Memory (Part 1) Sensory Memory and Short-term Memory Flashcards Preview

Mind, Brain & Behaviour 1 > Memory (Part 1) Sensory Memory and Short-term Memory > Flashcards

Flashcards in Memory (Part 1) Sensory Memory and Short-term Memory Deck (16):

What is Memory?

Memory is a set of systems for storing and retrieving information that is acquired through our senses

Memory contains mental representations of information that we have experienced, imagined and learned

“Learning is the process of acquiring new information, while memory refers to the persistence of learning in a state that can be revealed at a later time.” (Squire, 1987)


Memory in terms of three processes

1. Encoding: converting information into a comprehensible form (getting information into memory)

2. Storage: retaining information

3. Retrieval: gaining access to information stored in memory


Memory in terms of four processes

1. Acquisition: registers inputs in temporary sensory storage

2. consolidation: creates a stronger representation over time

3. Storage: maintenance of a permanent record to be used at later time

4. Retrieval: utilisation of stored information to create a conscious representation, or to execute a learned behaviour


The multi-store model of memory
Atkinson and Shiffrin (1968)

Three separate stores:

Sensory memory: duration: ultra-short, capacity: large

Short-term memory: duration: seconds to minutes, capacity: limited

Long-term memory: duration: years, capacity: unlimited


Sensory Memory (George Sperling 1960)

He named the sensory memory store “Iconic Memory”

This store is characterised by a large capacity, but rapid forgetting, as demonstrated in his famous experiment

He was interested in how much information can be held in sensory memory

For this experiment, he used the “full vs. partial report” method


George Sperling Memory experiment results

For a very brief time all information must be held in memory

The partial report technique demonstrated that the capacity of iconic memory is quite large – most people can recall 3-4 items (the entire row)

By varying the onset of the cue it has been shown that after ~500ms, the memory trace decayed and items can’t be recalled


Echoic Sensory Memory

There is a similar modality specific memory store for sounds

Moray, Bates, and Barnett (1965) simultaneously presented sequences of consonants from different locations

Echoic memory is similar to iconic memory in that there is more information accessible than can be recalled

Echoic memory is different in that the trace lasts longer, up to 3-4 seconds


What is the function of sensory memory?

Sensory memory prolongs briefly presented stimuli to increase the chance that they will be present long enough to be attended

Attended information can then be transferred into the short-term memory (STM) store

Information in STM can be held for several seconds, integrated with information from other sources, and then transferred into long-term memory (LTM)


Short- term Memory (STM)

capacity limited (5+ or -2 pieces of information), duration seconds to minutes

Information is kept active via maintenance rehearsal (repeating information without ascribing meaning to it)

non-rehearsed information is lost rapidly via decay


Peterson & Peterson (1959) Experiment with consonants and counting on Short-term memory

Participants had to remember 3 consonants (e.g. CHM)

They prevented prevent rehearsal by counting backwards in 3’s from a given number until stop (e.g. 506 – 503 – 501... )

They also systematically manipulated the retention interval from 3 to18s

If not rehearsed, the items are forgotten after some seconds

The consonants are displaced from memory


Short-term memory Chunking (manipulates capacity of STM)

George Miller (1956) conducted a study on STM Capacity:

Participants had to perform a digit-span task: This tests the ability to recall a sequence of random digits in the order they were presented (serial recall)

He also systematically manipulated retention interval from 3 to 18s

He found that participants were able to recall more items when they could group information into meaningful units (“chunks”)

Chunking can increase the amount of information contained in STM

Participants can store seven (5+/- two) units of information


The Serial Position Effect (Primacy and Recency Effect)

Rehearsal mediates the transfer between STM and LTM

Primacy Effect: better recall for items at the beginning of a study list as items have been rehearsed and transferred to LTM

Recency Effect: better recall for items at the end of a study list as items presumed to be retained and still rehearsed in STM


Craik & Lockhart (1972) Levels of processing

Challenged the multi-store model and the role of maintenance rehearsal in transferring information to LTM

They proposed that processing that involves more interaction with previously stored knowledge increases the chance that the information will be preserved

Their key concept is Levels of Processing: “deep” vs. “shallow” processing


Deep processing v.s Shallow processing

deep processing:
- involves a reorganisation of the information
- formation of association
- creation of (interactive) mental images

Deep processing results in richer ‘network of associations’ and increases the likelihood of recall

shallow processing:
- simple repetition of material (rehearsal)
- processing of surface features (i.e. upper or lower case letters instead of meaning)


Levels of processing: Experiment by Craik and Tulving (1975)

Participants were given a list of words with three different tasks (they did not expect the memory task later on):

Task 1: structural processing (shallow) Is the word written in capital letters?

Task 2: phonological processing (medium) Does the word rhyme with another word X?

Task 3: semantic processing (deep) Does the word fit in sentence Y?

Then, after the “experiment”, they were given an unexpected memory task...
The questions was whether the first task led to different performance in the memory task


Levels of processing: Experiment by Craik and Tulving (1975) Results

The results show that memory performance (measured in recognition rates and response latency) different between tasks

Deeper levels of processing in the initial task led to better memory performance later