Long Term Memory Retention Forgetting and False Memory Systems

Core Concepts and Questions in Long-Term Memory

• Fundamental Inquiries of Long-Term Memory (LTM):

  • What specific factors determine how efficiently humans retain and recall information over time?

  • What is the ultimate fate of stored information that appears to be "forgotten"?

  • What structural and operational insights into memory systems can be extracted from the study of amnesia?

Long-Term Memory Successes and Extraordinary Memory

• Very Long-Term Memory and Permastore:

  • Research by Bahrick demonstrated that certain long-term memories reach a state of permanence termed permastore.

  • Subjects evaluated decades after initial learning successfully recalled high school Spanish vocabulary and recognized the names and faces of high school classmates $30$, $40$, or even $50\text{ years}$ later.

• Case Studies of Extraordinary Memory Capacity:

  • Solomon Shereshevskii (Subject "S"):

    • Studied extensively by Russian psychologist A. R. Luria and featured in Mind of a Mnemonist.

    • Worked as a newspaper reporter who famously never took written notes during assignments.

    • Demonstrated practically unlimited memory capacity, easily repeating back lists of $70\text{--}80$ words or numbers after delays of several months or years.

    • Later worked professionally as a mnemonist.

    • Cognitive Drawbacks: S struggled with severe concentration issues because mental images would automatically appear and intrude upon present conscious thoughts. To intentionally forget information, he had to mentally picture himself writing the items on a blackboard and erasing them.

  • Conductor Arturo Toscanini:

    • Demonstrated incredible musical imagery and memory capacity.

    • Estimated to know by heart every note of every instrument for approximately $250$ symphonic works, the complete words and music for roughly $100$ operas, along with extensive chamber music, piano pieces, cello, violin compositions, and songs.

    • 
      

    

    • The Tale of the Panicky Bassoonist: In St. Louis right before a concert, second bassoonist Umberto Ventura approached Toscanini in distress after discovering that the key for the lowest note on his bassoon was broken. Toscanini shaded his eyes, reflected briefly, and reassured him that the specific broken note did not occur anywhere in that evening's concert program.

• Sensory Synesthesia:

  • Definition: A neurological condition characterized by unusual, involuntary cross-modal associations between different sensory modalities or cognitive representations. Individuals who experience this are referred to as synesthetes.

  • Color-Grapheme Synesthesia: A common variant wherein individual letters or numbers are perceived as inherently possessing specific, consistent colors.

  • Personification Synesthesia: A form where abstract numbers, days of the week, or months of the year evoke distinct individual personalities.

  • Subject "S"'s Synesthetic Associations with Numbers (Luria, 1968, p. 31):

    • $1$: A proud, well-built man.

    • $2$: A high-spirited woman.

    • $3$: A gloomy person.

    • $6$: A man with a swollen foot.

    • $7$: A man with a moustache.

    • $8$: A very stout woman—a sack within a sack.

    • $87$: A fat woman standing next to a man twirling his moustache.

• Visual Memory Capacity in Normal Individuals:

  • Roger Shepard (1967): Exposed subjects to $612$ pictures of distinct objects and subsequently conducted a recognition test.

  • 
    

  

  • Shepard (1967) Time-Delay Recognition Results:

    • Immediate Test: Near $100\%$ correct recognition.

    • $2\text{ Hours}$ Delay: Near $100\%$ correct recognition.

    • $3\text{ Days}$ Delay: Approximately $90\%$ correct recognition.

    • $7\text{ Days}$ Delay: Approximately $85\%$ correct recognition.

    • $120\text{ Days}$ Delay: Retention remained well above chance at approximately $58\%$ correct.

    • 
      

    

  • Standing et al. (1970): Expanded visual recognition testing to $2{,}560$ pictures, observing similarly exceptional retention.

  • Standing (1973): Tested subjects with $10{,}000$ pictures, demonstrating massive visual LTM storage capacity, with theoretical estimates projecting functional limits up to $1{,}000{,}000$ pictures.

Ebbinghaus, the Forgetting Curve, and Encoding Failures

• Hermann Ebbinghaus and Nonsense Syllable Research:

  • Pioneered the experimental study of memory by serving as his own subject to study the learning and retention of nonsense CVC (consonant-vowel-consonant) syllables (e.g., Koj, Bov, Wug).

  • Created a pool of $2{,}300$ meaningless CVC items on cards in a box, pulling out random sets for study trials until reaching $100\%$ error-free recall accuracy.

  • Because stimuli were explicitly designed to be meaningless, deep semantic processing was prevented according to Levels of Processing theory.

• The Ebbinghaus Forgetting Curve:

  • Calculated retention using the Savings Score formula:         $\text{Savings Score} = \frac{\text{Original Learning} - \text{Relearning}}{\text{Original Learning}} \times 100$

  • Retention Quantitative Data Over Time:

    • Immediate Recall: $100\%$ retention.

    • $19\text{ Minutes}$ ($20\text{ min}$): $58\%$ retention.

    • $63\text{ Minutes}$ ($1\text{ hour}$): $44\%$ retention.

    • $525\text{ Minutes}$ ($9\text{ hours}$): $36\%$ retention.

    • $1\text{ Day}$ ($24\text{ hours}$): $33\%$ retention ($38\%$ on graphical plot).

    • $2\text{ Days}$: $28\%$ retention.

    • $6\text{ Days}$: $25\%$ retention.

    • $31\text{ Days}$: $21\%$ retention.

  • The curve demonstrates an initial rapid drop in retention immediately following learning, which subsequently levels off over extended delays.

  • 
    

  

• Encoding Failure as a Cause of Forgetting:

  • Alexander Chase Quote: "Memory is the thing you forget with."

  • Apparent forgetting is frequently caused by a failure to encode the information in the first place.

  • Depth of processing principles dictate that shallow processing prevents the creation of stable long-term memory traces.

Factors Enhancing Retention and Recall

• Key Factors Improving Memory:

  • Spacing of learning sessions.

  • Structural organization of information.

  • State and context-dependent memory alignment.

• Encoding Specificity Principle (Tulving & Thomson, 1971):

  • Ability to retrieve a stimulus depends directly on the similarity between how the stimulus was processed during initial encoding and how it is processed at retrieval testing.

• The Spacing Effect:

  • First discovered by Ebbinghaus.

  • Massed vs. Distributed Practice: Massed practice (e.g., studying continuously for $3\text{ hours}$ in a single evening) produces inferior long-term recall compared to distributed practice (e.g., studying the same material for $1\text{ hour}$ per evening across three separate evenings).

  • Hypothesized Mechanisms behind Spacing:

    • Attention Attenuation: Sustained, unbroken exposure to identical material leads to declining attentional engagement.

    • Contextual Variety: Studying across distinct occasions exposes material to varied contexts, forming richer representations and multiple retrieval pathways.

    • Encoding Variability: Re-engaging with material at different times leads individuals to encode slightly different features or aspects of the material each time.

• Organization of Information:

  • Bower, Clark, Lesgold, and Winzenz (1969): Evaluated list recall when words were presented randomly versus structured in hierarchical trees.

  • Random List Example: Naples, World, Italy, Americas, Montreal, Bristol, Washington, Ottawa, Orlando, England, Europe, Dallas, Liverpool, Winnipeg, Rome, USA, London, Florence, Canada.

  • Hierarchical Organization Example:

    • World

      • Europe -> England (London, Liverpool, Bristol), Italy (Rome, Florence, Naples)

      • Americas -> USA (Washington, Dallas, Orlando), Canada (Ottawa, Montreal, Winnipeg)

  • Hierarchical organization substantially improves recall performance over unorganized word presentations.

Context-Dependent and State-Dependent Memory

• External Context-Dependent Memory Effects:

  • Godden & Baddeley (1975) Scuba Diving Experiment:

    • Subjects learned word lists either underwater ($\text{Water}$) while scuba diving or on the surface ($\text{Land}$).

    • Tested later for recall either underwater ($\text{Water}$) or on the surface ($\text{Land}$).

    • Results: Recall was roughly $40\%$ higher when testing context matched learning context ($\text{Water/Water}$ or $\text{Land/Land}$) compared to mismatched contexts ($\text{Water/Land}$ or $\text{Land/Water}$).

    • 
      

    

  • Grant et al. (1998) Noise Study:

    • Subjects studied material while listening to cafeteria noise over headphones ($\text{With noise}$) or in silence ($\text{In quiet}$).

    • Results: Test scores were significantly higher when testing occurred in matching acoustic conditions ($\text{Noise/Noise} \approx 6.2$, $\text{Quiet/Quiet} \approx 6.7$) than in non-matching conditions ($\text{Noise/Quiet} \approx 5.4$, $\text{Quiet/Noise} \approx 4.8$).

    • 
      

    

  • Infant Context Sensitivity (Butler & Rovee-Collier, 1989):

    • Infants taught to move a mobile by foot-kicking demonstrated strongest memory reactivation when tested in a crib with identical cloth bumper context patterns (e.g., red/black stripes vs. yellow/black squares).

    • 
      

    

  • Time-of-Day Context: Learning material at $3:00\text{ PM}$ yields superior recall when retested at $3:00\text{ PM}$ relative to retesting at $9:00\text{ PM}$.

• Internal State-Dependent Memory Effects:

  • Definition: Recall is improved when an individual's internal physiological or emotional state during testing matches their state during initial encoding.

  • Distinction: Context-dependent memory relies on external environmental variables; state-dependent memory relies on internal physiological/psychological variables.

  • Eich & Metcalfe (1989) Mood Experiment:

    • Subjects studied in either a $\text{Sad}$ or $\text{Happy}$ mood and were tested in either a $\text{Sad}$ or $\text{Happy}$ mood.

    • Results: Test scores were significantly higher under matched mood states ($\text{Sad/Sad} \approx 31$, $\text{Happy/Happy} \approx 26$) than mismatched states ($\text{Sad/Happy} \approx 16$, $\text{Happy/Sad} \approx 16$).

    • 
      

    

  • Clinical Implications (Bipolar Depression):

    • Information learned during a manic state is recalled best when testing occurs during a manic state.

    • Information learned during a depressed state is recalled best when testing occurs during a depressed state.

Theories of Forgetting and Information Availability vs. Accessibility

• Availability vs. Accessibility:

  • When an item cannot be retrieved from LTM, two distinct possibilities exist: whether the memory trace has physically vanished from storage (unavailable), or whether the memory remains stored but cannot be reached due to a lack of proper retrieval cues (inaccessible).

• Nelson (1971) Relearning Study ("It's Still There" Theory):

  • Subjects learned a paired-associate list (e.g., $43\text{-house}$, $67\text{-dog}$, $38\text{-dress}$, $77\text{-scissors}$).

  • After a $2\text{-week}$ delay, subjects completed a cued recall test (e.g., $43\text{-}\underline{\quad\quad}$).

  • Subjects successfully recalled $75\%$ of items; research focused specifically on the $25\%$ of items subjects "forgot".

  • Critical Manipulation: Subjects relearned the forgotten items under one of two conditions:

    • Same Pair Condition: Relearned original pairs ($38\text{-dress}$, $77\text{-scissors}$). Result = $78\%$ recall accuracy upon retesting.

    • Changed Pair Condition: Relearned cue paired with a new word ($38\text{-apple}$, $77\text{-kettle}$). Result = $43\%$ recall accuracy upon retesting.

  • Conclusion: Significantly higher performance in the Same condition proves residual memory traces persisted for "forgotten" items, demonstrating that forgotten LTM items remain available even when temporarily inaccessible.

• Theories Explaining Forgetting:

  • Decay Theory: Memory traces spontaneously weaken and erode over the simple passage of time through disuse.

  • Interference Theory: Forgetting occurs because competing memory traces interfere with retrieval.

    • Proactive Interference: Older long-term memories interfere with the retrieval of newer information.

    • Retroactive Interference: Newer long-term memories interfere with the retrieval of older information.

• Experimental Paradigms for Interference:

  • Retroactive Interference Paradigm:

    • Experimental Group: Learn List A -> Learn List B -> Delay -> Test List A.

    • Control Group: Learn List A -> [No Study] -> Delay -> Test List A.

    • Result: Experimental group recalls significantly less of List A than Control group because List B learning retroactively interferes with List A retrieval.

  • Proactive Interference Paradigm:

    • Experimental Group: Learn List A -> Learn List B -> Delay -> Test List B.

    • Control Group: [No Study] -> Learn List B -> Delay -> Test List B.

    • Result: Experimental group recalls significantly less of List B than Control group because List A learning proactively interferes with List B retrieval.

• Real-World Examples of Interference:

  • Retroactive Interference Example: Learning the rules of a new card game causes a player to forget or mix up the rules of a previously learned card game (New affects Old).

  • Proactive Interference Example: Opening a new bank account and going to an ATM, but repeatedly typing an old PIN because old memory interferes with recall of the new PIN (Old affects New).

Structure and Taxonomy of Memory

• Taxonomy of Memory Systems:

  • Memory

    • Sensory Memory

    • Short-Term Memory (STM) / Working Memory

    • Long-Term Memory (LTM)

      • Explicit Memory (Declarative Knowledge)

      • Implicit Memory (Non-Declarative Knowledge)

• Explicit vs. Implicit Memory Characteristics:

  • Explicit Memory: Involves conscious recollection and declarative knowledge. Tested via explicit tasks such as recall (reproducing stored items) and recognition (identifying previously encountered items).

  • Implicit Memory: Refers to unconscious cognitive or behavioral changes resulting from prior experience. Includes procedural knowledge (motor skills and habits), priming, and classical conditioning.

Neural Substrates and Types of Amnesia

• Brain Regions and Memory Taxonomy Mapping:

  • Explicit Memory Systems:

    • Semantic Memory (Facts): Medial Temporal Lobe, Hippocampal Region.

    • Episodic Memory (Personal Experiences): Medial Temporal Lobe, Hippocampal Region.

  • Implicit Memory Systems:

    • Skills and Habits: Striatum.

    • Priming: Cortex.

    • Classical Conditioning (Pavlovian): Cerebellum.

  • 
    

  

• Types of Amnesia:

  • Anterograde Amnesia: Inability to encode or learn new explicit long-term information following brain trauma (e.g., Patient H.M., Clive Wearing; depicted in films like Memento or 50 First Dates).

  • Retrograde Amnesia: Inability to retrieve explicit information learned prior to brain trauma. Typically temporally-graded, meaning memories for older events remain intact while memories immediately preceding the trauma are most vulnerable.

• Case Study: Patient H.M. (Henry Molaison, February 26, 1926 – December 2, 2008):

  • 
    

  

  • Surgical History: Underwent bilateral medial temporal lobe resection in 1953 to relieve intractable epilepsy, removing both hippocampi.

  • Cognitive Profile:

    • Severe anterograde amnesia (unable to form new explicit long-term memories).

    • Preserved pre-surgery facts and remote memories, preserved language abilities, and intact baseline intelligence.

    • Unable to recognize new people or remember major events occurring post-surgery (e.g., his mother's surgery).

    • Normal Working Memory: Could participate in active conversations normally, but would completely forget the interaction if the visitor left the room and returned.

  • Verbatim Subjective Experience Quotes:

    • "Right now I'm wondering, Have I done or said anything amiss? You see, at this moment everything looks unclear to me, but what happened just before? That's what worries me. It's like waking from a dream; I just don't remember."

    • "Every day is alone in itself, whatever enjoyment I've had, and whatever sorrow I've had."

• Role of Hippocampus and System Consolidation:

  • Memories are not stored permanently within the hippocampus; remote memories remain intact following hippocampal removal because they reside in neocortex.

  • The hippocampus and medial temporal structures function to "bind" distributed neocortical activity during encoding.

  • Standard Model of Consolidation:

    • Recent Memories: Retrieval depends heavily on the hippocampus because direct cortical connections have not yet matured. Hippocampal activation is High; Cortical activation is Low.

    • Remote Memories: As consolidation completes over time, robust connections form directly within neocortex. Hippocampal involvement is no longer required. Cortical activation is High; Hippocampal activation is Low.

    • 
      

    

• Preserved Implicit Memory in Amnesia:

  • While amnesic patients show severe impairments on explicit memory tasks, they demonstrate intact learning on implicit memory measures.

  • Classical Conditioning: H.M. acquired conditioned eye-blink responses (CS tone paired with UCS air puff to eye), driven by cerebellar learning structures.

  • Musical Preference: Amnesic patients form preferences for newly exposed melodies despite lacking explicit recognition of having heard them.

  • Mirror Reading Study (Cohen & Squire):

    • Compared Korsakoff's amnesics and Patient N.A. (severe anterograde amnesics) against normal controls on mirror-reversed word reading across $4\text{ days}$.

    • Stimulus Conditions: $50\%$ non-repeated words (implicit skill measure) and $50\%$ repeated words (implicit skill + explicit memory measure).

    • Results:

      • Non-repeated words: Amnesics and controls improved mirror-reading speed at identical rates, demonstrating intact implicit procedural learning.

      • Repeated words: Controls outperformed amnesics because controls benefited from explicit memory of specific words in addition to implicit procedural skill acquisition.

    • 
      

    

Eyewitness Memory and False Memory Phenomena

• Distortions in Eyewitness Testimony:

  • Eyewitness memory for complex events can be systematically distorted.

  • Individual witnesses may omit details that occurred, recall events that never occurred, or have memories altered by interim misinformation and questioning phrasing.

• The Misinformation Effect:

  • Definition: The alteration of memory caused by exposure to misleading post-event information.

  • Standard Experimental Design (Loftus): 1. View Event (e.g., video of car accident) -> 2. Post-event Misinformation (embedded in questions) -> 3. Memory Test.

• "Hit" vs. "Smashed" Experiment (Loftus & Palmer, 1974):

  • Subjects watched film clips of automobile accidents and answered questions regarding car speeds.

  • Verb Manipulation in Question: "How fast were the cars going when they [smashed / collided / bumped / hit / contacted] into each other?"

  • Estimated Speed Results:

    • Smashed: $41\text{ mph}$

    • Collided: $39\text{ mph}$

    • Bumped: $38\text{ mph}$

    • Hit: $34\text{ mph}$

    • Contacted: $32\text{ mph}$

  • Broken Glass Memory Test ($1\text{ week}$ later): Subjects were asked if they saw broken glass in the accident video (no glass was present).

    • Smashed Condition: $32\%$ incorrectly reported seeing broken glass.

    • Hit Condition: $14\%$ incorrectly reported seeing broken glass.

  • 
    

  

• Stop Sign vs. Yield Sign Experiment (Loftus):

  • Subjects viewed a slide sequence of a red Nissan driving down a street, stopping at a stop sign, turning right, and striking a pedestrian.

  • Interim Misinformation Question: "Did another car pass the red Nissan when it was stopped at the stop sign?" vs. "…when it was stopped at the yield sign?"

  • Forced-Choice Recognition Slide Test: Subjects chose between a slide depicting a stop sign and a slide depicting a yield sign.

    • Control Group (Non-misled): $75\%$ correctly selected the stop sign slide.

    • Misled Group: Only $41\%$ correctly selected the stop sign slide ($59\%$ selected the false yield sign slide).

• Implanting False Memories for Entire Events:

  • Hyman, Troy, and Billings (1995):

    • College students were interviewed about childhood events verified by parents, mixed with false events created by researchers (e.g., overnight hospitalization for high fever, spilling a bowl of punch on the bride's parents at a wedding reception).

    • Results across Three Interviews:

      • True Events: $80\%$ recall in 1st interview ($182/205$), climbing to $90\%$ ($191/205$) in 2nd, and $95\%$ ($195/205$) in 3rd interview.

      • False Events: $0\%$ recall in 1st interview ($0/51$); $18\%$ ($9/51$) recalled false events in 2nd interview; $25\%$ ($13/51$) recalled false events in 3rd interview.

    • 
      

    

• Imagination Inflation (Garry, Manning, Loftus & Sherman, 1996):

  • Subjects completed a Life Events Inventory (LEI), participated in guided mental imagination exercises of unexperienced events, and subsequently re-completed the LEI.

  • Results: Mentally imagining fictitious events significantly increases a person's belief and confidence that the events actually occurred in their past.

• Other Implanted False Memory Examples in Research:

  • Recalling a non-existent childhood birthday party.

  • Being lost in a shopping mall.

  • Evacuating a grocery store when emergency sprinklers went off.

  • Taking a hot air balloon ride.

  • Being attacked and bitten by an animal.

• Factors Moderating False Memory Susceptibility:

  • Event Plausibility: Plausible events are significantly easier to implant than implausible ones.

  • Repetition: Repeatedly presenting false suggestions increases false memory creation.

  • Imagination Inflation: Directing subjects to imagine the event visualizes context features, accelerating false memory formation.

  • Individual Susceptibility: Specific personality and cognitive profiles demonstrate increased vulnerability to false memories.