Memory

PSYC*2650- Memory


The Many Forms of Memory

  • Almost all behaviours are shaped by some form of memory

  • System vs. Processes

  • Forgetting is more like information being unavailable rather than information being gone


Traditional Memory Models

Sensory Memory

  • Sensory memory is the name given to the store that briefly retains sensory information

  • Hasn’t undergone meaningful processing

  • Thought to store sensory information in a raw or uninterrupted form

  • Sensory input from the environment contains far too much information for the brain to process all at once

    • We need something to filter what is important

  • The sensory memory system is thought to be responsible for two major phenomena in the visual system:

    • 1. Visible persistence

      • Both peripheral (retinal) and neural origins

      • Like an “afterimage”- visual information remains visible briefly after it’s physical presentation

      • Duration of persistence is inversely related to stimulus duration

    • 2. Informational persistence

      • Information can be extracted from a stimulus for a brief period of time after the stimulus itself is no longer present

      • How much information can be apprehended from a brief exposure to a visual stimulus?

      • People can typically only retrieve 3-4 out of 12 letters

  • Iconic memory

    • Large capacity

    • Brief duration


Short-Term/ Working Memory

  • Combines storage aspect of STM with active attention component

  • WM is a memory system that allows for short-term retention AND active manipulation of information

WM as a Trait

  • Individual differ in terms of their working memory abilities; WMC predicts many different outcomes

  • Working memory capacity (WMC): “is not really about storage or memory per se but about the capacity for controlled, sustained attention in the face of interference or  distraction” (Engle, 1999)

Distinction Between Working Memory and Long-Term Memory

  • Primacy Effect: Rehearsal of the first few items results in transfer to LTM

  • Recency Effect: ‘Dumping out’ of information in WM - recent info still left in the working memory

  • Predictions:

    • 1) preventing rehearsal of last few items should wipe out recency effect

    • 2) allowing for more rehearsal during item presentation should boost primacy effect

  • Variables that affect the primacy effect do not effect the recency effect as both effects are controlled by different  systems


Long-Term Memory


Declarative Memory

  • Episodic Memory

    • Bound to context (i.e. time and place)

    • Concerned with personal past

  • Semantic Memory

    • Not contextually-bound

    • Concerned with general knowledge knowledge and facts

Episodic Memory

  • Laboratory-based method that allows fro experimental examination of learning and memory encoding: study-test paradigm

    • Study: judge the following words in terms of your liking

    • Test: what were the words you say 20 minutes ago (recall) was this one of the words you saw (recognition)

  • Study phrase serves as episode whose details need to be recovered at test


Processing Principles

  • Craig & Lockhart (1972)

  • Critical how info is processed at time of encoding -> level/depth of processing

  • Different levels of processing words:

    • Perceptual feature analysis

    • Processing of sounds (phonology)

    • Processing of meaning (semantics)

  • Empirical test of framework with incidental orienting tasks (level of processing induced by experimenter)

  • Typical advantage of deep over phonological over shadow encoding on recognition memory test

  • LoP effect does not just reflect differences in time spent processing item at encoding

  • Instead type of processing, and integration with existing knowledge critical (integration more effective in yes decisions )

  • There are limits on when semantic processing is beneficial for memory

  • Memory performance depends not only on how information is encoded by also on the way in which it is tested -> transfer-appropriate processing

  • Memory is related to how it is encoded (meaning, sound, etc.)

  • Memory for an event depends on the interaction between properties of the encoded information and properties of the retrieval cue -> encoding specificity principle

  • Contextual reinstatement: change of circumstances between learning and test decreases memory

  • State-dependent memory: you will remember info better if you are in the same state as when you learned information in (e.g. drunk, sober, etc.)

  • Mood -dependent memory: you will remember info better if you are in the same mood as when you learned information in (e.g. happy, sad, etc.)


Autobiographical Memory

  • Childhood/Infantile Amnesia: little reliable memory for events experiences before four years old

    • Paucity of language (difficult to ‘label’ our experiences)

    • Difficultly binding different elements of an experience into a whole

    • We have difficultly binding memories which is why we have issues saving these memories

  • Reminiscence Bump:  enhanced memory for (episodic and semantic) facts of adolescences and young adulthood

    • Disproportionate enhancement of memory from 18-30 yo

    • Many autobiographically consequential experiences- experiences that shape a persons identity

    • Highly distinctive events- birth of first child, marriage, etc.

  • Semantic memory: information in semantic memory is though to be organized in a network with related concepts near each other

    • Things are connected by similarities (red= fire, apples, stop signs, etc)

    • Pieces of information are all connected and can connect each other

    • Spreading activation: activation of one concept ‘primes’ the activity of related concepts

    • Semantic priming- RT to targets is faster when preceded by a related primes  (i.e. nurse- BUTTER; bread- BUTTER)

      • Reaction time is much faster when words are related

  • Semantic priming is influenced by automatic spreading activation activation AND controlled expectations

    • When prime is BODY expect a target from the building category (i.e. BODY- door) RT will become faster over time than automatic processes RT as we have overwritten the automatic processes

    • Spreading activation yields facilitation at short SOAs, but inhibition at long SOAs as expectancy develops

Non declarative (Implicit)  Memory


Memory Without Awareness

  • Prior experience can shape behaviour without conscious awareness

  • Direct tests: probes memory with explicit reference to prior experience (e.g. recall; recognition)

  • Indirect tests: probes memory without making any reference to prior experience (e.g. priming)

    • Measures behaviour to demonstrate how memory affects behaviour

  • Example of direct vs. Indirect:

Sequence Learning (Nissan & Bullemer 1987)

  • Sequence learning- patients shown four boxes and asked to push a button (1-4) about where the circle shows up in the box

    • The circle should follow a subtle pattern

    • RT’s of when people press buttons get faster as they implicitly learn where to except the next

    • RT’s are faster to trained sequences as compared to random sequences

  • Implicitly Sequence Learning is often acquired without any explicit knowledge of actual sequence itself

  • Sometimes environment is predictable and the brain will implicitly pick up on these patterns even when we are not consciously aware of them

Jacoby experiment

  • Jacoby (1983) proposed that explicit memory was conceptually-driven and implicit memory was data-driven

    • Phase 1: read aloud a word preceded by (a) no context, (b) an antonym read silently, or (c) an antonym to which the word was generated

    • Phase 2: Given one of two tests

      • direct test- recognition test for words in Phase 1

      • Indirect test- identify briefly presented word

  • Recognition in direct test is best when word is generated in Phase 1

  • Perceptual identification is better in No Contact condition

  • Recognition is better in the Context and Generate conditions

  • The same manipulation has different effects depending on whether the test is direct or indirect- different processes involved in each type of test

  • If explicit and implicit were the same and playing by the same rules than they should respond the same thus proving they are different

  • Explicit contamination-Just because the experimenter doesn’t make reference to the encoding list doesn't mean that participants want intentionally use words from the lost to complete each stem

Method of Opposition (Jacoby, 1991)

  • How can we test apart implicit vs. Explicit influences on performance?

  • Two different conditions:

    • Inclusion: complete stem with word from list or with first word that comes to mind

    • Exclusion: complete stem using words no in the encoding list

  • If stems are still completed with studied words in the exclusion condition, then it must be an implicit influence of prior experience

  • THIS WILL BE ON THE MIDTERM

Memory Distortion and Illusions

  • Memory is not always accurate; susceptible to various forms of error

  • Important real-world implications

  • Legal system- wording of questions can inform an individuals memory

  • Medicine- questions the doctors ask can inform and shape our memory of our symptoms

  • Different types of memory errors

    • Are all errors in episodic memory errors of omission?

      • i.e. lack of availability of information due to poor encoding and/or forgetting

    • Or do we also make errors of commission

      • i.e. false remembering of episode or episodic detail that happened in different way or even never happened at all?

Memory Metaphors

  • Memory as a storage warehouse

    • Once memory is retrieved it is just as it was when it was stored

  • Memory as a detective

    • Memory as an active process of inference and re-construction

    • Reconstruct the past using clues, inferences and heuristics (rules of thumb)

Episodic Memory Illusions

  • Merry illusions studied with Deese / Reedier-McDermott (DRM) paradigm:

    • Highly robust effect of false remembering of critical lures in recall and recognition

    • Occur even when subject is informed about potential for memory illusions and instructed to avoid them

    • Underscores the powerful role of inference and re-construction in episodic memory

What Leads to False Memory?

  • Role of heuristics

    • Rules of thumb that guide our judgements

    • Serve to maximize cognitive cognitive economy

  • False remembering in DRM paradigm results from an over-reliance on gist information

  • Note that relying on such simple heuristics can have a profound effect on subjective experience

  • Prone to error but often right

  • When they work they work incredibly well

  • ‘False Fame’ Study by Jacoby et al. (1989)

    • Phase 1: participants read a list of fictitious names aloud for a study or pronunciation (spread and accuracy)

    • Phase 2: later, shown a lost of names along with old and new fictitious names. Test was either immediate or delayed (24 hr)

    • Reminded that the earlier list was non-famous name

    • Results: probability of judging a name as famous

      • Famous= .55

      • Non famous

        • New= .08

        • Old = .16

    • Some non famous names became famous overnight

    • That is the of “familiarity” was miss attributed to the fame of the name rather than the source (it was on the first list)

    • There is no such effect if the test is immediate because the surge of the familiarity is recalled-that is “I recall it being on the list”

The Role of Heuristic in Memory Judgements

  • Jacoby & Whitehouse (1989)- manipulate fluency of test items with masked priming

  • Showing individuals a brief flash of the word ‘river’ will make it more likely for participants to misattribute river as a word on the list

  • When unaware of the matching prime, participants attribute facilitated processing to prior exposure

  • When aware of the matching prime, participants attribute facilitated processing to the prime

  • Fluency Heuristic leads to false recognition

  • Fluency Heuristic- judgements of familiarity are the result of ease of processing (fluency)

    • Fluency may be due to prior exposure (as in false alarm)

    • Fluency may be due to experimental manipulation (illusion familiarity)

Episodic Memory: Suggestibility

  • Individual’s tendency to incorporate incongruent for from external sources into personal recollections from episodic memory

  • Situations in which suggestibility can be important factor:

    • Eyewitness tetsimony in police investigations and courtroom

      • Can reports from eyeywtiness reflect memory illusions due to suggestions during interrogations

      • When witnesses have high confidence in their memory they are often right

      • Witness’ memory can only be questioned once to avoid suggestion

    • Recovered memories from abuse in psychotherapy

      • Can they reflect memory illusions due to suggestions in therapy?

  • E.Loftus conducted groundbreaking research on suggestibility in memory

    • Demonstrated the misinformation effect when people who witnessed an event are later exposed to new and incongruent information about it their recollections often become distorted

    • Shown a video of a car pulling up to a stop sign and getting to a collision and then ask “how fast was the car going when they car pulled up to the yield sign” and people will misremember the sign as a yield sign

    • Wording also greatly important (smashed vs. Hit car)

Memory Illusions and Emotion

  • Are memories for emotional, impactful events immune to error

  • “Flashbulb memories”

  • Tend to be believed with greater confidence

  • Tend to be associated wth great vividness

  • But are they more accurate than everyday?

    • Talarico & Rubin (2003)

    • Asked 54 people to answer questionnaires about their memories for the 9/11 attacks and ordinary events right after 9/11

    • Re-tested either 1 week later, 6 weeks later or 32 weeks after

    • Participants recalled emotional events with higher confidence but were not anymore accurate than everyday events

Memory Disorders

Patient H.M

  • When told a joke, despite the fact H.M had no recollection of the joke he acted as though he had heard it before and was less amused

Amnesia

  • The hippocampus is thought to be critical for long-term declarative memory

  • Medial temporal lobes- in the temporal lobe and contains the hippocampus

  • Anterograde amnesia- inability to form new lasting memory

    • Most patients suffer at least a little anterograde

  • Retrograde amnesia- inability to remember events that occurred prior to injury

    • Usually only forget events around the time you suffered the injury (you won’t forget your entire life just more recent memories)

  • Global amnesia - having both anterograde and retrograde amnesia

Are all forms of memory impaired?

  • Amnesia often have preserved implicit memory and can still learn even if they cannot recall it

  • Impairment in tests of declarative memory (e.g.g free recall, cued recall)

  • Normal performance on indirect test of memory (e.g. stem completion- implicit memory)

  • Normal semantic priming

Neurodegenerative Dementia

  • Progressive and slow decline of nervous system

  • Symptoms referred to as “dementia”

Alzheimer’s Disease

  • Build up of neurofibrillary tangles and amyloid plagues

  • Plagues and tangles “choke off” neurons killing them

  • AD can only be formally diagnosed  post-mortem

  • Post-mortem analysis often finds plaques and tangles present in asymptomatic individuals

  • Strong genetic component

  • Difficult to diagnose as many diseases are similar to one another

  • Education and exercise seems to be a protective factor

  • Some have suggested it is more prevalent among women but it is unclear as to why (seems to be due to longer life expectancy)

  • Neurodegeneration often begins in medial temporal lobe structures (e.g. entorhinal cortex)

  • Results in progressive memory decline

  • Pathological changes can begin 10-20 years before symptoms appear

  • Episodic memories are typically first affected by degeneration spreads to other cortical areas, yielding additional cognitive impairment as disease progresses