Elsevier

Acta Psychologica

Volume 139, Issue 2, February 2012, Pages 327-334
Acta Psychologica

Action video game players form more detailed representation of objects

https://doi.org/10.1016/j.actpsy.2011.12.002Get rights and content

Abstract

Previous research has clearly demonstrated action video game improvements in visual and spatial attention. The present study investigated action video game related changes in the resolution of representations for both dynamic and stationary objects by comparing video game players (VGP) and non-video game players (NVGP). In a color wheel task (adapted from Zhang & Luck, 2008) where viewers were asked to freely recall the color of briefly presented objects, we found that VGPs were more accurate than NVGPs. Furthermore, in the Multiple Identity Tracking task (Horowitz et al., 2007), we found that VGPs were able to track not only more objects but also maintain identity of tracked objects better than NVGPs. Finally, we demonstrated that VGPs had greater attentional breadth and higher spatial representation resolution.

Highlights

► We compared action video game players (VGP) and non-video game players (NVGP). ► We examined visual representations of dynamic and stationary objects. ► VGPs outperformed the NVGPs in tasks that measure visual and spatial attention. ► VGPs were found to have more detailed object representations than NVGPs. ► VGPs were found to have enhanced spatial representations than NVGPs.

Introduction

In the last two decades, there has been a growing interest in the impact of action video games on cognitive skills. Action video games typically require players to process rapidly moving, peripherally located objects as efficiently as slow moving and centrally located ones; additionally players have to simultaneously track and process multiple objects. It is now known that playing such demanding video games improves performance on various tests of visual perception and attention. For instance, video game players (VGPs) are better at localizing targets in both the central and peripheral visual fields (Feng et al., 2007, Green and Bavelier, 2006a), and are faster at target detection compared to non-video game players (NVGPs) (Castel, Pratt, & Drummond, 2005). VGPs can also recover faster from attentional capture (Chisholm et al., 2010, Green and Bavelier, 2003, Green and Bavelier, 2007) and can simultaneously apprehend and track a greater number of items compared to NVGPs (Green and Bavelier, 2006b, Trick et al., 2005). In more perceptual domains, VGPs also outperform NVGPs; they are better at change detection (Clark, Fleck, & Mitroff, 2011), detecting changes in the direction of object movement (West, Stevens, Pun, & Pratt, 2008) and object contrast (Caplovitz and Kastner, 2009, Li et al., 2009). Moreover, a recent study showed that action video game related benefits transfer to non-visual auditory domains (Donohue, Woldorff, & Mitroff, 2010). Training studies have further confirmed that the aforementioned improvements are caused by video game playing and do not merely reflect higher ability individuals choosing to play video games (Achtman, Green, & Bavelier, 2008; for a review on training effects see Green and Bavelier, 2003, Green and Bavelier, 2006a, Green and Bavelier, 2006b, Green and Bavelier, 2007).

Even though much research has identified video game related improvements in visual and spatial attention abilities, no research to date has investigated whether there are action video game related changes in how objects are represented. Since action video games require fast and accurate identification of objects and their features, they are likely to influence how objects are processed and represented. The present study to our knowledge is one of the first studies that investigated the effects of playing action video games on the processing of both dynamic and stationary objects in visual short-term memory. In the next two sections, we briefly outline the main approaches to study dynamic and stationary object representations and discuss findings on the effect of action videogame playing in each domain.

Dynamic object processing refers to one's ability to process a constantly moving and/or changing object. One key paradigm through which dynamic object processing has been studied is Multiple Object Tracking (MOT). The MOT task requires one to track identical and randomly moving targets among identical distractors. Typically, people can track at least four items among eight identical ones for 10 s, with 85–90% accuracy1 (Pylyshyn, 2004). Even though much research has employed this particular paradigm, to date, there is no consensus on the mechanisms of how people are able to track multiple objects (for a review see Cavanagh & Alvarez, 2005). A major point of departure between views has been the extent to which MOT really involves the tracking of content-addressable individual objects. In other words, is MOT performance governed purely by pre-attentive processes or does VSTM processes also contribute to it?

According to the Fingers of Instantiation (FINST) theory, viewers can pre-attentively individuate objects and tag their locations without actually identifying them (Pylyshyn, 1989, Pylyshyn, 2004). In contrast, other researchers argue that one's ability to track objects is based on available attentional and visual short-term memory resources2 (Kahneman and Treisman, 1984, Kahneman et al., 1992, Yantis, 1992).

One of the strongest pieces of evidence arguing that object information is accessible to participants comes from studies employing the Multiple Identity Tracking (MIT) task (Horowitz et al., 2007, Oksama and Hyönä, 2004; but see Pylyshyn, 2004 as well). In the MIT task, instead of identical items, unique objects are tracked, making it a closer analogue of real-world tracking tasks. In the study done by Horowitz et al. (2007), there were 2 conditions in which participants tracked unique cartoon animals. In the Standard condition, participants reported the locations of all cued animals, making it analogous to the classic MOT task. However, in the Specific condition, viewers had to indicate where a specific animal was last seen, once the objects disappeared after the tracking phase. Performance was higher in the Standard than in the Specific condition. More critically, Horowitz et al. (2007) demonstrated that participants were able to accurately identify objects in the Specific condition. They took this finding to suggest that tracking partially depends on one's ability to maintain object information in visual short-term memory.

While previous research had investigated the effects of action video game playing on one's tracking ability, these studies’ exclusive reliance on the MOT task, made it impossible to make direct inferences about the effects of action video game playing on dynamic object processing. Nevertheless, these studies have suggested that VGPs and NVGPs may differ in how well they track unique objects. In MOT tasks, VGPs were shown to track two more items than NVGPs; a benefit explained by changes in VSTM processes, rather than in pre-attentive processes (Green and Bavelier, 2006b, Trick et al., 2005). If action video game playing results in changes in VSTM processes and the MIT task requires the tracking and maintenance of unique object information, it is possible that VGPs would outperform NVGPs in the MIT task. To test this idea, we compared VGPs and NVGPs on the MIT task. Consistent with previous research we expected participants to successfully track more items in the Standard than in the Specific condition (e.g. Horowitz et al., 2007). Given that the Standard condition in the MIT task was designed to be analogous to the MOT task, we expected VGPs to do better than NVGPs on the Standard condition. And finally, we predicted that VGPs may perform disproportionately better in the Specific condition than NVGPs because of the greater visual short-term memory resources they are thought to have.

If action video game playing influences how objects are represented in VSTM, then we expected that there may also be differences between VGPs and NVGPs in tasks which require maintenance of stationary objects. Indeed Boot and colleagues reported that VGPs outperformed NVGPs on a simple change detection task, with the video game related benefits getting larger at higher set sizes (Boot, Kramer, Simons, Fabiani, & Gratton, 2008). This finding suggests that VGPs have a larger VSTM capacity. However, no research to date has examined whether action video game playing improves the quality of VSTM representations. Thus, in this study we specially investigated action video game related improvements in the resolution of object representations.

Typical change detection tasks are suitable for determining VSTM capacity (Alvarez and Cavanagh, 2004, Luck and Vogel, 1997), and attentional demands on object representations (e.g. Johnson et al., 2008, Wheeler and Treisman, 2002), however they typically provide only an indirect means to the nature of representations in VSTM (but see Awh, Barton, & Vogel, 2007). Recently, Zhang and Luck (2008) used the color wheel task, where participants first studied a number of colored squares, and were then asked to choose the color of the probed square from a color wheel including 180 color options (see also Wilken & Ma, 2004). In this task, the dependent variable was the distance between the chosen and the studied color in a 3 dimensional color space, yielding a continuous measure of error in representation resolution (Zhang & Luck, 2008). In the current study we adopted the color wheel task to specifically investigate differences in representation resolution between VGPs and NVGPs. Since our primary goal was to determine differences in resolution we chose to keep set size constant at 3 items, within the limits of average VSTM capacity (Alvarez and Cavanagh, 2004, Cowan, 2001, Luck and Vogel, 1997). We expected that VGPs may form higher resolution representations than NVGPs; thus we investigated whether VGPs were more accurate at identifying colors of probed objects.

In addition to the visual short-term memory measures, we also investigated action video game playing benefits in the spatial domain. Previous research, using a Useful Field of View (UFOV) task, has shown that VGPs have wider attentional breadth (Feng et al., 2007, Green and Bavelier, 2006a). In the UFOV task, viewers are typically presented with 24 positions, arranged in 3 co-centric circles. A target appears in one of these positions for a very brief period of time (< 100 ms) and then viewers are asked to identify the correct position after a brief delay (~ 1000 ms). Performance is usually worse at outer than inner eccentricities and VGPs perform significantly better than NVGPs at all eccentricities, with action-video game playing benefits being larger at outer eccentricities. We expected to replicate these gains. Furthermore, we explored the nature of errors to determine the source of differences between VGPs and NVGPs. We specifically checked whether on error trials participants chose a neighboring position to the target or a random position out of the 24 possible locations. If participants were choosing a neighboring position as the target, this suggests that they had probably detected the initial target and formed a representation, albeit at a less-than perfect resolution. On the other hand, randomly chosen options suggest that the target was not detected at all, or the initial representation formed had an unacceptably poor resolution given the particular task demands.

Section snippets

Participants

Twenty-four male VGPs (Age = 19.9 ± .97) and 20 male NVGPs (Age = 21.1 ± 1.39) participated in the experiment. All participants were Bogazici University students, and they were given course credit for participating. Participants were chosen as in the Green and Bavelier, 2003, Green and Bavelier, 2006a, Green and Bavelier, 2006b, Green and Bavelier, 2007. VGPs reported that they had been playing action video games for at least a minimum of 3 to 4 days a week, no less than an hour per day for the last 6 

Multiple Identity Tracking (MIT) Task

For the MIT task, the dependent variable was the average number of correctly tracked items in both the Standard and Specific conditions. VGPs outperformed the NVGPs in both conditions (see Fig. 4). To compare each participant's performance across standard and specific conditions, we calculated the proportion of correct responses; thus, for both tasks scores ranged between 0 and1.

In the Standard condition, both groups performed close to ceiling, yet VGPs tracked significantly more objects than

Discussion

In this study we investigated the effects of action video game playing on object and spatial representations by comparing VGPs and NVGPs on both dynamic and stationary tasks. VGPs outperformed NVGPs in all domains tested. Specifically, VGPs (1) successfully tracked more objects in both the Standard and Specific conditions of the MIT task, (2) represented colors with greater precision in the Color Wheel task and (3) more accurately located targets across the visual field in the UFOV task, than

Funding

This research was partially supported by Bogazici University's BAP (09B703P), TUBITAK (109 K513), and TUBA-GEBİP funds granted to A.B. Authors declare no conflict of interest.

Acknowledgements

We would like to thank Todd Horowitz for sharing the MIT task with us. We would also like to thank Burak Guclu and Esra Mungan for comments on the project.

References (42)

  • G.A. Alvarez et al.

    How many objects can you track? Evidence for a resource-limited attentive tracking mechanism

    Journal of Vision

    (2007)
  • E. Awh et al.

    Visual working memory represents a fixed number of items, regardless of complexity

    Psychological Science

    (2007)
  • K. Ball et al.

    The useful field of view test: A new technique for evaluating age-related declines in visual function

    Journal of the American Optometric Association

    (1992)
  • R.W. Boot et al.

    The effects of video game playing on attention, memory, and executive control

    Acta Psychologica

    (2008)
  • D.H. Brainard

    The Psychophysics Toolbox

    Spatial Vision

    (1997)
  • G.P. Caplovitz et al.

    Carrot sticks or joysticks: Video games improve vision

    Nature Neuroscience

    (2009)
  • D.A. Castel et al.

    The effects of action video game experience on the time course of inhibition of return and the efficiency of visual search

    Acta Psychologica

    (2005)
  • J.D. Chisholm et al.

    Reduced attentional capture in action video game players

    Attention, Perception, & Psychophysics

    (2010)
  • N. Cowan

    The magical number 4 in short-term memory: A reconsideration of mental storage capacity

    The Behavioral and Brain Sciences

    (2001)
  • S.E. Donohue et al.

    Video game players show more precise multisensory temporal processing abilities

    Attention, Perception, & Psychophysics

    (2010)
  • M.W.G. Dye et al.

    Increasing speed of processing with action video games

    Current Directions in Psychological Science

    (2009)
  • Cited by (60)

    • Exploring staircases as architectural cues in virtual vertical navigation

      2020, International Journal of Human Computer Studies
    • Enhancing Attentional Control: Lessons from Action Video Games

      2019, Neuron
      Citation Excerpt :

      These include standard visuo-spatial working memory tasks, such as the Corsi block tapping task, to complex span tasks, to various spatial forms of the N-back task (Appelbaum et al., 2013; Blacker and Curby, 2013; Waris et al., 2019; Wilms et al., 2013). For example, in terms of reasonably simple spatial memory tasks, Sungur and Boduroglu (2012) found action video game players to have more precise visual working memory for colors than non-gamers on an adapted color-wheel task. Similarly, Green and Bavelier (2006) reported enhanced performance on an object enumeration task both in cross-sectional work comparing AVGPs with NVGPs as well as in an intervention study testing the causal role of action video game training.

    View all citing articles on Scopus
    View full text