Types of memory

Most psychologists agree that there are three types of memory - sensory register, long-term memory and short-term memory. 

They differ in duration, capacity and coding.

Temporarily stores information from the senses - constantly recieving info from all around us.

Unless we pay attention to it, it disappears quickly - spontaneous decay.

Has a limited capacity, and a very limited duration.

Information coded depending on the sense that has picked it up - e.g. visual, auditry & tactile.

Limited duration and capacity.

Coding is usually acoustic.

Potentially unlimited capacity & info is then theoretically permanent.

Coding usually semantic (the meaning of the information).

Three different types of LTM:

• Episodic - stores info about events that you've experienced. Can contain info about time, place and the emotions you felt. They are declarative - means that they can be subconciously recalled.
• Semantic memory - stores facts and knowledge that we have learned and conciously recall - simply knowledge.
• Procedural memory - stores the knowledge of how to do things e.g. walking. This information can't be conciously recalled.

Sperling (1960) - Investigation of the sensory register.

Method:

• Lab experiment, participants shown a grid of four letters for 0.05 of a second. They then had to immediately recall either the whole grid, or a randomly chosen row indicated by a tone played after the grid was shown.

Results:

• Whole grid recall - only 4-5 letters recalled of average.
• Particular row indicated - around 3 letters on average.

Conclusion:

• Participants didn't know which row was going to be selected, so they were able to recall 3 letters from any row, therefore almost the whole grid was held in their sensory register. They couldn't report the whole grid because the trace faced before they could finish recall. 

Evaluation:

• Lacks ecological validity 
• Variables were controlled so could be replicated

Method:

• Patricipants shown nonsense trigrams and asked to recall them either 3, 6, 9, 12, 15 or 18 seconds.
• During pause they were asked to count backwards in threes from a given number.
• Was an 'interference task'  to prevent them from repeating the letters internally. 

Results:

• After 3 seconds, participants could recall about 80% of the time.
• After 18 secons, recall was about 10%.

Conclusion:

• When rehersal is prevented, very little can stay in STM for longer than about 18 seconds.

Evaluation:

• Lab experiment - results are likely to be reliable
• Lacks ecological validity
• Meaningful or real life tasks may last longer in STM.
• Each participant saw many trigrams so it could have confused them - first trigram was the only realistic trial.

Method:

• 392 people asked to list the names of their ex-classmates - free recall test
• Then shown them and asked to name them - photo recognition test
• OR given names and asked to match them to a photo of the classmate - name recognition test

Results:

• Within 15 years of leaving school, participants could recognise about 90% of faces and names.
• Recall about 60% on free recall for within 15 years of leaving school.
• After 30 years, free recall had declined to about 30%.
• After 48 years, name-recognition was about 80% accurate and photo-recognition was about 40% accurate.

Conclusion:

• The study is evidence of LTM in a natural setting. Recognition is better than recall, so there may be a huge store of info, but it's not easy to access it all.

Evaluation:

• high ecological validity
• hard to control variables - results may not be reliable
• showed higher recall than other studies on LTM - may be because the info was meaningful
• The type of information could be rehearsed 
• Results can't be generalised beyond other types of info held in the LTM.

Method:

• Participats either presented with a string of digits or letters.
• Had to repeat them back in the same order.
• The number of digits or letters increased until the participant failed to recall the string correctly.

Results:

• Majority of the time, participants recalles about 9 digits and about 7 letters.
• Capacity increased with age during childhood.

Conclusions:

• STM has a limited storage capacity of 5-9 items.
• Individual differences were found, such as STM increasing with age, possible due to the use of memory techniques such as chunking.
• Digits may have been easier to recall as there were only 10 digits to remember, compared to 26 numbers.

Evaluation:

• Lacks ecological validity
• Meaningful info may be recalled better, perhaps showing STM to have a greater capacity 
• Previous sequences recalled by the participants might have confused them on future trials.

Reviewed research into the capacity of STM. He found that people can remember about 7 items. 

He said that the capacity of STM is seven, plus or minus two.

Suggested that we use chunking to combine individual letters or numbers into larger more meaningful units.

STM could probably hold about 7 pieces of chunked information, increasing STM's capacity.

In STM, er sometimes try to keep info active by repeating it to ourselves. This means that it generally involves acoustic coding.

In LTM, coding is generally semantic - it's more useful to code word s in terms of their meaning, rather than what they sound or look like.

Method:

• Participants given four sets of words that were either acoustically similar or acoustically dissimilar, semantically similar or semantically dissimilar. 
• Independent group design
• Participants asked to recall the words either immediately or following a 20 minute test.

Results:

• Participants had problems recalling acoustically similar words when recalling the word list immediately (from STM).
• If recalling after an interval (LTM), they had problems with semantically similar words.

Conclusions:

• Patterns of confusion between similar words suggest that LTm is more likely to rely on semantic coding and STM on acoustic coding.

Evaluation: 

• Lacks ecological validity
• Other types of LTM and coding that this experiment doesn't take into account.
• Independent groups design - Wasn't any control over participant variables.