working memory
Extremely limited in capacity
‘Digit span’ = how many #’s can you successfully repeat back
William james - ongoing train of thought and theres things that go away but come back
H.M.
Intractable temporal life epilepsy
Treated with hippocampal resection and got anterograde amnesia
LTM severely impaired but STM retained intact
Normal short term memory for auditory material (7 +/- 2 items)
Worked with same researchers for 50 years but didn't remember them
Deficit highlighted by extended digit span task - at their limit then +1 to get more
He could not recall a single string 1 higher despite many repetitions
E.P
Viral encephalitis - lost ability to create new LTM but normal STM
K.F.
Damage in left parieto-occipital region of the brain because of a motorcycle accident
Had normal LTM but impaired STM
Recency effect - (the last numbers said) will lose of there is distraction in between recall
Primary effect - had undivided attention to repeat so it stuck
5 word list errors -
made when words sounded alike (house vs mouse)
due to verbal rehearsal → STM
10 world list errors
semantically similar words (labor vs work)
Due to memory based on meaning of words → LTM
George sperling - felt like they faded too fast but he knew the info was there before
Partial report - cued letters
Performance plateaus at about 4-5 letters
Can report at least 3 items per row so 3*3=9 out of 12 is stored
Sperlings idea = STM = sensory and working
A high capacity ‘sensory memory’ which quickly decays (iconic/visual memory)
A limited capacity working memory of only 3-4 items
Duration is less than 1 second for visual and auditory info
Vulnerable to visual distraction (masking)
Phonological loop (wm)
Seems to store sounds
Short words are easier (causes differences between languages)
Doing a speaking task while remembering (tah tah tah)
concurrent articulation task reduces wm capacity
reduces sound-alike errors
Similar sounding words causes interference independent of meaning
Counting backwards or other speaking tasks interferes with this system
Visuospatial sketchpad
Manipulate number of things to remember (set size)
Concurrent verbal task WM (load vs no load) did not have effect
verbal and visual memory use separate resources
Performance decline as set size increased from 4-12 items
Are visual and spatial working memory the same processes
Remembering visual but have spatial interference - bad and vice versa
They hurt each other and they not overlapping
George miller - ‘my problem is that I have been persecuted by an integer’ (+/– 7 things)
Not in information but 7 items/chunks
16 #’s is hard but 4 years is easy (1941200119842025 = 1941 2001 1985 2025)
S.F - sophomore w avg grades
230 one-hour training sessions for year and half
Digit-span task and add 1 each time he recalled correctly
Half of trials he would describe how he did it
Used mnemonic associations (3492 = 3 min 49.2 sec mile time)
But after tested with letters instead of words he remembered 6 - wm didn't change
7 average capacity
Result of unconscious chunking but actual number is closer to 4
Interim memory - a set of ‘buffers’ for working with info that is no longer perceptually present
Why do we care?
Correlates to life success, iq, reading comprehension, sat score
Reading span - read or listen to a list of 2-6 sentence and recall last word of sentence
Operation span - solve simple math and read the word aloud
N-back - previous stim the same as the last one
The more “working” the task the better the correlations
WM → LTM?? Modal model
Wm is the process of encoding into LTM
Would predict that things we interact w frequently are well represented in LTM
Modal model says stm required for entry into ltm but KF had no stm and did have ltm
Modal says length of time in stm determine likelihood of ltm storage (what about the penny tho)
Rehearsal
Maintenance rehearsal - reciting
Relational or elaborate rehearsal - linking
Repeated exposure does not help memory much - but associations and linking matter a lot
Learning
Incidental - learning w/out intention to (deeply/neural processing improves)
Intentional - deliberate with expectation memory will be tested later
Connections promoter retrieval - organization
Learning as prep for retrieval
Encoding = putting something into your ltm/memorizing
Memory is context based → when will u need this info again
Memory is state dependent (better when states are the same at learning and recall)
Physical match (diving, smell) and emotional match (happy, sad,)
Adaptive function = most relevant mems are most accessible
Context dependent memory = you don’t just ‘know’ or ‘not know’ it - they're in your mind but hard to access in wrong circumstances
Spreading activation & priming
Spreading activation - travels from one concept to another via associative links
Helps explain context effects and why partial information helps in retrieval
Encoding specificity
State-dependent learning an context reinstatement
“Lifted” - cues something heavy vs “Tuned” - cues musical instrument
This kind of activation spreading happens automatically - explains repetition priming
Nurse and doctor related but bread and chair are not
Explicit vs implicit memory:
amnesia (from medial temporal lobes) does NOT impact all types of LTM
Explicit test of memory - recall facts or events
Implicit tests - guess words based on word fragments - amnesiacs often show intact implicit memory
Implicit memory tends to last longer
Priming
Aftereffects
Muscle memory (bike)
Skills (mirror tracing)
Source memory - if available, recognition responses are similar in mechanism to recall “i saw this word before”
Familiarity - recognition based on feelings of familiarity
Fictitious names primed were remembered and assumed famous
Don’t know why names feel familiar
Source confusion - eye witnesses select from a photo lineup based on familiarity and not recollection
Amnesia - episodic vs semantic memory
Episodic = memory for personal events (mental time travel for humans)
First day of a new job
Wedding
Guests at best friends bday party
Semantic = facts and knowledge
Grass is green
Names of colors
How to use the phone
Amnesiacs
Damaged hippocampus - no episodic memory, cannot form new memories of events
Semantic memory intact & can remember general information about the past
Retrograde = cannot remember old memories (brief when u get a concussion)
Recent memories continue to undergo neurological change and more susceptible to amnesia
Anterograde = cannot form new episodic memories
2/13/25
LTM 2
Forgetting:
Ebbinghaus forgetting curve - quick loss at first then slow decay (100 58 44 33 25)
Autobiographical memory - longer time course of consolidation
considerable loss for three years then fairly stable memory
Consolidation - time it takes to imprint or keep in mind after period of time so its available
Why should we have bad memory?
Virtually limitless memory
Could not forget irrelevant details
Bad at inductive reasoning (filling in the blanks)
Anderson and schooler 1991- It is efficient for out memory system to make recent and frequent memories more accessible
A way to prioritize what is important and make it more accessible
Reminiscent bump = violation of forgetting law
We can rehearse memories and re-encode them which strengthens them and slows decay
Memories follow a predictable ‘forgetting function’ which shows evidence of consolidation
Rehearsal can make memories stronger leading to on avg a reminiscence bump in autobiographical memory and for some, perfect memory for the things they repeatedly rehearse
Explanations for forgetting
Decay - memories fade and disappear over time
Interference - memory is still there but we can’t retrieve it
In recall test, relearning same words showed much better results meaning those memories had been interfered and not decayed
Interference
Retroactive - new memories interfere with recall of old info
Remember new phone number and not old one
Proactive - old memories interfere with recall of new info
Remember old home address and forget new one
Wickens et al 1976 - interference from shared meanings of words
Ex: Leaning all the fruits interfered with recall while the group that had professions and fruits had better recall
Sleep - no interference
Alcohol - causes anterograde amnesia but also enhances memory for info encoded before drinking bc it prevents interference
Studying
Massed practice = many trials with the same stimulus are undertaken without interruption (better for ST retention but not always)
Spaced practice = the trials with the same stimulus are separated by other stimuli (better for LT retention )
Encoding variability - in subsequent encounters of a stimulus different aspects of a stimulus are focused on during encoding (and remember context is always encoded with an item)
Spacing increases encoding variability there are more ways in which information can be accessed and retrieved
Effort - if you already have an item active in your mind you may pay less attention to encoding a new trace (alternatively, effortful retrieval may itself be helpful to memory, testing effect)
Spacing effects - convincing your brain you might need the info even if you haven't used it in a while
Testing effect:
More than just an assessment tool, enhances memory retention to a greater extent than additional studying
Spaced learning and repeated testing = keys to memory
Spaced learning = convince your brain you’ll need this info over long intervals
Testing effect = retrieval practice is much better than continued studying; maybe because it builds new connections to other cues
Understanding is more important than repetition or anything else - if you fully understand materials you'll be much better able to remember them than any other way
Memory is constructive, not a video camera
Reconstructive process rather than simple and accurate reproductive process
Not always accurate representations of past
Memory errors - DRM procedure (remembered a word that wasn't in list because it fit the category = mistake ‘recall’ of theme words
Source monitoring is a particular challenge
Memory is reconstructive and malleable
Filter out or fill in missing pieces of info to recall coherent ⇒ misinformation effect : incorporating misleading info into ones memory of an event
Elizabeth loftus
Source memory - process of determining origins of our memories
Source monitoring error = misidentifying source of memory
Very frequent source of false memories
Fictional names were thought to be famous
Emotional memory
Remembered more easily and vividly
Emotion improves memory and benefit becomes greater with time (may enhance consolidation)
General finding is ‘overfocus’ → remember the gun better but the perpetrator the same or even less well
Flashbulb memories
Memories for surrounding shocking highly charge important events (9/11)
Where you were and what you were doing
Subjective experiences - highly emotional vivid and very detailed
Can be inaccurate or lacking in detail despite confidence
We forget our individual memory of these vents at almost exactly same pave of other autobiographical memories
Narrative rehearsal hypothesis - repeated reviewing/hearing of event (constructive memory)
False memories
Can we implant false memories
Can remember the story but don't remember the source or how you heard it