STM and working memory

Short-Term Memory (STM)

temp store of current sensory info and relevant LTM to achieve current goals and make sense of surroundings

• brief duration - longer than sensory

• capacity = 7 ± 2

Maintenance Rehearsal

The process of keeping information active in STM through an ‘inner voice’ until transferred to LTM

Digit Span Task

Used to estimate capacity of verbal STM - hear number sequence and repeat it back in order = immediate serial recall. capacity estimated by longest sequence recalled correctly in 2 trials

Brown-Peterson Task

measures decay of STM - hear 3 letters and then a number. the number cues participants to start counting aloud backwards by 3s from that number until told to stop. Filling retention interval with task prevents rehearsal = 3 sec interval 50% of trials were recalled correctly and by 9 dropped to 20% and 0 by 12-18 seconds

Primacy Effect

easily recall first items in a list because opportunity to rehearse increases likelihood of being transferred to LTM = eliminated if rehearsal if prevented by introducing concurrent task (e.g. repetition of word)

Recency Effect

easily recall last items in a list because still contained in STM = reduced by introducing filled retention interval before recall

Craik and Tulving 1975

level of processing applied to material to be recalled is more important for determining if will be encoded in LTM than amount of rehearsal time

Working Memory

Baddeley and Hitch (1974) proposed multi-component system that supports meaningful encoding and reasoning to replace MSM. MSM focuses on maintaining info for immediate recall but working sees STM as mental workspace that helps achieve current goals and update understanding of world

Episodic Buffer (EB)

A storage system that combines inputs from loop and sketchpad and LTM to create multimodal representation of the current contents of awareness.

Phonological loop

temporary store for sound-based information

• capacity tested using digit-span backwards which actively manipulates info in memory rather than just maintain sequence

Visuo-spatial sketchpad

temporary store for visual and spatial based info

• capacity measured using corsi block tapping task = 9 blocks that examiner taps in pattern that participant repeats = length of sequence accurately recalled in 2 trials is capacity

Central Executive

set of central cognitive processes that control attention to info and selection of info from LTM for planning and coordinating complex behavior, e.g. goal orientation, focus attention, multitasking.

• measured using Stroop colour-word task = say colour of word and not read word = direct attention to target while suppressing interfering info

serial position effect

The tendency to remember the first and last items in a list better than those in the middle - u shaped serial position curve

Craik and Tulving 1975 method and results

Didn’t know was memory test just measured on time of response. 1 of 3 questions asked before each word to promote a different level of processing

• structural/perceptual = visual-based(shallowest)

• phonemic = sound-based (slightly deeper processing)

• semantic = meaning-based (deepest level of encoding)

5 minutes to do another task then given surprise memory recognition test to find which words they were shown

Results = remembered more for semantic with least for structural = deeper level produces greater LT retention

components working memory

central executive, phonological loop, visuospatial sketchpad and episodic buffer

components of LTM and their connection to working systems

• visual semantics = sketchpad

• verbal semantics = loop

• episodic LTM = buffer

2 way flow between working and LTM = info in working can be interpreted and informed by info from LTM and enables info in working to update LTM

neural basis of working memory

• executive processes = pre-frontal cortex including dorsolateral prefrontal and anterior cingulate cortex

• loop = left-hemisphere fronto-temporal lobe

• sketchpad = right occipital-parietal network

• buffer = parietal cortex

chunking

a memory strategy that involves grouping information into larger, more manageable units to enhance working memory capacity.