Psych 351A: Midterm 3 Review (Short Term and Working Memory)

Modal Model of Memory

An information processing model of memory developed by Atkinson and Shiffrin in 1968, argues memory acts as an integrated system with limited capacity, has 3 major structures (STM, LTM, sensory memory)

What are the 4 major processes present in the modal model of memory

1. Rehearsal

2. Encoding

3. Strategies to make a stimulus memorable

4. Strategies of attention

Sensory memory

Very short lived buffer for all or most information that hits our sensory receptors, holds large amount of info for a short period of time, has 2 subsets

2 Subsets of sensory memory

Iconic - persistence of vision, retention of the perception of light

Echoich - persistence of sounds, helps us integrate acorss a spoken word

Sperling’s sensory memory experiment

Flashed array of letters across a screen, participants asked to report as many as possible and had them either do a partial or whol report

Whole report (Sperling’s experiment)

Participants asked to report whole display, average of 4.5/12 letters

Partial report (Sperling’s experiment)

Participants heard a tone that told them which rows of letters to report, average of 3.34/4 letters

What would happen if one delayed the tone presentation for a fraction of a second after the letters were extinguished in Sperling’s experiment

Performance deteriorates rapidly, average of 1.5/4 letters after 1s delay

What conclusions can be drawn from Sperling’s sensory memory experiement

That sensory memory holds lots of information but decays rapidly

Sensory memory works like a buffer by briefly storing incoming stimuli so they can be processed

Short Term Memory (STM)

Stores small amounts of information for a short duration, information may be new or recalled from LTM

Brown and Peterson STM experiment

Individual asked to read 3 letters, then a number, then cunt backwards by 3. Then asked to recall the 3 letters after a set time

What was found after 3s and 18s in Brown and Peterson’s STM experiment

3s - 80% accuracy

18s - 10% accuracy

What is the reason Brown and Peterson’s STM experiment had people count backwards

Because they wanted to stop rehearsal, STM capacity is only about 15-20s, following that info will decay

Proactive interference (PI)

Occurs when information previously learned interferes with learning new info, memory from previous trials interferes with memory for new letters

Retroactive Interference (RI)

Newly learned information interferes with and impedes the recall of previously learned info

What is the digit span of STM

How many digits a person can remeber, average is 5-9 according to George Miller

Luck and Vogel’s change detection experiment

Showed participants a frame of coloured squares, then a delay, then another frame that was either identical or different from the original, asked them to point out any changes

What did Luck and Vogel’s change detection experiment find

That people did far worse than the 7±2, only about 4 remembered, with correctness decreasing as the number of squares increases, likely due to inability to rehearse

Chunking

Small units can be combined into larger meaningful units that are strongly associated with one another

Ericsson, Chase and Faloon poor college student experiment

Trained SF with average memory ability to use chunking such that after 230 1 hour sessions he could remeber up to 79 digits through chunking

Chase and Simon chess experiment

Looked at beginners and masters on remembering where chess pieces were located, pieces could either be placed randomly or strategically

What was found regarding memory in the random and strategic placement during Chase and Simon’s chess experiment

Random - no difference in memory

Strategic - Master’s did far better, able to remeber most pieces

STM Coding, Mental Coding and Physiological Coding

Coding - way information is represented

Mental Coding - how a stimulus is represented in the mind

Physiological Coding - how a stimulus is represented by firing of neurons

Conrad’s auditory coding experiment

Participants briefly saw target letters and were asked to write them down, looked at the mistakes people made when writing them down

What did Conrad’s auditory coding experiment find

People often made errors when letters sound alike, not when they looked alike, concluded STM for letters is auditory

Della Sala’s visual coding experiment

Presented visual information that was difficult to verbalize, had participant recreate patterns of a partially filled grid, found that STM can also be visual

Wicken’s semantic coding experiment

Had participants listen to 3 words, then count backwards for 15s, then attempt to recall the same 3 words, had two conditions

Wicken’s semantic coding experiment two conditions

Same - Use words from same category on every trial

Switch - Use words from new category on final trial

What is found regarding the “same“ condition in Wicken’s semantic coding experiment

Performance decreases as trials commence due to proactive interference

What is found regarding the “switch“ condition in Wicken’s semantic coding experiment

Performance will decrease on trials 2 and 3, but increase on trial 4, because of the release of proactive interference

What was concluded regarding Wicken’s semantic coding experiment

That semantic coding must play a role in STM as PI only occured when words were in the same category

Working Memory

A limited capacity system for the temporary storage and manipulation of information for complex tasks such as comprehension, learning and reasoning

T or F: working memory is different from STM

F, it is a part of STM

What are the 3 major components of Tripartite Model of Working Memory

1. Phonological loop

2. Visuospatial Sketchpad

3. Central Executive

Phonological loop

A small loop that holds digits in STM, has two subcomponents (phonological store and articulatory rehearsal process), tested using a memory span/serial recall task

Memory span/serial recall task

Series of words, letters, numbers, etc. are presented visually or auditory as participants asked to recal the sequence in the correct order directly after presentation

Memory span

List length that the participant correctly recalls 50% of the time

Phonological similarity effect

Words that are dissimilar are more easily remembered, presents evidence that info is stored based on how it is verbalized

Word length effect

Memory is better for short words than long words, because short words take less time to be rehearsed and are refreshed more often

Articulatory suppression effect

When one continually repeats a word out loud while trying to rehears others they will be unable to rehearse and memory span is reduced (only when reading words), provides evidence for phonological store and articulatory rehearsal process

Why does the articulatory supression effect only reduce the effect on read words

Because visual words must be rehearsed to enter phonological store, whereas auditory words can go right in

Sternberg Search Coglab

One asked to memorize a list of numbers, then given a probe itme and respond if probe item was in the list of numbers, looked at how we retrieve information stored in our phonological loop

What was the IV and the DV in the sternberg search

IV - set size, present/absent

DV - accuracy and reaction time

What were two hypothesis regarding strategies used in the sternberg search 

Serial - items retrieved one at a time

Parallel - items retrieved all at once

What would we assume if the strategy used was serial and the target was absent

RT would increase linearly with the size of the set, could be exhaustive (look through all values) or terminating (stop when value found, shallower slope)

What would we assume if the strategy used was parallel and the target was absent

Assumed that the RT would stay consistent regardless of the set size

What were the findings of the sternberg search

Consistent with the serial hypothesis and we engage in an exhaustive search

Visuospatial sketchpad

Acts as a rehearsal process for visually presented information, tested in Brook’s experiment

What were the two tasks in Brook's experiment

Verbal - memorize sentence and then consider each word to be a noun, respond yes/no

Visuospatial - memorize a block diagram and consider each corner to be an outside corner, point to Y/N on page

How did Brook’s study demonstrate dissociation between the phonological loop and visuospatial sketchpad

Found the verbal response interfered with the verbal tasks more than the visuospatial task and vice versa

Klauer and Zhao experiment

Had two memory tasks and two interfering tasks

Memory:

1. Spatial - memory for dot locations

2. Visual - memory for Chinese ideograph shapes

Interfering:

1. Spatial - movement discrimination

2. Visual - colour discrimination

What were the findings of the Klauer and Zhao experiment

That there was two subcomponents of the visuospatial sketch pad

1. Visual memory store - for colour and form, associated with ventral stream

2. Spatial memory store - for locations and movements, associated with dorsal stream

What 3 things is the central executive involved in

Discovered by Miyake et al study

1. Shifting

2. Updating

3. Inhibition

How was shifting found as part of the central executive in the Miyake et al study

Shift between tasks and mental sets, participants shown 4 quadrents and two of four quadrents were filled with number and letter. If top, report odd/even, if bottom, report consonant/vowel

How was updating found as part of the central executive in the Miyake et al study

Updating of working memory representations, participants shown words and asked to keep track of what the most recent metal, relative and country they had seen

How was inhibition found as part of the central executive in the Miyake et al study

Inhibition of dominant or prepotent response, participant focuses on target, a coloured square will appear followed by an arrow, and one is then asked to determine the direction of the arrow

What type of analysis was done on the Miyake et al study

Confirmatory factor analysis

Structural equation model

Can we explain performance on a complex task using shifting, updating and inhibition

How was the structural equation model evaluated

Wisconsin car sorting task, one is show a number of cards and asked to sort them, only response they get is if their sorting is incorrect or correct. Rule for sorting will swap randomly at some point during the task

What could we conclude from the wisconsin card sorting task

Working memory maintenance is critical to goal directed behaviour

Fluid intelligence

Capacity to think logically and solve problems in novel situations independent of acquired knowledge

What are the two measures of fluid intelligence

Ravens advanced progressive matrices - shown number of diagrams that follow a pattern and asked to fill in missing space based on pattern

Cattell cultural fair intelligence test

How can one measure working memory capacity

Change detection task

T or F: Working memory capacity is not correlated with fluid intelligence

F, It is highly correlated with fluid intelligence