Lect4 Attention Part 1 PDF

Attention (Part 1) In this lecture, we will discuss: Selective attention How do we focus on one thing?...

Attention (Part 1) In this lecture, we will discuss: Selective attention How do we focus on one thing? Three approaches Broadbent’s theory Treisman’s theory Late-selection model Divided attention KMF 1023 COGNITIVE PSYCHOLOGY How do we focus on two or more things at the Lecturer: Norehan Zulkiply same time? FSKPM Three factors to consider Universiti Malaysia Sarawak Attention Attention is involved in…. Process of concentrating on specific features Perception of the environment or specific thoughts or auditory, visual activities Memory Selective: excluding of other features of the Language environment Limited: in capacity and timing Problem solving Both overt and covert: we can consciously attend to information but some information grabs our attention What are some examples to show that you are paying attention? Selective attention Selective Attention Selective attention- ability to focus on Experiment by Cherry (1953) – Dichotic only one thing and ignore others Listening Task One message is presented to the left ear and Humans find it difficult to focus on another to the right ear information from two messages that are Participant “shadows” one message to ensure presented at the same time he is attending to that message Shadowing: ‘attend to’ and repeat the attended message out loud Example from experiments Didn’t remember much from the ‘unattended’ message - even when repeated MANY TIMES 1  Results of Dichotic Listening Participants could not report the content of the message in unattended ear Knew that there was a message Knew the gender of the speaker However unattended ear is being processed at some level Change in gender is noticed Change to a tone is noticed Cocktail party effect Ability to pay attention to one message and ignore all other messages at a noisy party when you are focusing In the shadowing procedure, a person repeats out loud on what someone is saying to you words he has just heard Theories of Selective Attention 1) Broadbent’s Filter Model of Attention Selective Attention as filtering Early-selection model Broadbent’s Filter Model of Attention Filters message before incoming information is (early-selection model) analyzed for meaning Treisman’s Attenuation Theory of Attention Early-selection model based on information (early and intermediate-selection model) processing Late-selection model (e.g., McKay, 1973) Theory of auditory attention Broadbent’s Filter Model of Attention Broadbent’s Filter Model of Attention Sensory Store Filter (penapis) Holds information for a short amount of Identifies attended message using physical time characteristics of the sound Transfers information to the filter voice, pitch, speed, accent All other messages are removed (filtered) except the attended message 2  Broadbent’s Filter Model of Attention Broadbent’s Filter Model of Attention Detector Short term memory Where unfiltered (attended) messages is Receives output of Detector processed to identify meaning Used immediately, or moved to long term Processes ALL information sent to it (to memory, or lost... determine higher-level characteristics of the message) Why is it called an Early-Selection Model? Channels of Attention Attended message can be separated from the Cherry’s dichotic listening task and unattended message early in the information processing system Broadbent’s theory assumes our left Filtered before meaning is : and right ears are different channels Like a sieve (penapis) Broadbent’s conclusion: it is difficult to Filters based on pitch, rate rather than size of particle switch attention between channels Problem with Broadbent’s theory Problem with Broadbent’s theory Later research showed information Moray conducted an experiment to show: (awareness of the meaning) is Information presented to the unattended ear is processed enough to give the listener some “detected” before being “filtered” awareness of its meaning Participants could switch channels and become Evidence against the filter model aware of the meaning of the unattended began to increase… information Eg. You may be talking, and laughing with some friends, then, above your conversation and all of the background noise, you hear your name 3  2. Treisman’s Attenuation Theory of The importance of Broadbent’s Attention theory Analyzed human thought processes in Theory of auditory attention terms of information being processed 2-stage theory that replaced filter with an attenuator Early (or intermediate) selection model through a sequence of stages Attended message can be separated from unattended Stimulated more research on message early in the information-processing system Selection can also occur later (i.e., at dictionary unit) ATTENTION Treisman’s Attenuation Theory of Treisman’s Attenuation Theory of Attention Attention Dictionary Unit Attenuator - Analyse incoming message in terms of: Contains stored words that have thresholds for being Physical characteristics (high or low pitched; fast or slow) activated (tahap pengaktifan) Language (grouped into syllables or words) Threshold- the smallest signal strength that can just be Meaning (meaningful phrases) detected Just enough processing to distinguish different incoming messages Common words/especially important – low threshold Attended message emerges – full strength but, Uncommon words/unimportant to the listener –high unattended message emerges – attenuated (Leaky filter thresholds (need stronger signal to activate the words) model) Treisman would say your name (if you are the listener) has a low threshold Treisman’s Attenuation Theory of Attention 3) Late-Selection Model of Attention The final output of this model: Selection of stimuli for final processing does not occur until after information has been analyzed for meaning The STRONG Attended message that gets through MacKay (1973): In attending ear, participants heard an ambiguous sentence + “They were throwing stones at the bank.” In unattended ear, participants heard a biasing word either “river” “money” The IMPORTANT parts of the weaker After many trials, participants were asked which of these two unattended message target sentences was closest in meaning to one they heard earlier. They threw stones towards the side of the river They threw stones at savings and loan association yesterday 4  Comparing Early & Late-Selection Models Late-Selection Model of Attention MacKay’s experiment found that” The meaning of the biasing word affected participants’ choice Even though participants reported not hearing the biasing words, they were more likely to pick target sentences that fit the biasing word This result suggests unattended information was processed to the level of its meaning (Broadbent) (MacKay) How to decide when to use Early or Late- Task Load Selection Models? There are evidences for both theories High-load task: difficult and requires lots of cognitive resources Some researchers suggest attention is Eg. When you want to solve a mathematical or programming problem, a tough problem task dependant (bergantung kepada Make use of early selection model of attention tugas) Require most of your cognitive resources, process only attended task The important factor is task load: the Low-load task: relatively easy and requires less amount of cognitive resources needed cognitive resources to complete a task Eg. Listening to music Make use of late selection model of attention Require less cognitive resources, so both attended and some unattended tasks can be processed Divided Attention Divided attention How do we focus on two or more things at Divided Attention the same time? When is this possible and when is it not? 5 Divided Attention a. Skill- Effect of Practise Paying attention to two different things simultaneously Results of research on Divided Attention: Simultaneously drive and talk With practice, people can learn to do 2 quite (not very good to drive and talk on your hand difficult tasks simultaneously phones) Possibility to divide attention depends on 3 Experiments by: factors: Schneider & Shiffrin (1977) - detecting ‘targets’ in a. Skill rapidly presented frames b. task difficulty, and, (memilih target dari rangka-rangka yang ditunjukkan dalam c. task type jangkamasa yang sangat pendek) Skill- Effect of Practise Skill- Effect of Practise Results: Schneider & Shiffrin (1977)’s experiment 1: consistent mapping condition First shown a memory set consists of letters or numbers Shown 20 frames (2.4 seconds) Each frame had four positions Dot pattern or a target or a distractor is placed at any one of Participants only had 55% correct answers at first, but… the four positions They reported the task became automatic after Distractors from different category than target about 600 trials (In the first 600 trials, they had to keep Same target is never used as a repeating the target items in order to remember them) distractor on another trial and occurs without intention/ consciously thinking about it same distractor never used as uses few cognitive resources targets => Automatic Processing happens after practicing 600 times => Divided attention became possible after practice b. Effect of Task difficulty Task difficulty If a task is difficult, divided attention is Hard/Difficult tasks may not eliminate sometimes not possible divided attention but… Evidence from experiments: Can negatively affect performance on Strayer & Johnston (2001) – Cell phone one or more of the performed tasks conversations and driving For example... 6 Task difficulty c. Effect of Task Type (jenis tugas) Study by Strayer & Johnston (2001) on driving while talking on handphone: Experiment by Brooks (1968) Results: Hands-free phones provide no safety 2 conditions: advantage First condition: You run pass more red lights and take longer to “John ran to the store to buy some oranges” hit the brakes (slower reaction time) Memorize the sentence above and in your mind say ‘yes’ or ‘no’ to whether a word in the sentence is a noun Second condition: Try again by pointing to the Y or N, to whether a word in the sentence is a noun “The bird flew out the window to the tree” Task Type Summary Results: Selective attention – focus on one thing & People found the first condition harder ignore the others compared to the second condition Early Selection Model - Broadbent’s theory : filters sound it harder to respond when they had to say ‘yes’ or ‘no’ based on physical characteristics compared to pointing to Y or N Early/Intermediate Model - Treisman’s theory: attenuator filters First condition: Verbal response matches the task sound based on meaning, language & physical characteristics or remembering the sentence Late Selection model – McKay’s model: filters sound based on meaning Second condition: The pointing response (spatial response) does not match the verbal task => Responding is harder when similar types of tasks compete for cognitive resources => Highlights competition between similar tasks Summary Attention (part 2) Divided attention – focusing on 2 stimulus simultaneously Ability to divide our attention depends on 3 factors: a. Skill KMF 1023 b. task difficulty, and, COGNITIVE PSYCHOLOGY Lecturer: Norehan Zulkiply c. task type FSKPM Universiti Malaysia Sarawak 7  In this lecture, we will discuss: What do people ‘see’? Eg. Visual attention how do we attend to objects that we see? What do people “see”? The spotlight of visual attention (2 models) Object-based visual attention Evidence from brain damaged patients What do people ‘see’? A scene’s gist even if the scene is flashed for a fraction of a second Even information not focused on Fei Fei Li (2002)’s study What do people ‘see’? Change Blindness Fei Fei Li (2002) Change Blindness -> difficulty in suggests we receive a majority of the detecting changes in scenes especially information from our environment even at smaller details very fast times Caused by failure to pay attention but this is not always true (especially when Eg. Compare the two scenes and try to see we consider smaller details) what is different. Change blindness phenomena 8 Stimulus 1 for change-blindness demonstration. Stimulus 2 for change-blindness demonstration. Change Blindness But if you add a cue to show which part of the scene had been changed, you can detect the changes much more quickly Shows the importance of attention = cue What do people ‘see’? Models for Visual Attention Three models for visual attention: Changes are usually difficult to notice but we Spotlight function quite well without being aware of Zoom Lens every detail in our environment (i.e., with our perceptual intelligent we can “fill in the *(describe attention as operating on whatever stimuli are at a particular location) blanks” of the scene based on past experiences)) Object based *(suggest that enhancing effect of attention Functional NOT to attend to every detail or can be located on a particular object rather else our perceptual system can be overloaded than at a particular location) very quickly Intelligent perception (Chapter 3) Role of top down processing 9  1. Spotlight Model of Visual Attention 1. Spotlight Model of Visual Attention Model of location-based Spotlight functions like a precue attention Precue: stimulus used to cue Visual attention is like a spotlight location of a target Information within “spotlight” processed “deeper” than outside the Improved information spotlight processing for stimulus at precued location (where attention is given) 2. Zoom Lens Model of Attention 3. Object-based Attention Modification to spotlight model of visual Attention given to one place on an attention object (object-based), and not where Attention can be spread (zoom out) or the object is located (location based) narrowed (zoom in) -that is, the spotlight is not a fixed width The effect of attention can spread Processing is best when the spotlight is as throughout the whole object narrow as possible (zoomed in) Also location-based Egly’s (1994) experiment Object-based Attention Object-based Attention Reaction time was fastest at A< B < C B & C are equal distances but people respond faster when target was seen at B compared to when it was seen at C Paying attention to A had caused the attention to spread throughout the same rectangle on the right Even if the cue was at A some enhancement happened at B Egly et al.’s (1994) object-based attention experiment. as well (a) Cue signal appears at the top or bottom of one of the rectangles to indicate where the target will probably appear; (b) target appears at one of the ends of the rectangle. Numbers indicate reaction times in msecs for when the cue appeared at the top of the right rectangle. 10  Object-based Attention: Evidence from Object-based Attention: Evidence Neuropsychological studies from Brain damaged patients Condition known as Unilateral Neglect which affects one’s attentional mechanism Unilateral = one sided result from brain damage to right parietal lobe cause people to ignore information on the left visual field Half the world does not exist! Neglect indicates a lack of attention to the left side of the world occurs for both locations AND objects Patients who suffer from unilateral neglect acts as if For example: only half of the object is there. Even though they can see the “absent side” Summary Summary of Models: Attention can be based on where a person is looking in the environment (location based) & Spotlight: location where a person is looking on an object based (object based) Zoom lens: modified Attention works like a spotlight that scans spotlight, location based different locations in general scenes, and can lock onto objects and follow them when they move Object-based: sticks to object Summary Damage to the right parietal lobe can result in Unilateral Neglect Unilateral Neglect – a condition where people fail to pay attention & ignore information in their left visual field 11