Sensory Memory Characteristics
- Brief storage of stimulus info (input buffer)
- Info enters automatically
- W/out attention, info decays (or is replaced)
- Iconic: Visual
- *Echoic**: Auditory
Visual Persistence
persistence of a visual stimulus beyond physical duration
Ex: Lightning-Sensory activity doesn’t outlast each bolt
White font/ black ground Exp
Iconic memory
- Task: “When you hear the tone, report all of the letters from the signaled row.”
- Changed to black on black
- Max duration in iconic memory: 5 sec
- Decay is faster for weaker contrasts
Bar/circle cue Exp
Iconic memory
Bar cue:
Results paralleled findings from earlier studies w/ tone as cue.
Circle cue:
Accuracy was much poorer
Backward masking: circle “overwrites” memory trace for letter (some Ps reported seeing only the circle)
Short Term Memory Characteristics
Temporary storage of info
– Limited capacity (smaller than SM or LTM)
– Duration depends
Rehearsal
– Maintenance
– Increases prob. of transfer into LTM
Serial Position Curve Exp
- 20 nouns presented one at a time - 5 s intervals
- “Study by repeating aloud words on the list during each interval”
- “free recall”
Primacy & Recency Effects
Primacy: Extra rehearsal for 1st few items (transferred in2 LTM)
Recency: Last few items still in STM
How to weaken recency effects
Distractor Task
- Backwards counting btwn study & recall (delay)
- Alter amount of time to rehearse (slower presentation=more stored in LTM but no effect on recency)
- Alter # of items on study list
- cant change primacy effects
Brown-Peterson Task
- You’ll see a 3-letter trigram (PWL)
- Then you’ll see a 3-digit #
- Count backwards (out loud) from that # by 3s until I say STOP
- Distractor task used to minimize rehearsal of “DKM”
- IV = Time spent counting backwards
- DV = Recall accuracy
- Conclusions W/out rehearsal, info forgotten quickly.
- BUT time & amount of interference confounded
Probe-Digit Task
Task
- Ps heard a list of 16 digits
- Final digit=repeat (probe digit). Cue to recall the digit that followed first appearance of probe on list
- IV1 = Reading rate, IV2 = # of items b/t probe & 1st appearance, DV = Correct recall
Conclusions:
Amount of forgetting similar for slow & fast # of intervening items is critical
Retroactive Interference (RI)
B-P Task
Memory for one event is impaired by later events.
- RI w/in trials
Proactive Interference (PI)
B-P Task
Interference
Memory for one event is impaired by earlier events.
- build-up of PI across trials
- stronger when TBR items across successive trials are more similar
Category switch Exp
- Fruit to fruit, 2. fruit to vegetables, 3. fruit to flower, 4. fruit to careers
- Demonstrates release from PI
Maintenance rehearsal
- rote repetition
- Less likely to leave permanent record of info in LTM
Elaborative Rehearsal
- think about what the TBR info means, how it is related to other things we know
Generation Effects & synonym/Rhyme Exp
Connections generated yourself are better remembered than those by others
- Generate syn (deep) or just read (shallow)
- Generate rhyme (deep) or read (shallow)
Conclusions: Support for DoP. Memory is better for gen-meaning(syn) than gen-sound(rhyme)
Transfer Appropriate Processing
probability of remembering better when conditions at test (retrieval) match those at study (encoding
aka
“Encoding specificity hypothesis” or “State dependent learning”
Working Memory
Limited-capacity system for temp storage & manipulation of info for:
- comprehension
- problem- solving
- learning
- reasoning
Baddeley’s model
Head honcho in charge of info processing = Central Executive
Assisted by 2 rehearsal buffers (Phonological loop, Visuo-spatial sketchpad)
Episodic Buffer: Storage system that can integrate info from different modalities.
- keep info around for current task(s)
- But it is deeper processing (in CE) that leads to LTM encoding
- Each can take input directly through perception or indirectly through imagery
Central Executive
coordinates processing: Focuses on specific parts of a task & switches attention from one part to another
Phonological Loop
“inner ear”
Articulatory Loop
maintenance rehearsal (process)
- sound-based, subvocal repetition - “inner voice”
Supporting evidence
- Phonological similarity effects
- Word length effects
- Articulatory suppression effect
Phonological Similarity Effect
- Memory is poorer when Ps rehearse a set of items that sound similar vs. not
- Intrusion errors during retrieval:
- Often phonologically similar to TBR words, Rarely similar in meaning
- Rehearsal in articulatory loop based on sound more than meaning
Word Length Effect
Found
- # of syllables appears to be more important than # of letters
- what predicts WM capacity appears to be
speed of rehearsal rather than # of items
