Foundations of Perception and Cognition - Short-Term and Working Memory: Part 2 PDF

 This presentation contains images owned by third-parties. They can be used for educational purposes in this presentation. The following copyright warning must be provided prior to presenting them: Copyright Warning This material has been copied and co...

 This presentation contains images owned by third-parties. They can be used for educational purposes in this presentation. The following copyright warning must be provided prior to presenting them: Copyright Warning This material has been copied and communicated under the Statutory Licence pursuant to s113P of the Copyright Act 1968 for the educational purposes of the University of Adelaide. Any further copying or communication of this material by you may be the subject of copyright protection.  Part 1 (Lecture 1) Part 2 (Lecture 2) What is Memory? Working Memory: Overview An Early Model of Memory The Phonological Loop Short-Term Memory (pre-1970’s) What Use is The Phonological Loop? Short-Term and Working Memory The Visuo-Spatial Sketchpad? What Use is The Visuo-Spatial Sketchpad? The Central Executive and Episodic Buffer Working Memory: Evaluation  Recommended Reading This week’s recommended reading is: Goldstein Chapter 5: Short-Term and Working Memory I discuss most major topics* but don’t always cover them in the same order. I also cover interesting topics that are not in the chapter *Example: I leave out the content from the ‘Working Memory and the Brain’ section, but please do read this section A Quick Recap Atkinson and Shiffrin’s (1969) Modal Model of Memory saw STM as a unitary system for temporarily storing small amounts of information. Then, in the early 1970’s: 1. Brain lesion research showed Verbal STM and Visuo-Spatial STM are distinct systems 2. STM underpins many cognitive abilities (e.g., reading, maths, problem solving) Baddeley and Hitch (1974) there proposed a new model of STM called The Working Memory Model, that had several systems and emphasised STM’s role in underpinning cognitive tasks Defining Working Memory Baddeley and Hitch (1974) defined WM as: “a limited-capacity system for temporary storage and manipulation of information for complex tasks such as comprehension, learning, and reasoning” Their original model had three systems, but Baddeley (2000) added a fourth (see right) STM vs Working Memory Before discussing the Working Memory Model, let’s clarify how STM and WM differ. Some people use both terms interchangeably, but most distinguish between:  STM tasks: Participants briefly study / immediately recall information (e.g., a digit span task)  WM tasks: Participants briefly study / immediately recall information, but also complete an ongoing task at the same time and/or manipulate the information (e.g., recalling digits in a new order) Most studies I discuss in this lecture are technically STM studies The Working Memory Model Two of the four systems are the: 1. Phonological Loop (Verbal STM)  Temporarily stores language / sound-based information (e.g., words when reading or listening to a conversation) 2. Visuo-Spatial Sketchpad (Visuo-Spatial STM)  Temporarily stores images / spatial location information (e.g., colours or object locations) The Working Memory Model The Working Memory Model The other two systems are the: 3. Central Executive  An attention system (not memory) that helps us focus attention, switch attention between tasks, and divide attention 4. Episodic Buffer  Integrates / temporarily stores information from the PL, VSSP, and LTM (e.g., chunking) The Working Memory Model Working Memory and Everyday Abilities Working Memory supports “complex tasks such as comprehension, learning, and reasoning” Working memory ability is a more powerful predictor of success in school than IQ, and strongly predicts workplace success In class, I will show a 1 min 45 video by Alloway and Alloway (see right) discusses its importance Overview Temporarily stores language / sound-based information (so Verbal STM). It has two systems: 1. The Phonological Store: Speech enters the store directly/is temporarily stored 2. The Articulatory Control Process: Written words enter here, are converted, and are transferred to the Phonological Store What do I mean when I say written words are converted?  Overview The Phonological Store stores information in a phonological (sound-based) code or format Speech is in a phonological code (it is made of sound) and can enter the store directly Information in the store is quickly forgotten (e.g., in 20 secs) unless rehearsed in the Articulatory Control Process The Articulatory Control Process is also called the Articulatory Loop or Articulatory Rehearsal Process  Overview Written words are not in a phonological (sound- based) code, so cannot enter the store directly They enter the Articulatory Control Process, are converted to a phonological code using subvocal articulation* (rehearsal) and are then sent to the Phonological Store. Information in the store is quickly forgotten (e.g., in 20 secs) unless rehearsed in the Articulatory Control Process *This is the voice you hear in your head each time you read a word. Testing the Phonological Loop Baddeley (1990) made several claims about the phonological loop. What evidence is there that: 1. Claim 1: Written and spoken letters/words are stored in phonological codes 2. Claim 2: Visually and aurally presented letters/words are processed differently:  Speech is in a phonological code so enters the Phonological Store directly  Written words are not in a phonological code. They must be converted into one to enter the store Studies supporting both suggestions will now be discussed Claim 1: Phonological codes Letters/words that sound similar can get confused on STM tests (the phonological-similarity effect) Their visual appearance or meaning has far less impact (so their sound seems to really matter)  Conrad’s (1964) participants incorrectly recalled more non-studied letters that sounded like studied ones (e.g., D, when C was studied) than looked like them (e.g., E, when F was studied)  Baddeley’s (1966) participants made more errors recalling phonologically similar words (e.g., Cat, Mat, Bat) than semantically similar words (e.g., Huge, Large, Big) Claim 2: Different Processing Mechanisms Baddeley et al.’s (1975) participants studied lists of 8 words: ▪ Some contained 1 syllable words (short lists) Word length effect ▪ Some contained 8 syllable words (long lists) should occur  The lists were presented either aurally or visually Visual words need converting On 50% of trials, participants engaged in Articulatory Suppression to stop them (1) rehearsing the words and (2) converting visual words to phonological codes Claim 2: Different Processing Mechanisms Baddeley et al.’s auditory presentation results: ▪ A word-length effect occurred, irrespective of articulatory suppression (sloping lines) ▪ Articulatory suppression stopped rehearsal and recall was impaired slightly (black line) What happened during the visual presentation when words were not in a sound-based format and needed converting? Claim 2: Different Processing Mechanisms Baddeley et al.’s visual presentation results: ▪ With no articulatory suppression (green line), written words could be converted, recall was good, and the word-length effect occurred ▪ When articulatory suppression occurred (yellow line), written words could not be converted, regardless of length. Recall was poor and the word-length effect disappeared The Phonological Loop and Language Baddeley et al. (1998) argue the PL evolved to support language acquisition They felt it enables us to temporarily store and repeat unfamiliar sound patterns (i.e., new words) whilst permanent records are being constructed This is important for longer words. If you struggle holding them in Verbal STM, learning is harder The Phonological Loop and Language Gathercole and Baddeley (1990) tested: Language impaired  8-year-olds: language skills of a 6-year-old 8-year-olds  6 & 8-year-olds: age-typical language skills struggled with non- word repetition Language-impaired 8-year-olds were worst on a Verbal STM test and at repeating longer non- words (e.g., woogalamic), as they should could not hold them in mind to process / repeat them The Phonological Loop and Language Baddeley et al. (1988) studied Patient PV, who Healthy controls learned foreign developed Verbal STM problems after a stroke word-pairs easily after 10 attempts Her digit span was 2 - 3 items, but she had normal visuo-spatial STM, LTM, and IQ PV could not learn foreign word-pairs after PV and controls could easily learn native language 10 attempts word-pairs, but could not learn foreign language word-pairs (controls learnt them in 10 attempts) Overview 1970’s researchers mainly studied the PL. In the 1980’s, Bob Logie studied the VSSP in detail. In Logie (1995), he concluded it has two systems: 1. The Visual Cache: Stores visual information about object shapes, patterns, and colours 2. The Inner Scribe: Stores spatial information Bob Logie worked with Baddeley on (e.g., object locations and movement) many of his early VSSP studies Evidence for Two Systems Cognitive neuropsychological research on patients with brain lesions identified a double-dissociation between performance on Visual Cache tasks (e.g., immediately recalling the colours of studied objects) and Inner Scribe tasks (immediately recalling the spatial locations of studied objects): Patient and Study Lesion Cause Visual Cache Tasks Inner Scribe Tasks Patient L.H. Traffic accident Poor Normal Farah et al. (1988) Patient M.V. Stroke Normal Poor Carlesimo et al. (2001) The Visual Cache (Capacity) Alvarez and Cavanagh (2004) tested the visual cache’s capacity to see if is the same as Verbal STM’s. They used stimuli of varying complexity:  Squares (poor accuracy if 4.4 items)  Polygons (poor accuracy for 2 items)  Cubes (poor accuracy if 1.6 items) The more complex a shape, the fewer could be temporarily stored/recalled The Inner Scribe (Capacity) Vandierendonck et al. (2004) studied the inner scribe’s capacity using Corsi blocks:  A researcher taps out a sequence on blocks. The participant must repeat the sequence On average, participants could correctly recall the order of 6 taps (so had a spatial span of 6) A child repeating a Corsi Block sequence  Demonstration In class, time permitting, we will test your Inner Scribe’s spatial memory capacity 5 items 6 items 7 items The Inner Scribe (Capacity) Corsi blocks were invented by Corsi (1972) and based on Knox’s (1913) Cube Imitation Task  Knox’s task was used to screen immigrants to the US for intelligence problems (see right) Random question: Can you outperform a chimp on a spatial span task? In class, you will find out Everyday Uses of the Visuo-Spatial Sketchpad The VSSP has many uses (e.g., temporarily storing what objects look like and their location in space, before permanent records are created in LTM) It plays a role in many tasks, such as learning new routes when walking (Garden et al., 2001) On the next two slides, its role in visual search and visual imagery are considered  The VSSP and Visual Search Aziz et al. (2001), and others, have shown the VSSP plays a role in visual searches:  It holds information about target features & previously searched distractors and locations For young adults participants (18-35), as VSSP capacity increased, search times decreased* A target (shown pre-search) and search array *Older adults verbal STM capacity predicted search times. They had worse VSSP capacity (it declines faster than Verbal STM with age) and their searches were highly verbalised (mentally rehearsing things like “find the man with the monocle and green scarf”) The VSSP and Visual Imagery The VSSP plays a role in visual imagery, which is an important ability in design-focussed careers:  architecture, engineering, art, etc Wilson et al. (1999) studied an artist who developed Visual STM problems. Her sculpting changed from detailed realistic works to abstract forms as she struggled visualising objects  The Central Executive Baddeley (2007) argued the CE is an attention system (not memory) with three major functions: 1. Focussing attention on tasks* 2. Dividing attention between tasks 3. Switching attention between two tasks In terms of Working Memory, it coordinates the PL and VSSP’s limited resources (see right) The Working Memory Model *This also includes ignoring distractions The Central Executive You are watching sport on tv (focussed attention) and processing the images with your VSSP/the commentary with your PL The phone rings, so you stop focussing on the match and answer it (task switching), using your PL to process the conversation As you talk on the phone, you keep one eye on the sport (divided attention) The Episodic Buffer Baddeley (2007) argued the EB temporarily stores integrated information from PL, VSSP, and LTM  It allows allowing ‘chunking’ discussed earlier Baddeley (2012) felt it stores approx. four chunks of multidimensional information (i.e., visual, auditory, and other information combined) The Working Memory Model Integration Can Boost Recall Darling et al (2017) presented digit span digits: a) One at a time (verbal STM only) b) On a standard keypad display c) On a random keyboard display Recall was best with (B). They felt this occurred as verbal STM, spatial STM, and LTM (keyboard knowledge) combined in the EB, boosting memory  Two Strengths Two Limitations There is convincing empirical evidence for all Baddeley (2012) acknowledges the model is components of the model oversimplified, as several kinds of information are ignored (e.g., STM of smell, touch, and taste) As Logie (2015, p. 100) noted, the model explains We need more research on the interactions among the findings “from a very wide range of research topics, for working memory components (e.g., how the EB example, aspects of children’s language development, integrates information from the PL, VSSP, and LTM) aspects of counting and mental arithmetic, reasoning and problem solving, dividing and switching attention, navigating unfamiliar environments”. By the end of Short-Term and Working Memory: Part 2, you should be able to 1. Differentiate Short-Term Memory tasks and Working Memory tasks 2. Draw Baddeley and Hitch’s (1974) Working Memory Model, correctly labelling its components 3. Describe each of Working Memory model’s major systems and their functions 4. Explain Working Memory’s is importance in everyday life 5. Explain the strengths and limitations of the Working Memory Model