PSC 130 MIDDY

Hermann Ebbinghaus

Curve of forgetting - humans forget information rapidly; 50% within an hour + 70% within 24 hours

• a lot of our experience is not immediately accessible

Sir Frederic Bartlett

Interested in cultural and social elements that contribute to how we remember

• remembering is reconstructed → act of remembering = act of imagination

short term memory

ability to keep limited information in mind for quick access to use for an immediate goal

• limited amount of things that we can remember (7 ± 2)

• we have limited chunks of knowledge - tend to separate things into groups and remember 4 ±1 chunks of information

Memory Span

number of items that can be repeated immediately in the correct order 50% of the time

Short term storage

Flexibility to handle incoming information + recall information that we’ve previously learned

• includes rehearsal, coding, decision, strategies

Long term storage

passive storage where memories are stored for later use

• includes things that are permanent - like a hard drive/database

Baddeley’s working memory model

central executive system responsible for our visuospatial sketchpad + phonological loop

• separate buffers to handle verbal vs visual information

  • phonological loop - mediates verbal maintenance/rehearsal

  • visuospatial sketchpad - mediates visual maintenance

different ways to store information: spatial vs sound/verbal information

central executive

attentional control system

• responsible for attention control of working memory

• updating contents of working memory

• manipulate info in working memory

• coordinating concurrent mental activities

• inhibition of distracting information

phonological loop

mediates speed-based rehearsal

• phonological storage

• articulatory control process

phonological similarity effect

hard to remember something if they all sound the same → immediate recall is impaired when items sound alike

• recall decreases w/ # of syllables we must learn (the more syllables = harder to remember something)

• recall decreases w/ length of time required to pronounce words

• sounds impact our ability to remember

Heyer & Barrett (1971)

asked participants to do position recall vs identity recall

• visual distraction impacts ability to remember → destroys position memory

• verbal distraction impacts identity recall → harder to remember the type of letters listed

dorsal

where/spatial

• top part of the brain

• stream connected to back of brain analyzing information

ventral

what/object

• more details about the shapes and intricate details of something

• differences between objects w/ meaning

• identity but no location of the memory

dorsolateral prefrontal cortex

critical for functions that are associated with “central executive”

• chemical flows are controlled by prefrontal cortex

• human prefrontal cortex allows it to operate the way it does

working memory storage will activate contents of a specific brain area → different areas of brain activated depending on the type of memory

effects of prefrontal damage

• dysexecutive syndrome - dysfunction in executive functions

• distractiability

• poor planning

• “perseveration”

D’Esposito & Postle (1999) + Prefrontal Cortex and Working Memory

span - requires immediate recall; testing working memory capacity (storage) but does not require maintenance across a delay

delayed response - requires retention across delay

• maintenance + control (keeping things in memory even when being distracted by others)

• patients with PFC lesions are impaired on delayed-response tasks

Levels of Processing (Craik & Lockhart) - Executive functions contribute to both WM and LTM

levels of processing lead to better ways of thinking

• rehearsing information does not increase the likelihood of something being remembered

• memory encoding is a byproduct of information → learn without trying if we are using the right strats

memory is stored in the way it is processed

brains go through different levels of processing

sensory - lowest

• shallow - seeing faces

• deep - forming connections

more information = more memory tracks

• more processing will have richer brain tracts

deeper processing = more distinctive memory = better memory performance

prefrontal cortex

critical for executive function

areas in the left prefrontal cortex (PFC) are more active during deep than shallow processing

• brain activity higher for things remember

• more activity in learning = remember on test

• deep processing = deeper memory trails to remember

generation affection

effect of elaboration is stronger for materials that you have to generate yourself

• more likely to remember words when you have to fill

if you can visualize = easier to remember something

• visual imagery to encode words & memory

• stores & making it concrete = remember something better

item-speciifc encoding

memory that stands out

• deep encoding produces more distinctive memory trace for each time → more distinctive memories are easier to remember later on

relational encoding

organizing and sorting information in memory

• focus on similarities among multiple events

• do well if u can organize information and make it as distinctive as possible

• associate new name w/ group they belong to and fun fact abt them

Relatedness can improve

recall!

• organize the information we’re trying to learn (ie. clumping/clustering)

• can remember words better if they’re related and form a relationship between them

BUT relatedness can also lead to memory distortions

retrieval process

• act of remembering based on retrieval cues that help you search for and activate memories for past events

• interaction between encoding & retrieval determines how you will remember an event

• retrieval cues help you find what you are looking for

Transfer appropriate processing

memory better if the type of information encoded is appropriately based on our retrieval test

• how you retrieve drives how you should encode information

deep encoding is better than shallow encoding → memory best if material is processed same was at study

Encoding specificity principle

effectiveness of retrieval cue depends on how well it aligns with the way an item was originally encoded

• experience of recollection relies on interaction between cue and memory trace

• consider the interaction between the retrieval cue and the way information is encoded

processing

maintenance rehearsal/shallow processing

elaborative rehearsal/deep encoding

item specific encoding

helps you form rich distinctive memories

relationship encoding

helps you form meaningful links between memories that might ultimately compete with one another

context

what gives memory power, binds episodic memory together

context dependent memory

information is remembered best when retrieval and encoding contexts match

• recall best when you’re in the same “environment” as when you learned something

• context is not a place → study of mind

• imagining context can make something easier to retrieve

mood-state dependent memory

mood-congruent information

• tend to remember things/retrieve the type of memory based on mental state that your’e in

• emotions are part of the context that we remember in

free-recall

context only impacts ability on free recall where you have to regurgitate information

• travel back up the information and context to remember

interference

harder to retrieve words if the words are similar → interference creates competition in memory, making it harder to remember

proactive interference

past impacts new memory

• learning #1 can cause you to learn #2 more slowly

retroactive interference

new impacts remembering old

• learning #2 causes you to forget #1

contextual interference

context is part of the memory and is a source of interference

• context can be a retrieval cue → more things learned = greater interference

how to beat interference

• distinctiveness

• organization/integration - remembering tool to generate possibilities of fruit being recalled

von restorff effect

we’re more likely to remember words that stand out in a cluster of similar words

• isolated items are typically remembered better than other items

• distinctiveness reduces interference during retrieval

semantic memory

meaning of what?

• organized knowledge a person possesses about words and symbols

Episodic memory

spatiotemporal context (where/when)

• receive and store information about temporally dated episodes or events

• exists on a timeline: we remember “rough” times and logic to decipher when something happens

• high temporal continuity: close to each other in time & how things are related

temporal context

in what pattern does something fall in the order it occurs

• time = change in mental contexts

• memories that occurred around the time are linked more closely than memories for events

• states change gradually over time → changes in minuscule things can influence memory & mental contexts shifts to gradually change over time

Recency effect

tend to remember the last thing of a list

• ability to recall a word determined by where on the list it falls

• delay that occurs - no more recency effect BUT random task - recency effect is back

• recency memory is not just due to work memory or short-term memory

items far away are lumped together making it harder to differentiate timescale at the end of the list

conditional recall probability

once someone recalls a certain event, they’re likely to remember events close to that timeline

• memory is like a time machine - you can travel close in time to things that you remember/context that you draw upon

• we remember things little by little - less likely to recall backwards since time moves forwards

Catastrophic forgetting

difficulty adapting new information into neural networks

• systems at a certain point stop learning because of limited ability to recall past vs present = continual learning problem

retrieval cues

pieces of information that elicits retrieval of a memory from prior event

• different types of tests = different types of retrieval cues

Signal detection theory model of recognition

W’ere always attempting to discriminate between signals in our brain

• decide quickly if we’ve seen something or not

D’ (discriminability)

average difference in memory strength between studied and unstudied items
- tell apart faces we’ve met vs never met before

recall is more prone to failure than recognition

strong memory = recall the item but weak memory = recognize it

generate-recognize midel - Kintch

if I recall a word → i must generate word on my own and then recognize it from the list I generated myself

• any word that can be recalled should be recognized BUT its hard to recognize things in isolation

• recognition → context-dependent

less likely to recognize

frequent words

more likely to recall

frequent words

recognition is supported by 2 separate and independent processes

familiarity
recognition

familiarity

knowing - aware that something exists

• all items elicit some familiarity so process never fails

• type of information - item

• fast retrieval

recognition

remembering - can relive the memory

• process sometimes fail

• type of information = item + context

• slow retrieval

response deadlines

familarity causes false alarms while recollection helps say no correctly

• recollection is aid to overcome bias

• more false alarms for high frequency words

remember-know method

retrieved memories can be distinguished on the basis of subjective experience

• remember - recollect specific details about an event

• knowing/familiarity - subjective feeling that items are familiar but you can’t recall any contextual information about it

familiarity

supports recognition by giving you sense of how strong memory is for a particular item

• does not support free recall/cued → lack deep encoding information

• feeling

recollection

supports free and cued recall

• supports recognition by helping you pull up detailed information about the item and study contex

Based on Ebbinghaus’ studies of forgetting, you would expect to show the most forgetting (i.e., the biggest difference in memory performance) between:

A test given 20 minutes after study vs. a test given 1 hour after studying

difference between the "long-term store" and "short-term store" in the Modal Model (e.g., Atkinson & Shiffrin, 1968)

The long-term store is responsible for storing information permanently, while the short-term store is responsible for temporarily holding information for immediate use.

Core functions of the central executive as described by Baddeley?

Coordinating tasks that are happening concurrently

Manipulating information in working memory

Inhibiting distractions

Cortical Areas in the Dorsal Stream are more involved with _____ Working Memory, whereas the Areas in the Ventral Stream are more involved with ____ Working Memory

Spatial, Object

Heyer & Barrett (1971) found that verbal distraction disrupts recall of letters (verbal recall), and visual distraction disrupts recall of position (visual recall). This suggests that

There are separate working memory systems for verbal versus visual information

Older adults with extensive white matter hyperintensities (WMH) on MRI scans have:

Reduced prefrontal cortex activation and worse working memory

Craik and Lockhart's Levels of Processing Framework proposed that

Elaborative rehearsal should result in better learning than maintenance rehearsal

In order to test the Levels of Processing Framework it was necessary to:

use incidental encoding tasks, so that experimenters could manipulate how information was processed

Alex is trying to memorize a list of words by sorting them into different categories. This is an example of:

relational encoding

Morris, Bransford, and Franks (1977) had subjects encode a list of words either by deciding whether each word had a certain sound in it, or by processing the meaning of the word. Next, subjects in one group were given an item recognition test on the words they studied, whereas subjects in another group were given a rhyme recognition test (i.e., “Did you see a word that rhymed with ‘eagle’?”). Based on the Levels of Processing Framework, one would predict that:

subjects who did meaning-based encoding would show better memory performance, regardless of the test condition

Morris, Bransford, and Franks (1977) had subjects encode a list of words either by deciding whether each word had a certain sound in it, or by processing the meaning of the word. Next, subjects in one group were given an item recognition test on the words they studied, whereas subjects in another group were given a rhyme recognition test (i.e., “Did you see a word that rhymed with ‘eagle’?”). Based on the Transfer Appropriate Processing Framework, one would predict that:

subjects who did meaning-based encoding would show better performance on the item recognition test, but subjects who did sound-based encoding would show better performance on the rhyme recognition test

According to Tulving’s Encoding Specificity Principle, successful memory performance depends on:

the interaction between a retrieval cue and the type of trace that was formed at encoding

According to Why We Remember, Endel Tulving said that a key characteristic of human consciousness is that we are “capable of mental time travel, roaming at will over what has happened as readily as over what might happen, independently of the physical laws that govern the universe.” Tulving is describing the experience of retrieving:

 

episodic memories

If you want to recall a specific piece of information, which type of memory test would benefit most from remembering the context in which the information was learned?

Free Recall

In a study by Goodwin et al. (1969), participants learned a list of items in two different states (either while sober or after drinking 10 oz of 80-proof vodka). They were then asked to recall the list either while sober or drunk. These results suggest that:

While drinking was not overall beneficial for memory, participants who drank alcohol before study did better if they had alcohol before the test phase than those who were sober at test. This is an example of state-dependent memory.

Based on the Von Restorff effect, we can expect that:

“Apple” would be better remembered if it was in a study list of furniture items than if it was in a study list of fruit.

The lag-recency effect provides evidence to support the idea that Episodic Memories are

temporally organized

On a recognition memory test, when a subject incorrectly says “old” for an unstudied item, we would call that a:

false alarm

Generate-Recognize models state that:

recall involves 2 processing stages, but recognition only requires one

In Tulving’s Remember/Know procedure, “Know” responses are thought to be based on:

familiarity

Research on word frequency effects suggests that:

an infrequent word like “armadillo” is less likely to be recalled, but more likely to be recognized on a memory test, relative to a more frequent word like “horse”.

The "mirror effect" in recognition memory refers to the fact that:

low frequency words are associated with more hits and fewer false alarms than are high frequency words