Lecture 13 - Training Attention? Flashcards
central executive
idea that you have a “you in charge” making the correct detection and selecting the correct response
yo can only have one response at a time
making decisions one at a time
responsible for shifting attention to the next task
attentional blink
you’re attentional system seems to pause when targets are rapidly presented
places a limit on attention
consistent with central executive bottleneck
can we change our ability to interact at the attentional level?
Suppose you are performing two tasks, but the second task is highly practiced (automatized). Where does the Psychological Refractory Period model predict the “Central control’ processing for Task 2 will begin?
central control
What happens is we make Task 2 harder (e.g. more sensory processing required)?
This shows that a bottleneck exists in central (executive) control, and is
not purely a function of sensory processing.
doesn’t matter, the central bottleneck remains: you can only make one decision and then you switch (when the next central control process can begin)
switch cost is fixed: minimum amount
this is a classic bottleneck.
If you can only perform one response selection at a time
(regardless of the task load),
• It suggests that the task choice/decisions use a common cognitive resource.
- constraint on what you can do
The central executive bottleneck may seem like a
disadvantage, but it can be very beneficial.
• It allows simultaneous sensory processing of different stimuli, maybe helping prime certain categories. [Bridging
early and late selection theories]
- not as much cost: bottleneck just stopping you from making that final task
• It may help separate motor commands so you don’t make errors during complex tasks (e.g. imagine trying to say two things at once).
• Importantly, it allows you to prioritize and plan actions,
stopping or revising in real time.
- if you have a number of diff tasks coming in (competing stimuli) you will choose the best action for that moment: match action to context (huge advantage)
areas in the brain where executive control is
- anterior cingulate gyrus: heavily involved in task difficulty
- prefrontal cortex (executive control, will power): area in the brain where this is housed
Bottlenecks while driving!
• The 100-car Naturalistic Driving Study
2006
placed dash-cameras in front and back windows.
− Documented 82 crashes and 771 near crashes in more than 2 million miles of driving.
− In 80% of crashes and 67% of near crashes, the driver was inattentive 3 seconds before event. Not enough time to switch back and make decision (e.g. brake!).
when something is automatized you give up control so the central executive may not be as
effective
If you were able to truly automatize a task so that it carried no attentional
load and was automatically completed without the central executive
deciding, this would be called a ______________________?
Fixed-action pattern
nothing to choose, inhibit or prioritize
straight from sensory to motor
Strayer and Johnston (2001)
central executive bottleneck with driving task
used a simulated driving task to test driving performance with cell phones and ‘hands-free’ cell phones (compared to listening to the radio).
- Brake as quickly as you can when you see the red light.
- Single task = just driving.
- Participants missed more than twice as many red lights if using cell phone.
- Braking time also increased.
• Results were the same with ‘hands-free’ phone.
- when you’re talking with someone there’s a heavy load on visio-spacial capacity: imagining what the speaker is talking about instead of focusing attention on immediate surroundings
- central control will always be involved so if you have to make a decision you have to have time to switch the task!!!! TASK SWITCHING!!!! THE TIME TO SWITCH THE TASK CANNOT BE AVOIDED
The central executive bottleneck can’t be eliminated.
- This means that a person cannot text and drive at the same time. They are switching between tasks.
- Hands-free or automated devices don’t reduce cognitive load, so are never automatized. (Even automatized tasks don’t eliminate the PRP.)
- People consistently overestimate their attentional capacities.
• Traffic fatalities are the overwhelming #1 cause of
preventable death for college-aged people.
People consistently overestimate
their attentional capacities.
Hands-free or automated devices
don’t reduce cognitive load, so are never automatized. (Even automatized tasks don’t eliminate the PRP.)
#1 cause of preventable death for college-aged people.
Traffic fatalities
Training attention?
It was noted that some things can make tasks easier (faster).
- Practice can make tasks with consistent mapping go faster. This may be because you are lowering the attentional load of the task.
- When tasks have varied mapping, practice may increase your capacity sharing.
• If the Psychological Refractory Period is a hard limit (and it seems to be), how can we train attention to be faster or have greater capacity?
Practice can make tasks with _____ consistent mapping
go faster
This may be because you are lowering the attentional load of the task.
When tasks have varied mapping, practice may
increase your capacity sharing.
not reducing cognitive load!!
switching between tasks and getting more efficient
Recall statistical learning
(through mere exposure)
allows you to detect contingencies between stimuli (sorta non-associative but you’re still making unconscious associations).
- perceptual learning
• If a stimulus predicts another stimulus, this should reduce the attentional task load associated with processing them together
(because it reduces computational complexity or cognitive load). [pretty baby easier than preby batty].
- your sensory systems have learned to guess what follows what
Perceptual learning (i.e. practice) should, therefore:
- Lead to faster reaction times in processing and identifying stimuli when mapping is consistent (or when probabilities are higher).
- Allow faster capacity sharing (switching) when mapping is varied.
- better at knowing where the pauses are going to be
The attentional blink
is a temporary decrease in the ability to detect a 2nd target soon after detecting a 1st target.
–> you’re going to be worse at detecting that second letter
how attentional blink is tested
- Letters are displayed using rapid serial visual presentation (RSVP), every 100 ms.
- Task 1: Respond to the S
- Task 2: Respond to the B
- The target letters can vary by lag – the distance in letters between them.
When the second target (T2)
follows the first target (T1) by 180 – 500 ms, participants have trouble reporting it.
- There seems to be a difficulty in processing T2 if it is delayed.
- This may be due to a capacity limitation in sensory processing.
- Additionally, there is the switch cost to identify the T2 after selecting the response to T1.
- The exact mechanism isn’t known, but the effect is vey robust.
Can the attentional blink be altered with extensive training?
Green & Bavelier (2003)
compared visio-spatial experts to novices on visual attention tasks.
• Action video game players (VGPs) were compared to nonVGPs.
• The thought was that game
expertise may train and improve attention in capacity - reducing bottleneck (making rapid identifications easier - reducing switch cost) and task switching.
Results: found
that the VGPs experienced less attentional blink than nonVGPs
open question: capacity or task switching?
CRAZY THING about attentional blink testing
you have the first target appear you become much worse not immediately after the first!
it’s not a problem right away (not sensory)
why? no real answer
Was the difference only for items in the center of vision where you focus on the game?
Green & Bavelier (2003)
• No. They found improvements for a peripheral target detection task at eccentricities outside a game environment.
• This suggests that the skills
learned in gaming generalize
outside the original viewing area.
• Generalization is a rare finding (e.g. chess skill doesn’t tend to generalize), so this is an intriguing finding.
