M8, T2, Attention and video game training Flashcards
Research questions in video game training (Green and Bavelier, 2003)
- Are there differences in visual attention skills for regular video game players (VGPs) compared to non-video game players (NVGPs)? That is, do VGPs have more attentional resources?
- What happens to the visual attention skills of NVGPs if they are trained on action video games?
Flanker task
“Which way is the central arrow pointing?”
- measure reaction time for compatible versus incompatible trials to see how attention spotlight is deployed
- used to measure attentional breadth/level of attentional resources a given individual possesses by measuring distractor processing
*look up image
Flanker high v low perceptual load task
- Six rings are shown on the computer screen on each trial (100ms)
- Participants decide if either a diamond or square (targets) was shown on each trial inside one of the rings
- Ignore singleton distractor shape presented outside of the six rings.
- Low and high perceptual load manipulated using additional distractors inside the rings
*look up image
Compatibility effects in flanker high v low perceptual load task
Low perceptual load -> compatibility effect is large (faster to process target compatible v incompatible trials)
-> Attentional resources free to process the distractors - broader attention breadth
High perceptual load -> compatibility effect small.
-> Less residual attentional resources for processing irrelevant distractors - narrow attention breadth
Flanker task and VGP v NVGP Green and Bavelier (2003)
- Compare VGPs and NVGPs on flanker task performance across low and high perceptual load
- If video game playing enhances attentional resources, VGPs should show larger distractor effects compared to NVGPs across low and high perceptual load tasks (i.e. have a broader attention breadth).
-> in NVGPs the flanker compatibility effect decreased with increased perceptual load
-> in VGPs the compatibility effect remained with increasing perceptual load
-> Playing video games is associated with a broader scope of attention in the flanker task (i.e. more attentional resources left over)
Enumeration task
- measures how much visual information we can process in a single moment
- Between 1 and 12 squares are presented on the screen for 50 milliseconds
- Participants’ task is to indicate the number of squares shown on each trial
Subitizing range = number of items apprehended at the same time without error
-> Subitizing range predicts how many items can be attended to at once (Most adults - value is 3 or 4 items)
-> Can also be used to measure counting ability when subitizing range is exceeded
Enumeration task, VGP v NVGP Green and Bavelier (2003)
- VGP could subitize more squares than NVGP
- average number items subitized was 4.9 for VGP v 3.3 items for NVGP
- overall, VGP more accurate at the task (78% v 65%)
-> suggest VGP have a broader scope of attention
Functional field of view - size of visual field
- Flanker task and Enumeration Task only measure attention over a small visual field (the size of the visual field we usually play video games on)
- Green and Bavelier (2003) argue that the functional field of view task can be used to see if attention also differs over a larger area of the visual field for VGPs and NVGPs
Functional field of view task
- Participants are briefly shown a display of 24 boxes that subtend 10, 20 or 30° of visual angle from the centre of the visual field.
- One box contains a target shape
- A visual mask is presented to avoid after images
- Participant indicates which spoke on the wheel the target (triangle within a circle) was located.
Functional field of view task, VGP v NVGP Green and Bavelier (2003)
- VGP have much higher accuracy for the task compared to NVGP at all eccentricities (distances)
- suggests that they have much broader scope of attention
- therefore, superior attention extends beyond the visual field in which video games are played
Attention blink task
- Measure of attention over time
- Participants presented sequentially with a rapid stream of letters in the same spatial location
- Letters are shown for 15 msec and the next letter appears 100 msec from the time the previous letter appears
- Most letters in the stream are distractors
- Participants detect two targets in the stream (T1 and T2) was X present?
Detection of the target, attention blink task
- usually T1 accuracy is high
- T2 accuracy varies, depends on how many distractor stimuli are presented between T1 and T2
- hypothesised that the attentional blink might occur due to attentional bottlenecking (Chun and Potter, 1995)
- T1 is attended to via ‘ filtering’ and gets through the ‘bottleneck’ to conscious awareness
- while this is happening, T2 has already occurred and does not get through the ‘bottleneck’
Attention blink task, NVGP v VGP Green and Bavelier (2003)
- VGP have enhanced attention across time- therefore VGP show a reduced attentional blink compared to NVGP
Green and Bavelier (2003) final experiment VGP v NVGP
- Participants’ performance on the enumeration, functional field of view task, and the attentional blink task was measured at baseline.
- Participants then completed 10 x 1-hour long sessions playing either an action video game (Medal of Honour) or a puzzle game (Tetris).
- Participants’ performance on the enumeration, functional field of view task, and the attentional blink task was measured after training
Findings from the final experiment, Green and Bavelier
- improved performance on all three tasks for the action video game training group, Tetris group no improvements
- enumeration task saw a increase of 1.7 items for those from action video game group
- at each eccentricity (ring of functional field of view) there were improvements for action video game group
- attentional blink was smaller after training, suggesting attention deployed more efficiently across time for action video game group
Bavelier et al (2012) research questions
- What are the neural correlates associated with improved attention from playing video games?
- Examined top-down attention by measuring BOLD signal (fMRI) in frontoparietal regions of the brain (top down control & regulate attention) as well as visual areas MT/MST (motion processing) for VGPs and NVGPs
Bavelier et al (2012) method low v high perceptual load visual search tasks
Video game and non-video game players completed low versus high perceptual load visual search tasks in the presence or absence of a distracting visual motion display
-> is a square or diamond present in the array?
- Perceptual load manipulation intended to vary level of top down attention control exerted via dorsal frontoparietal networks in the brain (FEF and IPS)
-> Higher load = more frontoparietal activation
Bavelier et al 2012, method motion task
- Motion task (random dots kinematogram) used to measure where attention spotlight was focused during the task.
- RDK processed by neural area V5/MT
- If RDK is in the focus of attention, activation in V5/MT will be increased compared to RDK outside of the attention focus
Bavelier et al 2012, participants
- Participants were 26 males (12 VGPs and 14 NVGPs)
- NVGPs - < 1hr action video game per week in last 12 months
- VGPs - played action video games minimum of 5 hrs per week in last 12 months
- Trained participants on flanker task prior to fMRI scanning
- fMRI scans during flanker task performance
Behavioural and imaging results Bavelier et al 2012
- The effect of perceptual load on RT was the same in both groups, but shorter RTs for VGPS than NVGPs
- But some evidence that the effect of irrelevant motion in periphery more disruptive for NVGPs than VGPs
- NVGPs - ↑ frontoparietal network activation with ↑ perceptual load
- VGPs – showed some ↑ activation in some regions with ↑ perceptual load
- Overall, VGPs showed lower levels of MT/MST activation than NVGPs
-> This may reflect improved distractor suppression in the VGP group (better allocation of space based attention to targets)
Bavelier et al 2012 conclusions
VGPs more efficient than NVGPs in operation of:
- Frontoparietal attention networks
- Visual processing
- Filtering out irrelevant movement information
