in stimulus-response mappings in reaction time basedmeasurements. Therefore, we used a spatial cueing task with varying cue lead times (CLTs) to investigate thespeed of covert attention shifts of 98male participants divided into 36NVGPs and 62VGPs based on theirweeklygaming time. VGPs exhibited higher peak andmean performance than NVGPs. However, we did not find any dif-ferences in the speed of covert attention shifts as measured by the CLT needed to achieve peak performance.Thus, our results clearly rule out faster stimulus-responsemappings as an explanation for the higher performanceof VGPs in line with previous studies. More importantly, our data do not support the notion of faster attentionshifts in VGPs as another possible explanation.

© 2016 Elsevier B.V. All rights reserved.

Keywords:Top-down controlSpatial cueingExogenous vs. endogenousInformation processingStimulus-response mappingVideo games

1. Introduction

Video games have been a research topic in science for the last threedecades (Latham, Patston, & Tippett, 2013). Despite their high preva-lence – 60% of juveniles in the U.S. play at least 1 h a day (Rideout,Foehr, & Roberts, 2010) – we still have no consistent evidence aboutthe consequences of playing video games. Some studies have showndetrimental effects like increased aggression (Anderson et al., 2010),or addiction symptoms (Gentile et al., 2011). However, it remains un-clear whether or not violence in video games can be blamed for aggres-sive behavior (Ferguson, San Miguel, Garza, & Jerabeck, 2012), or ifbeing “bad” in a video game improves moral sensitivity in the realworld (Grizzard, Tamborini, Lewis, Wang, & Prabhu, 2014). But videogame play has also been causally linked to positive effects like superiorcontrast sensitivity (Li, Polat, Makous, & Bavelier, 2009), enhanced con-trol over selective attention (Green & Bavelier, 2003), improved multi-tasking (Chiappe, Conger, Liao, Caldwell, & Vu, 2013; Strobach,Frensch, & Schubert, 2012), increased visual working memory capacity(Blacker & Curby, 2013) and encoding speed (Wilms, Petersen, &Vangkilde, 2013), faster information integration (Green, Pouget, &Bavelier, 2010) and even real-life ameliorations such as better surgicalskills (Schlickum, Hedman, Enochsson, Kjellin, & Fellander-Tsai, 2009),

, UniversityHospital Zurich, c/o5, CH-8092 Zurich, Switzerland.ack).

or improved reading abilities in dyslexic children (Franceschini et al.,2013).

In addition, correlational studies found more precise temporal pro-cessing (Donohue, Woldorff, & Mitroff, 2010; Rivero, Covre, Reyes, &Bueno, 2013) and better selective attention abilities (Cain, Prinzmetal,Shimamura, & Landau, 2014; Chisholm & Kingstone, 2012; Green &Bavelier, 2003) in video game players (VGPs) compared to non-videogame players (NVGPs). In general, VGPs show shorter reaction timesthan NVGPs in a multiplicity of tasks (Dye, Green, & Bavelier, 2009).We have shown that VGPs have shorter reaction times but do not pro-ducemore errors in an anti-saccade task (Mack & Ilg, 2014). Our resultsindicate that inhibitory control is not altered in VGPs. Paralleling our re-sults, a similar study using saccade targets and distractors also foundshorter saccadic reaction times in VGPs (Heimler, Pavani, Donk, & vanZoest, 2014). However, this study found a slight increase in error ratesof VGPs.

Recently, it was proposed that VGPs exhibit an enhanced ability in“learning to learn” (Bavelier, Green, Pouget, & Schrater, 2012), that is,the ability to adapt swiftly to new tasks. More specifically, allocationof attentional resources is increased, thereby enhancing the signal inquestion for the task. It is debated whether these attentional improve-ments are related to exogenous, bottom-up control (Cain et al., 2014),or to early distractor inhibition (Bavelier, Achtman, Mani, & Focker,2012; Mishra, Zinni, Bavelier, & Hillyard, 2011) and other, later compo-nents of endogenous, top-down control of attention (Chisholm, Hickey,Theeuwes, & Kingstone, 2010; Chisholm&Kingstone, 2012; Clark, Fleck,

12 D.J. Mack et al. / Acta Psychologica 169 (2016) 11–19

& Mitroff, 2011; Wu et al., 2012). Based on our eye movement results,we wanted to explore a third alternative: faster allocation of attentionalresources in VGPs, as suggested by Bavelier, Green, et al. (2012).We hy-pothesized that, in the framework of visuospatial attention, not just theallocation, but themere attentional orienting response is faster in VGPs.Within the famous “spotlight of attention” metaphor, this orienting isachieved through a covert attention shift (Posner, 1980; Posner,Snyder, & Davidson, 1980). This shift can be driven in a bottom-upman-ner through exogenous signals like sudden onset cues, or by top-downcontrol through endogenous signals like symbolic cues. It has been ar-gued that bottom-up processes precede top-down control of attention(Theeuwes, 2010) and that feature-based attention is closely relatedto bottom-up priming (Theeuwes, 2013). In our own study (Mack &Ilg, 2014), we found faster reaction times in VGPs for exogenously aswell as endogenously driven saccadic eye movements that are believedto be preceded by covert attention shifts (Shepherd, Findlay, & Hockey,1986). Since both saccade types were similarly affected, a faster covertshift of attention might be the best explanation for these results. In ad-dition, this would account for the shorter reaction times of VGPs in anytask which involves some form of spatial attention.

However, in reaction time based experiments, faster responses canbe explained alternatively by more efficient stimulus-response map-pings (Castel, Pratt, & Drummond, 2005). An examination of purely at-tentional effects must therefore use a paradigm without any motorinvolvement. Although there have been perceptual studies usingperfor-mance based signal detection tasks (Bejjanki et al., 2014; Green &Bavelier, 2003; Schubert et al., 2015; West, Stevens, Pun, & Pratt,2008; Wilms et al., 2013), none of them looked explicitly at the speedof covert attention shifts. Using the theory of visual attention to modelthe effects of video game play on various aspects of visual attention, ithas been shown that VGPs have a higher processing speed and lowerperceptual thresholds compared to NVGPs (Schubert et al., 2015;Wilms et al., 2013).

The spatial cueing task of Nakayama and Mackeben (1989) is an el-egant way to measure the speed of attention shifts without any motorinvolvement. In the “Nakayama task” the shifting speed is derivedfrom discrimination performance. As in other spatial cueing tasks, par-ticipants have to detect the presence of an oddball in a search array.The oddball is defined by a feature conjunction of orientation andcolor (Treisman & Gelade, 1980). In contrast to normal conjunctionsearch tasks, the oddball's location is cued, reducing the conjunctionsearch to a simple neighbor comparison. The crux of the task is thevery brief presentation of the search array for only 17 ms. The durationof the cue indicating the upcoming oddball location in the search array(“cue lead time”; CLT), is systematically varied between trials. With in-creasing CLT, an attentional enhancement of the signal and thus betterdiscrimination performance can be observed until a certain point. Forlonger CLTs, the attentional enhancement decays and performancedrops substantially. At a specific CLT, the attentional enhancement willbe strongest, resulting in a performance peak. This CLT for peak perfor-mance is a direct measure for the speed of covert attention shifts.

Nakayama and Mackeben (1989) suggested that the time course ofperformance reflects an early peaking bottom-up part as well as a lateplateauing top-down component. At short CLTs, the orientation re-sponse is transient and bottom-up triggered. At long CLTs, the responseis sustained and under top-down control. It has been shown that the at-tentional selection of the signal before the covert attention shift is re-sponsible for the transient component (Wilschut, Theeuwes, & Olivers,2011). This study also found that task difficulty is echoed in a transitionfrom shorter to longer CLTs for peakperformance. The authors proposedthat this reflects a shift from a bottom-up defined to a top-down con-trolled strategy in the participants. In summary, the CLT for peak perfor-mance in the Nakayama task reflects differences in subjective taskdifficulty, as well as type and speed of attention shift.

Very recently, Schubert et al. (2015) showed, that the attentionalsystem of VGPs is especially better in the lower visual field. Processing

speed seems to be higher for stimuli in this region. In their experiment,the authors presented five letters arranged in a vertical column to mea-sure visual processing speed on a fine-grained eccentricity level. Sincethe stimuli in our Nakayama task are arranged in a circular manner, itis possible to examine effects of retinal position on an isoeccentriclevel and to elaborate on the effects of eccentricity in a future study.

1.1. Research questions

We used the Nakayama task to pursue four research questions:

(1) Do VGPs perform better in the Nakayama task, thereby lendingfurther support against a faster stimulus-response mapping?

(2) Do potential attentional differences in VGPs result from faster at-tention shifts?

(3) Do VGPs and NVGPs differ in the balance between top-down andbottom-up control of attention?

(4) Are differences in performance between VGPs and NVGPs espe-cially pronounced in the lower visual field?

It is important to note that our study is a correlational study. Obvi-ously, it is impossible to infer any causal relationship between differ-ences in performance and video game play from our results. Incontrast, we intended to explore the nature of these differences as pre-cisely as possible.

2. Methods

2.1. Experimental setup

The experiments were performed with a PC (Compaq dc5750) run-ning under Windows XPwith extended desktop settings for two moni-tors. The stimulus screen (HP L1950; screen diagonal: 19″, refresh rate:60 Hz, resolution: 1280 × 1024 pixels) was connected via the VGA-portof the graphics adapter (ATI Radeon Xpress 1150). Stimuli were pre-sented using the Psychophysics Toolbox Version 3 (Brainard, 1997;Pelli, 1997) and MATLAB R2008a (The Mathworks, Natick, MA). View-ing distance of the participants was kept constant at 57 cm throughuse of a head rest.

2.2. Task

Fig. 1 shows the sequence of events in our version of the Nakayamatask (i.e. “Experiment 5: effect of retinal eccentricity” in Nakayama &Mackeben, 1989). Participants had to indicate if the bar at the cued loca-tion matched the other bars in its feature combination. These featureswere ‘orientation’ (horizontal/vertical) and ‘color’ (black:luminance ≤ 1 cd/m2; white: 125 cd/m2). Stimuli were presented on agray background (30 cd/m2). Each trial started with a white fixationcross (size: 19 × 19 arcmin, line width: 2 arcmin) at the center of thescreen. After a random fixation time (250–500 ms), a red square (size:44 × 44 arcmin, line width: 4 arcmin) cued the oddball location. TheCLT was parametrically chosen from 14 values (0, 17, 33, 50, 67, 83,100, 117, 133, 150, 200, 300, 400 and 600ms). Subsequently, the searcharray, consisting of 12 bars (each sized 30 × 16 arcmin) in a circular ar-rangement of 4° radius, was shown for 17 ms. Bar centers were equallyspaced at the 12 clock positions with a center-to-center distance of 2.1°of visual angle. Six bars were randomly assigned to one feature combi-nation (e.g. “horizontal-black”) and the remaining six to the oppositefeature combination (e.g. “vertical-white”). Thus, the two groups ofbars were always different in both feature dimensions. The bar at thecued location (aka the oddball) either differed in its orientation fromthe rest (e.g. “vertical-black” or “horizontal-white”) or not at all (e.g.“horizontal-black” or “vertical-white”). All other bars were randomlyassigned to any of the 11 remaining positions. The cue stayed on the

Fig. 1. Sequence of events in the Nakayama task used in the current study. The three possible responses are indicated at the right (the correct response in this example would be “1”). Theinset in the upper right shows the timing of the stimulus. F: fixation, C: cue, S: search array, M: mask, and R: response indicator.

13D.J. Mack et al. / Acta Psychologica 169 (2016) 11–19

screen during the presentation of the search array to avoid any offsettransient. After the search array, a black-and-white concentric circlepattern was shown for 250 ms to mask the search array and diminishthe influence of afterimages. Finally, the mask disappeared and the fix-ation cross changed its color fromwhite to black, signaling that the par-ticipant should now respond to the oddball orientation (horizontal,vertical or none). Responses were collected via keyboard buttonpresses. Non-numeric keypad digits were used to indicate if the oddballwas horizontal (button “1” press; left middle finger), vertical (button“2” press; left index finger), or wasn't an oddball at all, i.e., matchedthe other bars (button “0” press; right index finger). The three oddballtypes reduced the chance level to 33% and assured that mere guessingwas not a rewarding strategy. Each trial ended after the participantpressed the button. Therefore, the measurement was self-paced andcould be halted in between two trials. After the response, the nexttrial started with an inter-trial interval of 300 ms.

Participants performed three training blocks of 42 trials each (1 rep-etition for the 14 CLTs and the 3 oddball types), each with decreasingsearch array durations (117, 83 and 50 ms). This was meant to avoidlearning effects in the subsequent main experimental block (294 trials;7 repetitions for each CLT and oddball type). Training and main experi-mental blocks were preceded by an additional demonstration of the en-tire sequence of events and all possible response options. Completion ofthe whole experiment took between 40 and 75 min.

2.3. Data processing

Trialswere considered ‘correct’ if the participant's responsematchedthe oddball type. The performance at each CLTwas defined as themeancorrect performance at that CLT. Baseline performance was derivedfrom performance at 0 ms CLT and was designed to measure perfor-mance without attentive influences. Mean performance over all CLTswas computed as a general measure of participants' task achievements.Peak performance was calculated as the maximum performance, andthe corresponding CLTwas used as ameasure for the speed of the covertattention shift. Some participants had more than one peak, in whichcase the first CLT for peak performance was taken to get a lower limiton the speed of the attention shift. Based on the suggestion of Nakayamamentioned in the Introduction, participants were separated accordingto their CLT results into a transient class (CLT for peakperformance b 150 ms) and a sustained class (CLT for peakperformance ≥ 150 ms). This threshold was derived from the troughin the distribution of the CLTs for peak performance, which can beseen in Fig. 3.

2.4. Participants

98 male participants were enlisted from German high school seniorclasses. To avoid differential motivation effects in any of the groups, we

covertly recruited participants over the course of 5 months. Before theexperiment, participants were only told that the study was on attentionin juveniles. No further information was given until the end of the ex-periment, wherein the video game aspect of the study was revealed. Af-terwards, participants completed a questionnaire (see SupplementaryMaterial S1 for a translated version), which included date of birth, sex,and a self-report of the time per week spent playing video games(“weekly gaming time”; WGT) as well as doing sports (“weekly sportstime”; WST). All values had to be given over the course of the last12 months. The WST was used to control for a potential confound ofsports. Based on the WGT, participants were classified as either NVGPs(WGT b 4 h) or VGPs (WGT≥ 4 h) irrespective of the genre they played.From anecdotal reports of our participants, VGPs tend to play in onelong session once per week, rather than in several short ones everyday. Thus, we chose a threshold of 4 h to include also moderate VGPswhomight play only for one evening per week. This choice was backedup by the distribution of the reported WGTs which had a minimumaround 4 h (see Supplementary Fig. S1b). On average, the WGT in theNVGP group was 0.9 ± 0.2 h (mean ± SE) and in the VGP group12.3 ± 1 h.

Originally wemeasured both sexes, but failed to find enough femaleVGPs. To avoid confounding video game and gender effects, we exclud-ed the females from the analysis. All participants had normal orcorrected to normal vision. All experiments were performed in accor-dance with the Declaration of Helsinki.

2.5. Statistics

The data were processed and analyzed using self-written Matlabscripts. For the comparison of the twomain factors, attention type (tran-sient/sustained) and player group (NVGPs/VGPs), 2-factorial ANOVAswere computed. All post-hoc tests were performed using the Tukey-Kramer method. Results were considered significant at an alpha levelof 0.05. Effects sizes are reported as partial eta-squared (η2).

To support the validity of the classification into transient andsustained attention types, a repeated-measures ANOVA for the within-subject factor CLT (0, 17, 33, 50, 67, 83, 100, 117, 133, 150, 200, 300,400 and 600 ms) and the between-subject factors attention type andplayer group was performed. If the transient type just reaches its peakperformance at an earlier CLT than the sustained type, and both typesachieve a similar level of performance for the later CLTs, no interactionbetween attention type and these later CLTs should be present and thegroup differences should be negligible. The performance at earlierCLTs in the transient type should also be clearly higher. On the otherhand, if the classification indeed separates two attention types generat-ed by different functions as suggested by Nakayama and Mackeben(1989), then the performance differences should be smaller for theearly CLTs between the attention types. However, there should be sig-nificant interactions between attention type and the later CLTs and the

14 D.J. Mack et al. / Acta Psychologica 169 (2016) 11–19

performance should be substantially higher for these CLTs in thesustained type.

For the analysis of potential differences with respect to target posi-tion, a second repeated-measures ANOVA with the within-subject fac-tor cue position (for the 12 clock positions), and the between-subjectfactors attention type and player group was performed.

3. Results

The performances of typical VGPs and NVGPs for both attentiontypes are shown in Fig. 2. For these examples, the VGPs outperformedthe NVGPs with respect to peak, mean, and baseline performances.However, all reached their peak performance at the same CLTs with re-spect to attention type. Whereas both participants of the transient typehave a clear early peak at 33 ms (Fig. 2a), this transient component ismuch less pronounced in the sustained participants, who show a lateplateau around 300 ms instead (Fig. 2b).

Table 1 shows the statistics of all 98male participants, separated into36NVGPs and62VGPs.With respect to attention type,we observed pro-portionally more sustained VGPs (34%) than NVGPs (22%), but this dif-ference did not reach significance (p = 0.223; asymptotic Fisher test).2-factorial ANOVAs revealed no significant differences in age or WSTwith respect to attention type or player group (F's(1,94) ≤ 2.53,p's ≥ 0.115, η2's ≤ 0.026; see Supplementary Fig. S1c and d). For theWGT, the influence of attention type was not significant(F(1,94) b 0.01, p = 0.998, η2 b 0.001), whereas the effect of playergroup was significant by definition (F(1,94) = 57.6, p b 0.001,η2≤ 0.38; see Supplementary Fig. S1a). The analysis of thenumber of in-dividual peak performances showed a significant effect of attention type(F(1,94)=5.93, p=0.017, η2=0.059; see Supplementary Fig. S2)withfewer peaks in the sustained type. VGPs did not differ from NVGPs intheir number of peaks (F(1,94) = 0.39, p = 0.535, η2 = 0.004).

The distribution of the CLTs for peak performance across all partici-pants is shown in Fig. 3. The distribution was clearly bimodal, with atransient peak at 33 ms, a sustained peak around 300 ms and a trougharound 150 ms.

Over all participants, the performance curves of the transient andsustained types were different as shown in Fig. 4. The repeated-measures ANOVA yielded a significant influence of CLT

Fig. 2. Single participant data. (a) Examples of transient attention types: anNVGP (age: 17 yearsWGT: 7 h,WST: 6 h, P0: 76%, PMean: 70%, PMax: 86%, tMax: 33ms). (b) Examples of sustained atten300 ms) and a VGP (age: 17 years, WGT: 10 h, WST: 2 h, P0: 52%, PMean: 63%, PMax: 86%, tMa

performance, tMax: CLT for peak performance, WGT: weekly gaming time, and WST: weekly sp

(F(13,1222)= 3.13, p b 0.001, η2 = 0.032) and formed a significant in-teraction with attention type (F(13,1222) = 8.11, p b 0.001, η2 =0.079). The post-hoc test showed that the sustained type performedbetter at the longer CLTs (150, 200, 300, 400 and 600ms). Neither play-er group nor its interaction with attention type were significant at thesingle CLT level (F's(13,1222) ≤ 1.57, p's ≥ 0.086, η2's ≤ 0.016).

As shown in Fig. 5a, there was no difference in baseline performancewith respect to attention type (F(1,94)=0.01, p=0.919, η2 b 0.001) orplayer group (F(1,94)=0.33, p=0.562,η2=0.004). For themeanper-formance shown in Fig. 5b, the sustained type was slightly better thanthe transient one, although this difference was not significant(F(1,94) = 2.58, p = 0.111, η2 = 0.027). However, player group had asignificant influence on mean performance (F(1,94) = 8.24, p =0.005, η2 = 0.081): VGPs of both attention types performed betterthan NVGPs.

With respect to the cue position, the repeated-measures ANOVA re-vealed a small effect of worsemeanperformance at the lower comparedto the upper right positions (one o'clock position compared to six andseven o'clock) as shown in Fig. 6a and b (F(11,1034) = 3.51,p b 0.001, η2 = 0.036). Interestingly, there was no position-specific ef-fect of player group (F(11,1034) = 1.47, p = 0.137, η2 = 0.015). Simi-larly attention type and its interactionwith player grouphad no effect atthe single position level (F's(11,1034)≤ 1.27, p's≥ 0.235, η2's≤ 0.013).Since there was no main effect of attention type on mean performanceat the overall as well as the position-specific level, we performed afollow-up repeated-measured ANOVA with only player group asbetween-subject factor. This sharpened the general effect of cue posi-tion (F(11,1056) = 4.85, p b 0.001, η2 = 0.048) with better perfor-mance at one, two, three and nine o'clock compared to seven o'clockas well as one and two o'clock compared to six o'clock. But again therewas no position-specific effect of player group (F(11,1056) = 1.17,p = 0.3, η2 = 0.012).

In the case of peak performance shown in Fig. 7a, the effect of atten-tion type was significant (F(1,94) = 7.48, p = 0.007, η2 = 0.074). Thesustained type had higher peak performances than the transient one.The influence of player group was also significant (F(1,94) = 8.44,p = 0.005, η2 = 0.082): VGPs performed better than NVGPs. Since at-tention type had a significant influence on peak performance, we con-firmed these results in an additional regression analysis:

, WGT: 0 h,WST: 4 h, P0: 38%, PMean: 43%, PMax: 71%, tMax: 33ms) and a VGP (age: 17 years,tion types: anNVGP (age: 18 years,WGT: 2 h,WST: 5 h, P0: 43%, PMean: 53%, PMax: 81%, tMax:x: 300 ms). P0: Baseline performance at 0 ms CLT, PMean: mean performance, PMax: Peakorts time.

Table 1Descriptive statistics of the 98male participants. Values are given as:mean± standard er-ror (minimum–maximum).

# Age [years] WST [h] WGT [h] nMax

All 98 18 ± 0.2(15–27)

6 ± 0.4(0−22)

8 ± 0.8(0−30)

1.6 ± 0.1(1–5)

NVGPs Transient 28 18 ± 0.2(16–20)

6 ± 0.9(0–22)

1 ± 0.2(0–3)

1.8 ± 0.2(1–4)

Sustained 8 17 ± 0.4(15–18)

5 ± 1.1(1−12)

1 ± 0.4(0–3)

1 ± 0(1–1)

VGPs Transient 41 18 ± 0.3(16–27)

6 ± 0.6(0–18)

13 ± 1.3(4–30)

1.7 ± 0.2(1–5)

Sustained 21 18 ± 0.3(15–20)

5 ± 0.9(0–16)

12 ± 1.3(5–25)

1.4 ± 0.1(1–3)

#: number of participants, nMax: Number of individual peak performances, WGT: weeklygaming time, and WST: weekly sports time.

15D.J. Mack et al. / Acta Psychologica 169 (2016) 11–19

NVGPs: PMax (tMax) = 0.028 · tMax + 68%, R2 = 0.118, p = 0.04

VPGs: PMax (tMax) = 0.019 · tMax + 76%, R2 = 0.067, p = 0.042.

In both groups, peak performance (PMax) was significantly correlat-ed with the CLT for peak performance (tMax). VGPs' peak performancewas approximately 8% higher. The comparison of the slopes of the twolinear regressions revealed no significant difference (p = 0.303).

The results from the CLTs for peak performance can be seen in Fig.7b. The influence of attention type was significant by definition(F(1,94)=192.74, p b 0.001,η2=0.672). Surprisingly, therewas no ef-fect of player group (F(1,94) = 0.811, p = 0.37, η2 = 0.009).

All unreported interactions of the computed ANOVAs in the resultssection were not significant (p's ≥ 0.113, η2's ≤ 0.012).

4. Discussion

According to our initial research questions, we obtained four results:

(1) VGPs performed significantly better than NVGPs in mean as wellas peak performance.

(2) There were no significant differences in the CLTs for peak perfor-mance of VGPs and NVGPs.

(3) Therewas a tendency for the proportion of VGPs using top-downcontrol of attention to be higher compared to NVGPs. However,this difference did not reach significance.

Fig. 3. Distribution of the cue lead times for peak performance across all 98 subjects. Notethe non-uniform scaling of the x-axis. Bars represent relative frequency of occurrenceswithin each group. The gray vertical line indicates the border for the division of the twoattention types.

(4) All participants performed worse at the lower cue positions butthis effect was not different between player groups.

In the following discussion, we try to interpret our results in view ofsuperior attentional mechanisms in VGPs.

4.1. The VGP advantage cannot be explained by more efficient “buttonpressing”

In line with other performance-based perceptual studies (Bejjankiet al., 2014; Green & Bavelier, 2003; Schubert et al., 2015; West et al.,2008; Wilms et al., 2013), we can exclude a faster stimulus-responsemapping (Castel et al., 2005) as explanation for the superior perfor-mance found in VGPs. Since we did not measure any reaction times,but still found better performance in VGPs, our results lend further sup-port against thenotion of VGPs just being faster in “pressing thebutton”.Therefore, the superior performance of VGPs has to be rooted in alter-ations of the attentional system. Two studies employing the usefulfield of view task attributed the better overall performance of VGPs(Green & Bavelier, 2003, 2006) to an enhanced allocation of spatial at-tention. Since these authors compared performance before and aftervideo game training, they were able to demonstrate a causal relation-ship between video game play and the observed effects.

By modelling the neuronal processes underlying visual motion dis-crimination, Bavelier and colleagues showed that there is no differencein the time necessary for motor preparation and execution with respectto video game play (Green, Pouget, et al., 2010). Instead, VGPs actuallywere integrating necessary information faster than NVGPs. In a similarline of evidence, VGPs performbetter in an attention taskwhichwas ex-plained by a faster processing speed and a lower perceptual threshold(Schubert et al., 2015; Wilms et al., 2013).

4.2. VGPs are not shifting their spotlight of attention faster

Initially, we hypothesized that the shorter reaction times of VGPSarose from faster shifts of attention. Contrary to this idea, VGPs in ourstudy did not exhibit faster covert attention shifts, as indicated by thelack of differences in the CLTs for peak performance. Still, VGPs per-formed better than NVGPs and we offer three possible explanationsfor these results:

i) Faster information processing in VGPsAttention not only increases the discriminability of stimulithrough distractor inhibition or signal enhancement, but also ac-celerates the rate of visual information processing (Carrasco &McElree, 2001). In addition, attention is needed to form preciserepresentations in visual short term memory (Persuh, Genzer,& Melara, 2012), and VGPs possess a better visual sensitivity(Appelbaum, Cain, Darling, & Mitroff, 2013). Since VGPs have amore precise representation of the stimulus, they acquire moreinformation than NVGPs in the same amount of time. Further-more, VGPs possess a lower perceptual threshold, which enablesthem to start earlier with the processing of visual stimuli. Com-bined with a faster processing of these stimuli (Schubert et al.,2015; Wilms et al., 2013) this could lead to the higher perfor-mance seen in VGPs without the need for faster attention shifts.

ii) Higher rate of information accumulationVGPs may not only have an increased rate of information pro-cessing, but indeed a higher rate of information accumulation.Probabilistic inference is increased by action video game playand yields an improvement, for instance, in a simple motion dis-crimination task (Green, Pouget, et al., 2010). The authors of thisstudy estimated an approximately 20% higher rate of informationaccumulation for VGPs which could also account for the betterperformance of VGPs.

Fig. 4. Performance for each CLT across all 98 subjects. The gray shaded areas show 95%confidence intervals for the means of the two attention types (thick black and graylines). The stars at the bottom indicate significant differences between the two attentiontypes for particular CLTs. The thin lines give the means for the player groups at eachattention type (solid: transient, dashed: sustained).

16 D.J. Mack et al. / Acta Psychologica 169 (2016) 11–19

iii) Reduced backward masking in VGPsAnother explanation would be reduced backward masking inVGPs (Li, Polat, Scalzo, & Bavelier, 2010). An attenuated inhibito-ry effect of the mask in combination with the lower perceptualthreshold mentioned above (Schubert et al., 2015; Wilms et al.,2013) would give VGPs more time to extract crucial informationfrom the search array presented for only 17 ms.

The present study cannot ultimately distinguish between thesethree alternatives. However, our data clearly excludes a speeded atten-tion shift or an improved stimulus-response mapping as reason for thesuperior performance of VGPs.

Fig. 5. (a) Baseline performance at 0 ms CLT and (b)mean performance across all CLTs. ANOVAp≤ 0.01). Error bars show groupmeans and according95% confidence intervals. The smallmarknot significant.

4.3. Balance between bottom-up and top-down control

Nakayama and Mackeben (1989) previously concluded that the re-sponse functions observed in their task are the sum of a transient,bottom-up part and a sustained, top-down component. Whenwe sepa-rated our participants by attention type and plotted the mean perfor-mance for each CLT, this became clearly visible. Nakayama speculatedthat the lack of a downturn in the performance curve of some partici-pants may be due to ceiling effects which would flatten the transientcomponent in cases of high mean performance. This is not supportedby our data since the mean performance in the sustained group wasnot significantly different from the transient group. Moreover,Wilschut et al. (2011) proposed that greater task difficulty leads ob-servers to rely more on top-down control at longer CLTs, thus leadingto an increase in the late plateauing component of the performancecurve.

In the present study, the proportion of participants expressing thesustained attention shift was slightly higher in VGPs, although this dif-ference was not significant. Previous studies found better top-downcontrol of attention in VGPs (Cain et al., 2014; Chisholm & Kingstone,2012; Chisholm et al., 2010; Clark et al., 2011), which supports our ob-servation. Nevertheless, additional experimental findings are requiredto prove the notion of better top-down control in VGPs.

4.4. No difference in baseline performance

We included the baseline performance in our analysis as a controlcondition without any attentional influences. No significant differenceswere observed with respect to player group in this condition. Frommonkey physiology, it is known that attention is only able to boost theneuronal response if a stimulus is present in the receptive field (Treue&Maunsell, 1996). Directed attention without a stimulus does not elicita neuronal response. In the present study, the cue could not attract at-tention to the location of the oddball at 0 ms CLT since the oddball ap-peared simultaneously with the stimulus. This prevented anyalterations in attention from affecting performance in this condition.Since our results clearly point in the direction of better attentionalmechanisms in VGPs, the lack of differences in baseline performancealso supports this notion.

results are depicted at the bottom by black horizontal lines and symbols (n.s.: p N 0.05, **:ers represent single participants. Chance level ismarked by thedotted horizontal lines. n.s.:

Fig. 6.Mean performance at each cue position in (a) the transient attention type and (b) the sustained attention type across all 98 subjects. Themarkers and lines indicate themeans of theplayer groups at each of the 12 target positions (see Fig. 1). The shaded areas show according 95% confidence intervals. The black circlesmark themeans pooled over the player groups foreach attention type (transient: 57%, sustained: 62%).

17D.J. Mack et al. / Acta Psychologica 169 (2016) 11–19

4.5. No differential performance in the lower visual field

Schubert et al. (2015) recently found that VGPs possess a higherprocessing speed especially at the lower visual field. Since we didnot find differences in the speed of attention shifts with respect toplayer group, higher processing speed is a likely cause of the ob-served performance differences as discussed above. The results oftheir study would also imply a larger performance difference be-tween VGPs and NVGPs at the lower target positions in the presentstudy. Similar to Schubert et al. (2015) we found generally lowerperformance at the lower target positions across all participants.However, we failed to find any position-specific difference with re-spect to player group. To the contrary, the difference between thegroups became slightly smaller at the lower positions (most visibleat the six o'clock position in Fig. 6a). It has been reported that theasymmetry in attentional resolution is more pronounced betweenthe central and peripheral part of the visual field than it is between

Fig. 7. (a) Peak performance and (b) corresponding cue lead time for peak performance. ANOVAp≤ 0.01, ***: p≤ 0.001). Error bars show groupmeans and according 95% confidence intervals. Thorizontal line. n.s.: not significant.

the lower and upper ones (Ellison & Walsh, 2000). The same, ofcourse, is true for the spatial resolution of the retina. In addition,Ellison andWalsh (2000) also showed, that the asymmetry betweenupper and lower visual field vanishes if the stimulus is presented inboth fields simultaneously. Since Schubert et al. (2015) presentedtheir stimulus array on a grid in the periphery, the different retinaleccentricities together with the higher spatial resolution of VGPs(Green & Bavelier, 2007) might have exaggerated the effects of tar-get position. This explains the complete lack of position-specific ef-fects of player group in our study. Nevertheless, we used only asingle eccentricity of 4° and thus cannot exclude the possibilitythat differences in performance between VGPs and NVGPs becomemore pronounced at more peripheral locations. Future researchmight employ a more fine-grained analysis of the effects of videogame play on attention at different eccentricities and retinal loca-tions. It is noteworthy that vertical differences in performance withrespect to eccentricity (Carrasco, Talgar, & Cameron, 2001) or retinal

results are depicted at the bottom by black horizontal lines and symbols (n.s.: p N 0.05, **:he smallmarkers represent single participants. Chance level ismarked in (a) by the dotted

18 D.J. Mack et al. / Acta Psychologica 169 (2016) 11–19

position (Abrams, Nizam, & Carrasco, 2012) are in the opposite di-rection (better performance in the upper vs. the lower positions) asreported by Schubert and us.

4.6. Basic considerations for the comparison of VGPs and NVGPs

Due to the cross-sectional nature of our study, we cannot determineif the differences between VGPs and NVGPs are causally related to videogame play or are mere reflections of preexisting group disparities. Onemight speculate that people with superior attentional abilities performbetter in video games and thus play more often (Boot, Blakely, &Simons, 2011; Kristjansson, 2013). In this line of thought, the reporteddifferences between the player groups would be the consequence of asampling bias. However, this idea is disputed by increasing evidencefrom multiple training studies, showing a causal relationship betweensuperior performance and video game play (e.g. Franceschini et al.,2013; Green & Bavelier, 2003; Green & Bavelier, 2006, 2007; Kuhn,Gleich, Lorenz, Lindenberger, & Gallinat, 2014; Li et al., 2009; Li et al.,2010; Oei & Patterson, 2013).

Another problem with cross-sectional studies is expectation bias.The knowledge that VGPs are recruited due to their expertise might af-fect their motivation and thus their performance (and vice versa forNVGPs; Boot et al., 2011). We are convinced that our results were notinfluenced by expectation biases because our participants had no priorknowledge about the experiment's real purpose. They also could notinfer the purpose from the video game questionnaire, since it was ad-ministered after completion of the experiment. Thus, participantscould not build up any group-related expectations, which might haveotherwise influenced their performance. Similarly, the experimenterhimself did not know the participant's player group assignment at thetime of the measurement. Therefore, any observer-expectancy effectscould not possibly have contributed to our data.

We also did not include the data of female participants in our analy-sis to avoid confounding gender effects in the NVGP and VGP groups.

The last point addresses the video game genre. It has been increas-ingly emphasized that only these games elicit the beneficial effectsfound in VGPs (Appelbaum et al., 2013; Bavelier, Achtman, et al.,2012; Blacker & Curby, 2013; Cain, Landau, & Shimamura, 2012;Chisholm et al., 2010; Green & Bavelier, 2003; Green, Li, & Bavelier,2010). This is a valid claim, since most action video games are visuallycomplex, fast paced and often involve the parallel execution of severaltasks. Nonetheless, the exact features needed to elicit beneficial effectsare unknown and different genres might have different and possiblycontrary outcomes. A recent study showed that players of real-timestrategy games, which are commonly not defined as action videogames, possess better multiple-object tracking skills than players offirst person shooter games, which are generally accepted as the arche-type of action video games (Dobrowolski, Hanusz, Sobczyk, Skorko, &Wiatrow, 2015). Interestingly, multiple-object tracking has previouslybeen associated with superior performance only in action-VGPs(Green & Bavelier, 2006). Our definition of player group on the otherhand is purely based on the amount of weekly video game play. Thissurely includes different genres and thus might reduce the size of theobserved effects. Nevertheless, this widespread approach is beneficialsince it reduces the possibility to neglect potential effects of a specificsubgroup as for example the real-time strategy games.

5. Conclusion

All together, we provide further evidence against faster stimulus-response mappings as an explanation for higher task performancefound in VGPs. More importantly, our results do not support the as-sumption of faster covert attention shifts as an explanation for the ob-served differences between VGPs and NVGPs. Most likely, acombination of faster information processing and more effective atten-tional control provide the basis for the better performance of VGPs.

Disclosure statement

No competing financial interest exists.

Acknowledgement

We like to thank Ziad Hafed for his valuable support during the ini-tial phase of the study. The first author was financially supported by astipend of the Werner Reichardt Centre of Integrative Neuroscience.

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.actpsy.2016.05.001.

References

Abrams, J., Nizam, A., & Carrasco, M. (2012). Isoeccentric locations are not equivalent: Theextent of the vertical meridian asymmetry. Vision Research, 52(1), 70–78. http://dx.doi.org/10.1016/j.visres.2011.10.016.

Anderson, C. A., Shibuya, A., Ihori, N., Swing, E. L., Bushman, B. J., Sakamoto, A., ... Saleem,M. (2010). Violent video game effects on aggression, empathy, and prosocial behaviorin eastern and western countries: A meta-analytic review. Psychological Bulletin,136(2), 151–173. http://dx.doi.org/10.1037/a0018251.

Appelbaum, L. G., Cain, M. S., Darling, E. F., & Mitroff, S. R. (2013). Action video gameplaying is associated with improved visual sensitivity, but not alterations in visualsensory memory. Attention, Perception, & Psychophysics, 75(6), 1161–1167. http://dx.doi.org/10.3758/s13414-013-0472-7.

Bavelier, D., Achtman, R. L., Mani, M., & Focker, J. (2012a). Neural bases of selective atten-tion in action video game players. Vision Research, 61, 132–143. http://dx.doi.org/10.1016/j.visres.2011.08.007.

Bavelier, D., Green, C. S., Pouget, A., & Schrater, P. (2012b). Brain plasticity through the lifespan: Learning to learn and action video games. Annual Review of Neuroscience, 35,391–416. http://dx.doi.org/10.1146/annurev-neuro-060909-152832.

Bejjanki, V. R., Zhang, R., Li, R., Pouget, A., Green, C. S., Lu, Z. L., & Bavelier, D. (2014). Actionvideo game play facilitates the development of better perceptual templates.Proceedings of the National Academy of Sciences of the United States of America,111(47), 16961–16966. http://dx.doi.org/10.1073/pnas.1417056111.

Blacker, K. J., & Curby, K. M. (2013). Enhanced visual short-term memory in action videogame players. Attention, Perception, & Psychophysics, 75(6), 1128–1136. http://dx.doi.org/10.3758/s13414-013-0487-0.

Boot, W. R., Blakely, D. P., & Simons, D. J. (2011). Do action video games improve percep-tion and cognition? Frontiers in Psychology, 2, 226. http://dx.doi.org/10.3389/fpsyg.2011.00226.

Brainard, D. H. (1997). The psychophysics toolbox. Spatial Vision, 10(4), 433–436. http://dx.doi.org/10.1163/156856897x00357.

Cain, M. S., Landau, A. N., & Shimamura, A. P. (2012). Action video game experience re-duces the cost of switching tasks. Attention, Perception, & Psychophysics, 74(4),641–647. http://dx.doi.org/10.3758/s13414-012-0284-1.

Cain, M. S., Prinzmetal, W., Shimamura, A. P., & Landau, A. N. (2014). Improved control ofexogenous attention in action video game players. Frontiers in Psychology, 5, 69.http://dx.doi.org/10.3389/fpsyg.2014.00069.

Carrasco, M., & McElree, B. (2001). Covert attention accelerates the rate of visual informa-tion processing. Proceedings of the National Academy of Sciences of the United States ofAmerica, 98(9), 5363–5367. http://dx.doi.org/10.1073/pnas.081074098.

Carrasco, M., Talgar, C. P., & Cameron, E. L. (2001). Characterizing visual performancefields: Effects of transient covert attention, spatial frequency, eccentricity, task andset size. Spatial Vision, 15(1), 61–75. http://dx.doi.org/10.1163/15685680152692015.

Castel, A. D., Pratt, J., & Drummond, E. (2005). The effects of action video game experienceon the time course of inhibition of return and the efficiency of visual search. ActaPsychologica, 119(2), 217–230. http://dx.doi.org/10.1016/j.actpsy.2005.02.004.

Chiappe, D., Conger, M., Liao, J., Caldwell, J. L., & Vu, K. P. (2013). Improving multi-taskingability through action videogames. Applied Ergonomics, 44(2), 278–284. http://dx.doi.org/10.1016/j.apergo.2012.08.002.

Chisholm, J. D., & Kingstone, A. (2012). Improved top-down control reduces oculomotorcapture: The case of action video game players. Attention, Perception, &Psychophysics, 74(2), 257–262. http://dx.doi.org/10.3758/s13414-011-0253-0.

Chisholm, J. D., Hickey, C., Theeuwes, J., & Kingstone, A. (2010). Reduced attentional cap-ture in action video game players. Attention, Perception, & Psychophysics, 72(3),667–671. http://dx.doi.org/10.3758/APP.72.3.667.

Clark, K., Fleck, M. S., & Mitroff, S. R. (2011). Enhanced change detection performance re-veals improved strategy use in avid action video game players. Acta Psychologica,136(1), 67–72. http://dx.doi.org/10.1016/j.actpsy.2010.10.003.

Dobrowolski, P., Hanusz, K., Sobczyk, B., Skorko, M., & Wiatrow, A. (2015). Cognitive en-hancement in video game players: The role of video game genre. Computers inHuman Behavior, 44, 59–63. http://dx.doi.org/10.1016/j.chb.2014.11.051.

Donohue, S. E., Woldorff, M. G., & Mitroff, S. R. (2010). Video game players show moreprecise multisensory temporal processing abilities. Attention, Perception, &Psychophysics, 72(4), 1120–1129. http://dx.doi.org/10.3758/APP.72.4.1120.

19D.J. Mack et al. / Acta Psychologica 169 (2016) 11–19

Dye, M. W., Green, C. S., & Bavelier, D. (2009). Increasing speed of processing with actionvideo games. Current Directions in Psychological Science, 18(6), 321–326. http://dx.doi.org/10.1111/j.1467-8721.2009.01660.x.

Ellison, A., & Walsh, V. (2000). Visual field asymmetries in attention and learning. SpatialVision, 14(1), 3–9. http://dx.doi.org/10.1163/156856801741323.

Ferguson, C. J., San Miguel, C., Garza, A., & Jerabeck, J. M. (2012). A longitudinal test ofvideo game violence influences on dating and aggression: A 3-year longitudinalstudy of adolescents. Journal of Psychiatric Research, 46(2), 141–146. http://dx.doi.org/10.1016/j.jpsychires.2011.10.014.

Franceschini, S., Gori, S., Ruffino, M., Viola, S., Molteni, M., & Facoetti, A. (2013). Actionvideo games make dyslexic children read better. Current Biology, 23(6), 462–466.http://dx.doi.org/10.1016/j.cub.2013.01.044.

Gentile, D. A., Choo, H., Liau, A., Sim, T., Li, D., Fung, D., & Khoo, A. (2011). Pathologicalvideo game use among youths: A two-year longitudinal study. Pediatrics, 127(2),e319–e329. http://dx.doi.org/10.1542/peds.2010-1353.

Green, C. S., & Bavelier, D. (2003). Action video game modifies visual selective attention.Nature, 423(6939), 534–537. http://dx.doi.org/10.1038/nature01647.

Green, C. S., & Bavelier, D. (2006). Enumeration versus multiple object tracking: The caseof action video game players. Cognition, 101(1), 217–245. http://dx.doi.org/10.1016/j.cognition.2005.10.004.

Green, C. S., & Bavelier, D. (2007). Action-video-game experience alters the spatial resolu-tion of vision. Psychological Science, 18(1), 88–94. http://dx.doi.org/10.1111/j.1467-9280.2007.01853.x.

Green, C. S., Li, R., & Bavelier, D. (2010a). Perceptual learning during action video gameplaying. Topics in Cognitive Science, 2(2), 202–216. http://dx.doi.org/10.1111/j.1756-8765.2009.01054.x.

Green, C. S., Pouget, A., & Bavelier, D. (2010b). Improved probabilistic inference as a gen-eral learning mechanism with action video games. Current Biology, 20(17),1573–1579. http://dx.doi.org/10.1016/j.cub.2010.07.040.

Grizzard, M., Tamborini, R., Lewis, R. J., Wang, L., & Prabhu, S. (2014). Being bad in a videogame can make us more morally sensitive. Cyberpsychology, Behavior and SocialNetworking. http://dx.doi.org/10.1089/cyber.2013.0658.

Heimler, B., Pavani, F., Donk, M., & van Zoest, W. (2014). Stimulus- and goal-driven con-trol of eye movements: Action videogame players are faster but not better. Attention,Perception, & Psychophysics. http://dx.doi.org/10.3758/s13414-014-0736-x.

Kristjansson, A. (2013). The case for causal influences of action videogame play upon vi-sion and attention. Attention, Perception, & Psychophysics, 75(4), 667–672. http://dx.doi.org/10.3758/s13414-013-0427-z.

Kuhn, S., Gleich, T., Lorenz, R. C., Lindenberger, U., & Gallinat, J. (2014). Playing SuperMario induces structural brain plasticity: Graymatter changes resulting from trainingwith a commercial video game. Molecular Psychiatry, 19(2), 265–271. http://dx.doi.org/10.1038/mp.2013.120.

Latham, A. J., Patston, L. L., & Tippett, L. J. (2013). The virtual brain: 30 years of video-gameplay and cognitive abilities. Frontiers in Psychology, 4, 629. http://dx.doi.org/10.3389/fpsyg.2013.00629.

Li, R., Polat, U., Makous, W., & Bavelier, D. (2009). Enhancing the contrast sensitivity func-tion through action video game training. Nature Neuroscience, 12(5), 549–551. http://dx.doi.org/10.1038/nn.2296.

Li, R., Polat, U., Scalzo, F., & Bavelier, D. (2010). Reducing backward masking through ac-tion game training. Journal of Vision, 10(14). http://dx.doi.org/10.1167/10.14.33.

Mack, D. J., & Ilg, U. J. (2014). The effects of video game play on the characteristics of sac-cadic eye movements. Vision Research, 102, 26–32. http://dx.doi.org/10.1016/j.visres.2014.07.010.

Mishra, J., Zinni, M., Bavelier, D., & Hillyard, S. A. (2011). Neural basis of superior perfor-mance of action videogame players in an attention-demanding task. The Journal ofNeuroscience, 31(3), 992–998. http://dx.doi.org/10.1523/JNEUROSCI.4834-10.2011.

Nakayama, K., &Mackeben,M. (1989). Sustained and transient components of focal visualattention. Vision Research, 29(11), 1631–1647. http://dx.doi.org/10.1016/0042-6989(89)90144-2.

Oei, A. C., & Patterson, M. D. (2013). Enhancing cognition with video games: A multiplegame training study. PloS One, 8(3), e58546. http://dx.doi.org/10.1371/journal.pone.0058546.

Pelli, D. G. (1997). The VideoToolbox software for visual psychophysics: Transformingnumbers into movies. Spatial Vision, 10(4), 437–442. http://dx.doi.org/10.1163/156856897x00366.

Persuh, M., Genzer, B., & Melara, R. D. (2012). Iconic memory requires attention. Frontiersin Human Neuroscience, 6, 126. http://dx.doi.org/10.3389/fnhum.2012.00126.

Posner, M. I. (1980). Orienting of attention. The Quarterly Journal of ExperimentalPsychology, 32(1), 3–25. http://dx.doi.org/10.1080/00335558008248231.

Posner, M. I., Snyder, C. R., & Davidson, B. J. (1980). Attention and the detection of signals.Journal of Experimental Psychology, 109(2), 160–174. http://dx.doi.org/10.1037//0096-3445.109.2.160.

Rideout, V. J., Foehr, U. G., & Roberts, D. F. (2010). Generation M: Media in the lives of 8 to18-year-olds. Menlo Park, California: Kaiser Family Foundation.

Rivero, T. S., Covre, P., Reyes, M. B., & Bueno, O. F. (2013). Effects of chronic video game useon time perception: Differences between sub- and multi-second intervals.Cyberpsychology, Behavior and Social Networking, 16(2), 140–144. http://dx.doi.org/10.1089/cyber.2012.0103.

Schlickum, M. K., Hedman, L., Enochsson, L., Kjellin, A., & Fellander-Tsai, L. (2009). Sys-tematic video game training in surgical novices improves performance in virtual re-ality endoscopic surgical simulators: A prospective randomized study. World Journalof Surgery, 33(11), 2360–2367. http://dx.doi.org/10.1007/s00268-009-0151-y.

Schubert, T., Finke, K., Redel, P., Kluckow, S., Muller, H., & Strobach, T. (2015). Video gameexperience and its influence on visual attention parameters: An investigation usingthe framework of the Theory of Visual Attention (TVA). Acta Psychologica, 157,200–214. http://dx.doi.org/10.1016/j.actpsy.2015.03.005.

Shepherd, M., Findlay, J. M., & Hockey, R. J. (1986). The relationship between eye move-ments and spatial attention. The Quarterly Journal of Experimental Psychology A,38(3), 475–491. http://dx.doi.org/10.1080/14640748608401609.

Strobach, T., Frensch, P. A., & Schubert, T. (2012). Video game practice optimizes executivecontrol skills in dual-task and task switching situations. Acta Psychologica, 140(1),13–24. http://dx.doi.org/10.1016/j.actpsy.2012.02.001.

Theeuwes, J. (2010). Top-down and bottom-up control of visual selection. ActaPsychologica, 135(2), 77–99. http://dx.doi.org/10.1016/j.actpsy.2010.02.006.

Theeuwes, J. (2013). Feature-based attention: It is all bottom-up priming. PhilosophicalTransactions of the Royal Society of London. Series B, Biological Sciences, 368(1628),20130055. http://dx.doi.org/10.1098/rstb.2013.0055.

Treisman, A. M., & Gelade, G. (1980). A feature-integration theory of attention. CognitivePsychology, 12(1), 97–136. http://dx.doi.org/10.1016/0010-0285(80)90005-5.

Treue, S., & Maunsell, J. H. (1996). Attentional modulation of visual motion processing incortical areas MT and MST. Nature, 382(6591), 539–541. http://dx.doi.org/10.1038/382539a0.

West, G. L., Stevens, S. A., Pun, C., & Pratt, J. (2008). Visuospatial experience modulates at-tentional capture: Evidence from action video game players. Journal of Vision, 8(16),1311–1319. http://dx.doi.org/10.1167/8.16.13.

Wilms, I. L., Petersen, A., & Vangkilde, S. (2013). Intensive video gaming improvesencoding speed to visual short-term memory in young male adults. ActaPsychologica, 142(1), 108–118. http://dx.doi.org/10.1016/j.actpsy.2012.11.003.

Wilschut, A., Theeuwes, J., & Olivers, C. N. (2011). The time course of attention: Selectionis transient. PloS One, 6(11), e27661. http://dx.doi.org/10.1371/journal.pone.0027661.

Wu, S., Cheng, C. K., Feng, J., D'Angelo, L., Alain, C., & Spence, I. (2012). Playing a first-person shooter video game induces neuroplastic change. Journal of CognitiveNeuroscience, 24(6), 1286–1293. http://dx.doi.org/10.1162/jocn_a_00192.