Class Notes (807,339)
Canada (492,708)
Psychology (3,452)
PSY270H1 (191)
David Chan (64)
Lecture 8

PSY270H1 Lecture 8: Attention + Action

11 Pages
Unlock Document

University of Toronto St. George
David Chan

Divided Attention + Failures of Attention This is Feature Integration Theory (by Treisman). Imagine you look at a water bottle. The first stage (pre- attentive stage) you break up the object into its individual features (colour, shape, orientation, size). All of these features are “free-floating” – they are floating around in a theoretical mental space in your mind. It’s pre-attentive because it doesn’t require any cognitive resources (it happens automatically whether you’re engaged in another task or not). The focused attention stage is where you focus on the specific features and combine them to create a holistic representation and perceive it (which is conscious). This all happens unconsciously. He did a demo where he flashed a picture of four shapes along with two numbers on the left and right side of the screen. The task is to remember the features of all four shapes as well as find the sum of the two numbers (he divided attention between the two tasks). What happens is that when people made errors they tended to make errors based on swapping features (ex. Red triangle and Green circle --- “I saw a red circle”). Since all the features are free floating, if you limit your ability to attend these features your brain will randomly combine the features (called illusory conjunction). To do this you can either divide your attention (limit the amount of cognitive resources allocated to the task) or you can limit the amount of time in the flash (ex. you can do this really well if the shapes stayed for 2 seconds). Can we make a really difficult task (something that uses a lot of cognitive resources) automatic? In consistent, target is always a number and distractors are always letters (easy). In variable: Target was always a letter and distractors were always letters (which makes it difficult). In the second trial, a previous distractor became a new target and targets in previous trials became distractors (which makes it even more difficult). Can we practice the variable condition so much that we become just as good as the consistent condition? The results say no. At around trial 600 you plateau in both conditions. They concluded that there are some tasks that are so cognitively demanding that they never become fully automatic (they’ll always require some form of controlled processing aka cognitive effort). Another example of this is driving: No matter how long you’ve driven you will always require some level of cognitive effort. Side note: Highway blindness – The idea in which people will miss salient events in their visual field while driving as a result of being disengaged in the task. E: P’s come into lab and use a driving simulator. Have P’s drive on a course and see how many red lights they miss while either talking on a cell phone or not. Results are obvious: People with cell phone missed double the amount of red lights. When talking on the phone you are pulling attentional resources away from driving in order to talk on the phone (even if you’re an experienced driver). One possible confound: Maybe it’s not that you’re pulling attention away from driving but that you’re driving becomes worse as a result of only using one hand (since you’re using the phone with your other). They re-did this study using Bluetooth. The results showed that Bluetooth made you perform just as poorly as a phone (which shows that talking pulls attentional processes away – they even did a separate experiment that showed that experienced drivers are just as good as driving one-hand as two-handed). They did another study to test what the effect is of having a conversation with a friend who is physically with you. Results show that it is not as bad as using a cellphone (but still worse than no cell phone). The reason is that your friend is sharing the same visual experience as you and so naturally the conversation dampens (the amount of words said goes down). People understand that the driver is using a cognitively demanding task and so we automatically scale back on our ability to draw their attention (listening to music is even less since that is a passive task). Inattentional blindness – When we are focused on a task that demands our attention, we often fail to notice unexpected events (the video of passing a basketball but fail to notice gorilla). Change blindness – The failure to notice large changes between scenes that are otherwise easily detectable (A real life example: Movies filmed over multiple days and so in multiple takes there might be slight differences in ex. their hair). One way to test this is using the flicker paradigm (make two images, one version has slight difference – flick them back and forth and test how long it takes for P to detect the difference). This shows that we don’t have a rich representation of our environment (which is why this task takes long). In fact if you take experts who do have a rich representation of specific contexts, they will be good at those flicker paradigm versions (ex. Pro chess player will notice difference on a chess board compared to novice). These two types of blindness show the limits of our attention – We often miss salient events when we are focused elsewhere (even if we’re looking for them). When we attend to a stimulus it comes at the cost of attending other things. Orienting Attention Selective attention is the focusing of attention on a specific stimulus (at the cost of something else). Orienting is the alignment of attentional systems with the source of information. In addition to overt and covert there are two other ways in which we can shift attention. Exogenous cueing – When you have something in the periphery that captures attention automatically (ex. a flash of light). Endogenous cueing – Cues that direct attention based on meaning (ex. show arrow that points to the right – the arrow has a meaning because we’ve been taught its meaning). Alfred Yarbus wanted to find out if eye movements are informed by social context. Have P’s come into the lab and look at a painting. They were told to look at the painting depending on the instructions (ex. Is this family in the picture rich or poor…Do you think the visitor in this painting is happy to be there – essentially giving them a social context). They then tracked their eye movements. Ex. If they’re asked to figure out how old a person is the eye movements seem to focus on the faces of the people. Therefore eye movements reflect social understanding (what you know about the social environment). Taking this one step further: Autism is a developmental disorder characterized by withdrawal of contract from people & differences in attention (they don’t understand social cues/context). If eye movements tell us about social understanding, can we use this to diagnose children with autism (if so, this would be better than our traditional ways of currently diagnosing by studying their language development – which sucks because we can’t study that in a 2 or 3 year old child but we can study eye movement). E: Show kids movie scenes and track where their eyes move. Normal children focus more on the eyes (they tell us a lot socially) as well as what the person is looking at (because they are trying to grab a social understanding). Autistic children focus all over faces and they also focus on the background (things that are not socially informative). Endogenous orienting Endogenous cues generally exist in center fixation (so periphery things are not endogenous cues) and endogenous cues must be predictive and meaningful. If you do this task and realise the arrow doesn’t give you a correct meaning, you are no longer making it an endogenous cue. The finding is that you are faster at responding in the valid trials. Because endogenous cues are based solely on meaning, we can use almost anything and make it an endogenous cue (ex. an airplane or a fish orienting to the left). Exogenous orienting Exogenous cues generally exist in the periphery and are non-predictive. If a flash of light starts to become predictive then what you’re doing is giving that cue meaning and thus not exogenous. E is on the left. You are told to keep your eyes fixated on the middle box. One of the boxes on the right or left will become bright and you are told to ignore it (but nonetheless it’s meant to capture your attention). The cue goes away and the target appears on the left or the right. Valid = cue and target on the same side, Invalid = cue and target on opposite sides. Results are on the left. CTOA was varied and it is the time difference between the cue and the target. In the first grey section, you are faster in the valid condition. But this is only true if CTOA is 200ms or less. But after 200ms, this reverses. This slower effect in the valid condition is called inhibition of return – The effect in which people are slower to respond to a recently attended location. Ex: Wife loses her keys, tell her to look in her purse and it’s not there. Tell her to check again and when she checks it again she will pay less attention because she already knows it’s not there – i.e. there’s resistance to check – because of this resistance it takes you more cognitive resources to search the second time. This example is similar to what happens in this task: When the cue shows up your attention shifts to that side. If the target is taking a long time to show up, your attention will shift back to the middle and I will ignore that area. When the target comes it will take more cognitive resources to re-check that area because there is resistance to check that area again because you are under the impression that it’s not there (and so it’s easier to move to a new location hence the faster invalid). This only happens in exogenous cues. Cost of Action (Can the cost of doing something actually physically change our perceptual experience? – see a different colour then what’s really there) Proffitt wanted to look at how people judge the slant of hills. P’s come into the lab and look at a bunch of hills and indicate the angle in one of three ways. Verbal: Say it out loud how many degrees. Visual: A cardboard circle—move label Haptic: A board that can tilt—have P put their arm on it and tilt it to the angle they think the hill is. The results show that people tend to overestimate the angle of the hill when they are told to respond either verbally or visually. In the haptic condition they are closer to the true angle of the hill. Proffitt took this further and asked can we change how people judge these hills based on the cost (the cost of climbing a hill). They argued that if you can make something seem more costly, you can influence how you perceive it. So they had P’s come in and judge hills. Same experiment as before but add in an additional group who judged the hills while they were wearing a backpack. The left graph: No back condition shows same results as last experiment. Once you put on the backpack, you judge the hills to be even steeper but only for the verbal and visual but not haptic. They argue that when you look at the hill and the cost of action is greater you over exaggerate because now theoretically it takes you more energy to climb the hill. Haptically you resist this over exaggeration because we have this unconscious desire to accurately interact with the world when using our sense of touch. Unlike slants, we tend to underestimate distances because we are all capable to walking long distances (it’s easy for us). Then theoretically, if cost of action influences how we perceive the world, we should also be able to manipulate the cost to change our perception of distance. This E used older (55+) and younger adults (the rest). They argued that when you’re older, walking distances become harder and so this should affect perception. Bring P’s into a hallway, put a target far away and ask to estimate distance. Results: Older adults estimate the distance as far away compared to younger adults (older adults require more energy). Another way to affect cost of action is with pain. Have P’s judge distance. Patients were people in the hospital. Same results as last experiment. Now what happens if there is no cost, but some type of advantage (based on enjoyment or ability), would this show the opposite effect? Ability-based scaling – If perception is influenced by physiological potential, then ability (in addition to cost of action) should scale the environment. Taylor looked at people called traceurs (people who do parkour). So if they have this ability they should perceive the height of walls different than people who don’t do parkour. Give P’s three walls and asked to estimate height. They also asked P’s to say how confident they are in being able to climb the wall. The perception and confidence in the first two walls are not significant (so their perceptions are the same). Once you make the walls harder to climb for the novices but still easy for the traceurs, the traceurs perceive it as shorter than the novices. If you were to give them a fourth wall and it was some ridiculous height, the perceptions for both groups would be the same (because both of them are just as bad at climbing those ridiculously high walls). Therefore
More Less

Related notes for PSY270H1

Log In


Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.