Three-component model (Atkinson and Shiffrin)
- Sensory register
If attended to what you’re sensing will go to the short-term store / short-term/working memory
- Short-term store
If you don´t rehearse/reactivate information in the working memory it will fade away.
- Long-term store
The Spelling experiment
Sensory registers
Brief pre-attentive representation of input from sensory modalities after the orginal input has ceased
Iconic memory
representastions of visul stimuli (duration 200-500 ms)
Echoic memory
representation in auditory system (duration: 3-4 seconds)
Haptic memory
representation in somatosensory system (duration: about 2 seconds)
Pre-attentive
It is there regardless if we attend to it.
Magical number 7 (George A. Miller)
- The short-time memory is able to store 7+-2 bits of information at a time
- Chunking: organising smaller bits of information into larger units (“chunks”) of meaning, you can remember 7 chuncks
- The short-term memory system has a limited capacity
Experiment: Word list task
Primacy effect:
if one reads 15 words and repeat the words it´s easier to remember the first word, (you have capacity)
- Recency: it´s also easier to repeat the most recent word (short-term memory)
- After doing the experiment ten times the participants were given five minutes to write down as many words as she could from all ten lists, she got a negative recency effect where she remembered the last word the least
- The loss of the recency effect for delayed recall indicates the exixstence of a short-term memory system which decays over time
Working memory
Not just a short-term memory storage, but rather a mental workspace which serves goal-oriented cognition
Includes both a storage component (containing the mental representation) as well as a processing component which transforms the stored/active representation (mental operations) -> Baddeley model
Baddeley´s multicomponent working memory model
- Visuo-spatial sketchpad: buffer in short-term memory that deals with visual/spatial information
- Phonological loop:
- deals with verbal auditory information
- Phonological buffer: holds speech-based info for short period of time 2-3 seconds
- Articulatory control process, allows for the maintenance of info in the phonological buffer (sub-voal rehearsal) and conversion of visual information (written word) to speech-based phonological form
- Episodic buffer:
- A global workspace that is accessed by conscious awareness
- Storage of the central executive that binds together information from the phonological loop, visuo-spatial sketchpad and long term memory into chunks/episodes
- Limited capacity (4 chunks/episodes)
- Controlled by the central executive
- Comparable to Conan’s Working memory model
- Central executive: tries to utilize the info of the slave systems
- Contention scheduling system: automatiske kontrollsystem som gjør det mulig for oss å gjøre rutiner/andre innlærte oppgaver som for eksempel å sykle
- Supervisory activating system (SAS): kicker inn når man gjør noe komplisert, her må man bruke SAS til å override den automatiske reaksjonen slik at man kan rette oppmerksomheten mot et mål
Sub-vocal rehearsal
- Rehearsel, i.e. Sub-vocal (inner) articulation, revives the memory trace, and the time it takes to rehearse determines memory span
- Note: this contradicts the idea that we have a maximum number of items we can keep up in working memory (re. G Miller´s magical number 7)
Word length effect
- An experiment where words with the same syllables but with different spoken durations (coerce VS bishop) are tested.
- Result: it´s not about the units, it´s about the length
The spoken duration not the length in syllables determines WM span
Evidence for the role of sub-vocal rehearsal
- Word length effect
Immediate recall of word list is better for a more quickly pronounceable words - interpretation: sub-vocal rehearsal Is faster - The effects of articulatory suppression
Overt articulation of distractor words while being engaged with the primary word list affects performance negatively - interpretation: overt articulation interferes with sub-vocal articulatory rehearsal - The irrelevant speech effect
Presenting irrelevant speech in the background during learning of target word list affects performance negatively - interpretation: interference of sub-vocal articulatory rehearsal by phonological input to the phonological buffer - The phonological similarity effect
Recall is poorer for list of verbal items when the items sound alike, relative to performance when items do not sound alike (and having comparable meaning) - interpretation: shared phonological “fragments” interfere with each other
This is evidence that the phonological loop is articulation based.
Conan´s WM model
- There’s no separate working memory, believes that the working memory is an activation of the long term memory
Visual cache
Passive store of visual or spatial information for a short period
Inner scribe
Allows maintenance of visual information in visual cache in a rehearsal-like process
Experiment: Irrelevant pictures effect in visual working memory
Participants were either doing a visual task or a “blinded” pegboard (spatial test) before the retrieval of the visual patterns test. The visual interference task had a stronger negative effect on visual working memory than the spatial interference task.
Dysexecutive syndrome
- Associated with dorsolateral prefrontal damage.
- If one only has contention scheduling system, can only do well-practiced actions.
- Cannot learn to master new types of task.
Boundary extension errors
Participants recall info that extends beyond the shown scene. Widens the frame for the picture to make more sense, one creates a context that’s wider than what has actually been shown.
