![[DaveInman]](../../../../../www.sbu.ac.uk/~dave/images/daveicon.gif)
Dave Inman ![[SCISM Home]](../../../../graphics/scismhome.gif){kind=small}
SCISM ![[SBU Home]](../../../../../www.sbu.ac.uk/graphics/homebutton.gif)
SBU Home
This spreading to certain parts of such networks is not entirely under an individual's control. This is known as associative priming.
Experiments observing change in LTP have shown that recognition time is an inverse reflection of the growth of strength that is happening internally. As the strength of the record increases, the performance measures get better.
1. Preview
2. Questions
3. Read
4. Reflect
5. Recite
6. Review
Memory is the subject concerning the concepts that the brain employs to store and retrieve information, or the process referred to as remembering. This process enables us to use our memory's while conducting other tasks where the information that we have stored would be useful and necessary.
Memory can be split up into various subjects, and each subject has underlying concepts which aim to model the way in which our brain stores these memory's. Also the way in which they are recalled. The concepts which we are looking at here are where and how we store this information. In particular we are to look at the first accepted theories of memory and also the current thoughts. With work and research of scientists and psychologists such as Ebbinghaus in the very beginning to modern names such as, Miller, Barnes, Bobrow and Bower. From short term memory through to working memory, and with a look at activation and levels of processing, a picture will be formed of how we use our memories.
1. HUMAN MEMORY
The Ebbinghaus experiment (1885) was the first real experiment into memory, in particular the remembering of trigrams ( i.e. DAX, LOC). These all had no meaning but were all distinct monosyllable words. He measured the retention that he himself had of the words by counting the number of relearns required to achieve perfect recall and recite them correctly. His results can be shown as a graph and shows that retention decreases over time, but the rate of forgetting slows down. He also looked at the affects of overlearning and found that through overlearning a higher score can reached on his retention test.
RISE AND FALL OF SHORT MEMORY
The idea of Short Term Memory was introduced in the '60 's , and it immediately could explain things that were difficult before. There were so many people working at Short Term Memory there was a lot of work done, although the most thorough work was done by Atkinson and Shiffrin, and is still regarded as the best work on Short Term Memory, even if it has now been overtaken by new areas of research.
Short Term Memory was supposed to be an area where information was stored until it had been rehearsed enough to be placed into Long term memory. Short Term Memory was thought to have limited space, and its capacity was defined as its memory span. The memory span is the number of elements that could be repeated straight back after learning. It was found the normal memory span is 7, which is similar to the Magic Number 7+-2 proposed by George Miller along with the idea of chunking; the splitting up of information into understandable units). It was thought though that if a memory was not rehearsed enough then it would be lost forever to make way for new information, in the limited space of Short Term Memory.
Shepard and Teghtsoonian worked on this and with experimentation found that 60% of inputs stayed and were rehearsed enough to make it to Long Term Memory. The other 30% was lost rapidly from Short Term Memory.
This rapid forgetting led researchers into thinking that there was more to the memory system than the simple view of Ebbinghaus. Short term memory it seemed supported this finding. There was one question that was raised and needed to be answered had Short term memory any long term future.
Data still strongly supported Short term memory, such as the effects of rehearsal. The more anything was rehearsed the more likely it was to be remembered or stored in Long term memory, and it needed to be kept somewhere first, hence Short term memory.
Another theory was introduced to counter this. It was not the amount of rehearsal that was important but the quality of depth of processing involved in the rehearsal. Experiments showed that when no effort was placed in rehearsal then the memory performance was poor. So it was added that there may be no memory before Long term memory but the way in which the information is rehearsed lays stronger or weaker paths to the information. Poorly rehearsed information has weak paths and is therefore harder to find.
2. REHEARSAL AND WORKING MEMORY
Although the theory of short term memory seems incorrect there is some limit to the information that can be processed at any time. Opposed to the Short term memory theory, Baddeley said that the memory span was controlled by the speed of the rehearsal. This was called the articulatory loop. His experiments proposed a duration of 1.5 seconds of remembering. The quicker we process therefore the more we remember in this fixed duration. Some evidence also points to a link with speech and the way in which information is remembered. With like sounding words being confused with each other.
Baddeley also promoted another rehearsal mechanism called the visuospatial sketch pad. This along with the articulatory loop maintain the information in the working memory. These are controlled by a central executive. The difference between this and Short term memory is that the information does not have to spend time in either the visuospatial sketch pad or the atriculatory loop before entry to long term memory.
3. FRONTAL CORTEX AND WORKING MEMORY
It is known that the frontal cortex has an important role in working memory tasks. A small region has been found in primates to have more activity when working memory tasks are being performed. Also in humans there is evidence that there is an increase of blood flow to this area when it is thought the working memory is being used to perform a task.
3.1 ACTIVATION AND LONG-TERM MEMORY
This is the memory where the information is stored and can be retrieved over long periods of time. This memory is infinite. Some people believe that information is never lost from it but becomes less and less accessible.
A number of theories, such as the ACT theory, have appeared which assume that different pieces of information in long-term memory can vary from moment to moment in terms of how available they are. According to this theory information is encoded in all or none manner into cognitive units and the strengths increases with practice and decays with delay. Activation determines both the probability of access to memory and the rate of access. The underlying idea is that after we have used a piece of information, it will be highly unavailable but it will rapidly become unavailable if not used or rehearsed.
Two factors determine the level of activation of a memory. One is how recently we have used the memory. The other is how much we have practised the memory. Anderson's experiments illustrated how speed of retrieval varies with both recency and frequency of practice. He concludes that the there is little effect of the amount of study for items tested at a short interval. Whereas when the interval is longer the amount of study is a determining factor.
3.2 Spreading Activation
Spreading activation is the proposal that activation spreads along the paths of a propositional network, an interconnected network which consists of all cognitive units, to activate associated memories. Activation spreads from present material to activate associated material. It is the notion that memory traces becoming active when associated concepts are presented.
This activation is called associative priming. It is not entirely under an individual's control, and sometimes it primes knowledge from a part of this network unconsciously.
Several other experiments have been performed which produced several output. Meyer and Schvaneveldt (1971) concluded that we can read material that has a strong associative coherence more rapidly than incoherent material in which words are unrelated. Ratcliff and McKoon (1981) studied the effects of automatic spread of activation and the asymptotic levels that priming can reach. It can be inferred that this decrease in latency can be taken to reflect the rate at which activation spreads through the network.
We can see that activation will spread form presented material to activate associated material. The more the activation spreads to material, the more rapidly it can be retrieved. It turns out that the amount of activation spread to a memory depends on the strength of that memory.
4.1 PRACTICE AND STRENGTH
Two separate quantities can be associated with a memory trace: activation and strength. The activation level of a memory trace can undergo rapid shifts, but traces also can rapidly lose activation. Strength changes quantity more gradually. Each time we a use a memory trace, it will increase a little in its strength. The strength of a trace determines in part how active it can get and hence how accessible it will be. Strength of a trace can be gradually increased by repeated practice.
Memory performance improves as a power function of practice (T = 1.40 P-0.24 ). It is called power function because the amount of practice P is being raised to a power. This power relationship between performance and amount of practise is an omnipresent phenomenon in learning. This has been referred as the power law of learning. Blackburn (1936) came to the conclusion after practising addition problems for 10,000 trials, that as memory traces become stronger, they can receive more activation and so can be more rapidly retrieved.
4.2 Long-Term Potentiation and the Power Law
Long-Term Potentiation is one kind of neural learning which occurs in the hippocampus and cortical areas, and appears to follow a power law. It is a form of neural learning that seems related to behavioural measures of learning. Barnes (1979) carried out certain experiments, measuring the percent increase in excitatory postsynaptic potential (EPSP), which showed that neural activation changes with practice just as do behavioural measures. The assumption is that a performance measure like recognition time is an inverse reflection of the growth of strength that is happening internally. As the strength of the record increases the performance gets better (which means shorter times and fewer errors).
5. LEVELS OF PROCESSING
Mere study of material will not lead to better recall. It is important how one processes the material while studying it. More meaningful processing of material results in better memories.
There is evidence that shallow study results in little memory improvement. Nelson (1979) had subjects read paired associates that were either semantic associates (e.g. tulip-flower) or rhymes (e.g. tower-flower). Better memory (81 percent) was obtained for the semantic associates than the rhymes (70 percent). Presumably, subjects were more inclined to process the semantic associate meaningfully.
In addition to more meaningful processing, there is evidence that more elaborative processing results in better memory. Elaborative processing involves embellishing a to-be-remembered item with additional information.
Bobrow and Bower's experiments revealed that the critical factor is not whether the subject or the experimenter generates the elaborations. Rather, it is whether the elaborations are such that they constrain the to-be-recalled material.
- Good memory for material results when it is processed
more elaborately.
5.2 Meaningful Versus Nonmeaningful Elaborations
Some psychologists have argued that the entire levels-of-processing literature is to be accounted for in terms of memory for the processes involved in the original study of the material. Kolers has found that subjects can remember more about upside-down sentences. He argues that the extra processing involved in reading the typography of upside-down sentences provides the basis for the improved memory. It is not a case of more meaningful processing but one rather of more extensive processing.
- More elaborate processing will result in better memory even if that processing is not focused on the meaning of the material.
5.3 Incidental Versus Intentional Learning
Whether a person intends to learn or not really does not matter. What matters is how the person processes the material during its presentation. If the individual engages in identical mental activities when not intending to learn as when intending, he or she gets identical memory performance in both conditions. People typically show better memory when they intend to learn because they are likely to engage in activities more conducive to good memory, such as rehearsal and elaborative processing.
--Level of processing, and not whether one intends to
learn, determines the amount remembered.
5.4 Elaborative Processing and Text Material
Frase (1975) has found evidence for the benefit of elaborative processing with text material. He compared two groups of subjects on their memory for a text. The "advance organiser" group were given topics to think about before reading the text and a control group simply studied the text without advance topics. The advance organiser group did only slightly worse than the control group on topics for which they had not been given advance warning, but much better on topics for which they had been given advance warning.
A technique which promotes elaborative processing of text is the PQ4R technique. It derives its name from the six phases it advocates for studying a chapter in a textbook:
a) Preview. Survey the chapter to determine the general topics being discussed. Identify the sections to be read as units. Apply the next four steps to each section.
b) Questions. Make up questions about the section. Often, simply transforming section headings results in adequate questions.
c) Read. Read the section carefully, trying to answer the questions you have made up about it.
d) Reflect. Reflect on the text as you are reading it, trying to understand it, to think of examples, and to relate the material to prior knowledge.
e) Recite. After finishing a section, try to recall the information contained in it. Try answering the questions you made up for the section. If you cannot recall enough, reread the portions you had trouble remembering.
f) Review. After you have finished the chapter, go through it mentally, recalling its main points. Again try answering the questions you made up.
The central feature of the PQ4R technique is the question-generating and question-answering characteristics. There is reason to suspect that the most important aspect of these features is that they encourage deeper or more elaborative processing of the text material.
Reviewing the text with the questions in mind is another important component of the PQ4R technique. Experiments seem to show that reviewing the text with questions in mind may be more generally beneficial.
- Study techniques involving question generation
and answering lead to better memory for text material.
6 ENCODING VERSUS RETRIEVAL
One cannot talk about any memory experiment without considering what is involved in getting that information out. Many of the above issues considered are considerably complicated by retrieval issues, especially elaborative processing. There are important interactions between how a memory is processed at study and how it is processed at test. We have discussed the effects of factors like practice without discussing the activation-based retrieval processes which are facilitated by these processes.
CONCLUSION
Anderson has considered and presented research findings that help in understanding how information is encoded and stored into the human memory system.
The theory of short term memory is being succeeded with the belief that what is significant is that we process information in a way that is conducive to establishing a long-term memory trace. It seems that information can go directly from sensory stores to long-term memory. Clearly rehearsal aids memory encoding and storage but Baddeley has expanded on this by proposing the existence of a visuospatial sketchpad and phonological loop as mechanisms for rehearsing material. Experiments have revealed that different areas of the frontal cortex appear to be
responsible for maintaining different types of information in working memory.
The speed and probability of accessing a memory are determined by its level of activation, which in turn is determined by how frequently and how recently we have put the memory to use. Activation spreads through a network from presented items to activate associated memories but note that the spread-of-activation process is not entirely under an individual's control.
As a memory is practised, it is strengthened according to a power function known as the "power law of learning". Long term potentiation (LTP) is a kind of neural learning that occurs in the hippocampus and cortical areas. There is evidenc
e that LTP underlies the power law of learning. There appears to be a diminishing increase in LTP with amount of practice.
There is evidence to support the theory that memory for information is improved if the information is processed at deeper levels of analysis. Shallow or mere study of material will not lead to better recall. Elaborative processing, where an item is embellished with additional information, has been shown to lead to better memory. The level of processing, and not whether one intends to learn, determines the amount remembered. Elaborative processing is greatly beneficial forthe study of text material. Many popular techniques such as PQ4R have arisen for studying text material and many involve question generation and answering.
No doubt research paradigms will continue to emerge on human memory to assist in the search for that elusive "better memory". Let us not forget that the encoding of information into memory must be considered together with issues of retrieval and indeed human behaviour in general.
By understanding the "how and why" of human memory we hope we will be able to design and develop computer systems that can both interact more seamlessly with the user and work more intelligently. We found the preparation of this stack extremely useful both in professional and personal terms - exams? no problem!
Dave Inman SCISM SBU HomeLast Change Wed, Jan 29, 1997