Please enable JavaScript.
Coggle requires JavaScript to display documents.
Working memory reading p2 - Coggle Diagram
Working memory reading p2
How does WM interact with other functions- WM and attention- Broadbent (1958) suggested what we attend to determines what we temporarily maintain- explored by more (Cowan et al 2024)- partly bottom up driven by environment input eg stimulus that is more attention grabbing more likely to be retained- also top down under our control- can choose what stimulus of most value in that moment- flexibility- stay alert for environment changes but prioritise at same time
Prioritisation techniques explore this- ppt said to focus on some items within set of stimuli - some items worth more points (Allen 2025)- relationship with attention= bidirectional- more likely to attend info in environment if it overlaps with what we’re holding in WM- or more likely to get disrupted if distractions overlap with what’s being held- indicates attention guided by content specific representations being held in WM
How WM interacts with others- WM and LTM- bidirectional relationship between WM and LTM- already seen how verbal WM can support vocab learning- can be stored in long term- LTM also affects what’s learned in short term - on next bit
environmental info processed/ manipulated in WM usually has some relation to existing semantic knowledge (general facts/ knowledge context free from experiences) which includes concepts and word meanings- helps learning Short term- WM better for words than non words- also better recall for concrete highly imaginable words eg fruit compared to abstract words eg trend, wise- (Allen and Hulme 2006) also better visual memory for real world objects than simple shapes (Brady and Stormer 2022)
More beneficial effects of LTM can be seen when going past individual item putting words into familiar sentence structures can improve recall of 15+ words in sequence compared to around 5 (Baddeley 2009) - demonstrates stored knowledge concerning word forms, meanings, syntactic structures can boost working memory for single items and item sequences- can happen with making richer / robust representations when encoding- reidintegration can help recall- degraded memory reps repaired with support of LT knowledge
WM and emotion- WM may serve as cog control centre for emotional processing eg better WM = better emotional regulation (Schmeichel et al 2008) - emotional stim also captures attentional resources more likely to be retained in WM- interfering with other cog processes (Chainay et al 2023) - so better WM- better value based focus and bidirectional emotions affect WM- emotional states as well as emotional stimuli can affect WM- eg negative or intrusive thoughts occupying attention and WM (Ellis and Ashbrook 1988)
Anxiety or depression impairs WM too (Baddeley 2013) - but therapeutic possibilities can be offered - WM involved in mental imagery and visual info/ imagery can influence the experience of negative intrusive thoughts (Andrade 2023)
Vividness and impact can be reduced by engaging with visuospatial tasks during or shortly after event encoding
Functional imagery training- vividly imagining pictures associated with desired goal can suppress negative images related to worry/ addiction cravings- so negative emotions can occupy WM- but can occupy instead with positive processing
Cog models- working memory- frameworks that try to understand how we encode, hold process manipulate and retrieve - 1 multicomponent frameworks- first described by Baddeley and Hitch 1974- replaces the outdated idea of a single, passive short-term memory box with an active workspace made of specialized, interacting parts
Parts
The Original 1974 Framework
The classic model consists of a central attentional controller and two modality-specific "slave" storage systems:
• Central Executive: The "boss" of the system. It manages attention, coordinates information flow, and switches between tasks, though it does not store information itself.
• Phonological Loop: Handles speech-based and auditory information. It is split into the phonological store (an "inner ear" that holds words briefly) and the articulatory process (an "inner voice" that uses subvocal rehearsal to stop memories from fading).
• Visuospatial Sketchpad: The "inner eye". It temporarily holds and manipulates visual patterns, shapes, and spatial layouts. (Note: Baddeley later added the Episodic Buffer in 2000 as a backup store to link working memory with long-term memory
Updated multicomponent model
Recent reviews and updates by Graham Hitch and colleagues re-evaluate the model after 50 years of cognitive research:
Refining Decay and Rehearsal: Early ideas of automatic time-based phonological decay have been challenged. Rehearsal is now viewed as more flexible and less rigid than first assumed. [1]
Automatic Feature Binding: Past theories argued that combining different types of features (like a shape and a color) required heavy central executive resources. Newer findings show that feature binding in working memory often happens largely automatically during encoding. [1]
Broadened Functions: The updated framework places more emphasis on how attentional refreshing works and how the system dynamically supports complex real-world tasks like language learning, mental math, and emotional regulation.
Time based resource sharing- Barrouillet and Camos (2014)- has multicomponent view for WM but puts emphasis on role of time decay- assumes memory traces fade over time unless attention sometimes redirected to refresh, - storage and processing share common attention resource- more time spent processing- less time maintaining memory representations
Supporting evidence comes from cog load paradigms- pots must retain info eg letters while doing concurrent processing tasks eg saying if numbers= even or odd- according to TBRS- memory performance= based on flexibility to switch attention between processing and storage- and on time availability to refreshing items by focussing attention
Attentional frameworks - embedded processes framework (Cowan et al 2024)- Unlike older "modular" theories (like Baddeley's model) that treat working memory as a collection of separate, physical storage containers or "slave buffers" (like a visual scratchpad or verbal loop), Cowan's model states that working memory is not a separate structural box in the brain. Instead, working memory is simply the temporarily activated portion of your long-term memory (LTM) combined with your current focus of attention.
Each layer embedded in the next
LTM- The massive, base repository of everything you know—including facts, vocabulary, past events, and learned skills. Most of this information remains dormant and inactive until it is needed.
Activated LTM-
When you perceive an environmental stimulus or think about a concept, a subset of your LTM undergoes a temporary state of heightened activation. [1, 2]
How it works: If someone says the word "apple," your LTM representations for "fruit," "red," and "pie" might automatically spark into an active state
This activation is time-limited and prone to decay unless it is kept alive by attention or refreshing. It acts as a passive background holding area.
Focus of Attention-Embedded directly inside that activated pool is the Focus of Attention, which represents what you are consciously aware of right now.- How it works: This is the most highly accessible state where actual information processing, manipulation, and new concept "binding" take place.
FoA limits- Limitation: The FoA is strictly capacity-limited. Cowan's research shows that a normal adult can hold only about 3 to 5 discrete chunks of information in their focus of attention at one time, regardless of whether that information is visual, verbal, or spatial.
Embedded processes how it works together- The Role of the Central Executive: A top-down control system (the central executive) helps guide your voluntary attention to select which items from the activated LTM move into the main focus of attention./
The Role of Chunking: Because the focus of attention can only hold roughly 4 units, prior knowledge changes your working memory capacity. An expert can group ("chunk") complex data into a single meaningful unit in their LTM, freeing up slots in their FoA, whereas a novice treats every single detail as an individual item, quickly overloading their focus.
New Learning & Binding: When multiple items are brought together in the FoA, they can "bind" into a new concept structure. As your attention shifts to something else, this new structure is offloaded back into the activated LTM to stay briefly before eventually consolidating into a permanent long-term memory.