Wikipedia:School and university projects/Psyc3330 w12/Group16

Fuzzy-Trace Theory

Fuzzy-Trace Theory
Fuzzy-Trace Theory (FTT) is a gist driven view of cognitive development. A gist driven view regards the central idea of an entire memory. It does not encode specific details involving the memory, it is only concerned with the main concepts to allow one to "fill-in-the-blanks" of the memory at a later time when the memory is recalled. It is separated from other theories of memory because of its reliance on the interplay between memory and higher cognition. It is the interaction of cognitive triage that exploits memory recall to the fullest.

Background
The Fuzzy-Trace Theory was created based on theoretical advances associated with the Information Processing Theory. The research supporting FTT has been greatly influenced and dominated by the work of C. J Brainerd and V. F. Reyna since 1990. It concerns the encoding of memories and focuses on the retrieval of these memories, especially regarding errors in retrieval. As an interpretation of cognitive development, FTT is often referred as being derived from a cognitive triage. The cognitive triage consists of three different constituent parts: memory strength, episodic activation and output interference. It is the interaction and balance of this cognitive triage that promotes the greatest recall from memory. The following detailed explanation regarding the further understanding of the cognitive triage should be interpreted in terms of memory recall. The recall of memories can be considered as occurring in many different forms, including recalling a list (containing words, concepts and ideas) and also recalling a specific personal event from memory (Episodic memory).

Memory strength
The first component of the cognitive triage is memory strength. FTT describes memory strength as a gist driven form of memory, meaning memories are encoded as a main or central idea describing the entirety of a memory. Gist memories are general descriptions and thus lack specific, fine detailed information regarding the memory. Verbatim memory is just the opposite: memories are recalled exactly as they were encoded, including all fine details and the sources for these memories. Verbatim memories are not typically the brain's "first choice" when encoding memories. These memories tend to be sensitive to retroactive interference, and so risk the inability to be recalled from memory stores. Verbatim memories are also subjected to faster decay caused by the abundance of information contained within the memory and thus, a large amount of cognitive resources consumed during both encoding and retrieval of the memory. The amount of information associated with verbatim memories makes them more difficult to encode than gist memories. This introduces the concept of a "fuzzy-trace" or a gist memory. A fuzzy-trace is simply a trace (memory) of general information contained within the memory and hence are "fuzzy" because they are lacking detailed knowledge. For example, if given a list of words to memorize verbatim memory would create a separate store for each of these words and would also encode associations with the word itself (a picture that came to mind when the word "apple" was read off the list, or maybe a thought regarding the delicious apple the person ate for lunch prior to memorizing the word list) and associations within the environment (the room in which they were sitting was very warm and it was cooler in the room where they ate their apple earlier in the day). This abundance of word and environmentally related information encoded, is very accurately detailed but also very cognitively demanding. Gist related encoding may categorize the entire list of words as relating to one central theme (fruit). During recall, gist information will be much more readily recalled because the individual has an overall concept of word relating to "fruit" than verbatim information where each word was individually stored ("apple" because I ate this for lunch, "banana" because it's my favourite fruit, etc.). Memory strength also depends on factors unrelated to gist or verbatim encoding, including the frequency of the word in spoken memory, the meaningfulness of the word and the concreteness of the word. Using the previous example, those words that are used more frequently in everyday speech are more likely to be remembered (apple is more likely to be remembered when compared to guava), those words that have some meaningful context (banana is my favourite fruit) and the concreteness or abstractness of the word (you can picture an apple, but its harder to picture honesty). It is important to remember, the example of memorizing a list of words can just as easily be replaced with memorizing a list of instructions, or personal episodic memories.

Episodic activation
Episodic activation, the second factor of the cognitive triage associated with FTT, deals with traces that are recalled based on their relationship with a previously recalled word (spreading activation). The fuzzy-trace, similar to a gist memory, is recalled in its broad sense by spreading activation such that the gist memory will activate many different related concepts in memory stores. Episodic activation is dependent upon memory strength; the stronger the memory, the larger the number of connections within an episodic memory network. Recall that the strength of a memory depends on frequency, meaningfulness and concreteness. Those memories that are more frequent (hence are a stronger memory) will activate other related memory stores that are also more frequent, meaningful and/ or concrete. The activation (recall) of one memory will lead to the activation of other related memories within the same associative network. In the above example, utilizing gist memory to encode the list of fruit, upon recall you will know that the list of items you are being asked to recall falls under the category name "fruit". From the name of this category, related nodes will be activated within the episodic memory pathway to allow for the recall of other traces. Such that recalling the category name fruit will lead to those items on the list.

Output interference
The third component of the cognitive triage output interference, incorporates both memory strength and episodic activation to summarize the basic theory behind FTT. Output interference states that memory recall will occur in the order of weak-strong-weak activated words. This order of activation is explained by the amount of interference associated with the recalled items. Weak words will be activated first because they are being recalled at the beginning of the list and thus, have a lower level of interference due to the recency effect (see recency effect, see image serial position curve). After the words relevant to the recency effect are recalled, spreading activation occurs to recall those words with a strong memory trace related to the first words recalled. Finally, since the words with a strong memory trace were recalled, there is once again, less interference. It is proposed that the interfering words are those of strong memory traces and thus, the interference is no longer applicable and those weaker words can be recalled. Using the previous example to explain FTT's output interference model of producing weaker-stronger-weaker activated words, fruit names that occurred at the end of the list (recency effect) will be recalled first by participants after a short delay, due to a lack of output interference associated with these words. Next, words from the fruit list that have a strongly activated link associated with the 'recency effect' words will be recalled next. Finally, those words that are weakly activated on the associative network of nodes will be recalled next due to a lack of interference with the strongly associated words. This weaker-stronger-weaker activation ordering is in conflict with most other activation models which state there is a strong to weak activation sequence.

To summarize, Fuzzy-Trace Theory uses a cognitive triage (memory strength, episodic activation and output interference) to explain its unique model of recall strength; weaker-stronger-weaker activation from memory stores. It is again important to note that although the provided example dealt with memorization of a categorized word list, FTT is in regards to the recall of any memory trace (lists of words, concepts, ideas, as well as the recollection of personal episodic memories). The order of recall is firstly dependent on the memory strength. Memory strength includes encoding method (gist or verbatim), the familiarity of the word measured by the frequency of the word in spoken language, personal meaningfulness, and concrete or abstractness of the trace. Order of recall is subsequently dependent of episodic activation. Episodic activation is influenced by memory strength in terms of relationships between memory stores (nodes). Finally, FTT's order of recollection is based on output interference, which is dependent on both memory strength and episodic activation. The interference is determined by the strength of competing words within the episodic activation model, where words that are strong competitors will interfere with the initial recall of weaker words.

Memory development
The overarching principles of FTT focus on two different levels of memory development. Firstly, there is focus on how information is represented in memory, and secondly, there is focus on how these representations are retrieved and preserved. An important distinction made by FTT is the difference between verbatim traces (remembering something exactly as it was experienced; word for word), and gist traces (retaining the general meaning of things).

In 2004, Brainerd and Reyna identified five principles that contribute to memory development:


 * 1) Parallel storage of verbatim and gist traces
 * 2) Dissociated retrieval of verbatim and gist traces
 * 3) Differential survival rates for verbatim and gist traces
 * 4) Retrieval phenomenology
 * 5) Developmental variability in verbatim and gist memory.

Parallel storage of verbatim and gist traces
Common sense would lead one to believe that verbatim memories are encoded first, and that the gist is then extracted from these memories and transferred to semantic memory in a serial manner. However, a significant body of research suggests that this is not the case, and that instead there is a parallel storage of verbatim and gist traces - that they are in fact coded at the same time. The most intriguing evidence for this comes from experiments where subjects have been shown to acquire semantic information about targets in less than 50ms after onset.

Dissociated retrieval of verbatim and gist traces
There appears to be a dissociation between the retrieval of verbatim and gist memories. This was studied in children by using sentence recognition tasks. Children listened to three-sentence narratives (e.g. The bird is in the cage. The cage is on the table. The bird has green feathers). They were then immediately required to respond to recognition tests that consisted of the target probes (e.g. the cage is on the table), probes that maintaned the meaning of the narratives (e.g., the table is under the bird), and finally probes that went against the meaning of the narratives (e.g., the bird is under the table). It was hypothesized by the researchers that the correct recognition of the target probes would be almost exclusively based upon the retrieval of verbatim traces, and that the false recognition would be based overwhelmingly upon the retrieval of gist traces. However, after the results were computed these two types of performance were statistically independent of each other, suggesting a verbatim-gist retrieval dissociation.

Differential survival rates for verbatim and gist traces
There are differences in how long verbatim and gist traces are stored in memory. Verbatim traces appear to be less likely to survive consolidation than gist traces. An explanation for this could be that verbatim traces are more sensitive than gist traces to proactive and concurrent interference.

Retrieval phenomenology
FTT suggests that the retrieval of verbatim traces creates realistic recollective phenomenology, whereas gist traces usually create a vague familiarity phenomenology.

Developmental variability in verbatim and gist memory
There are distinctive developmental differences in verbatim and gist memory (see Development)

Neural activation for recognition tasks
Results consistent with those predicted by FTT can be tested in many ways, most commonly through recognition tasks. In a recognition task, target items are shown as a list or slideshow format, then after a delay period more items are shown and participants are asked to correctly identify the objects as "new" (not on the previously shown list of items) or as "repeat" (target) items (recognition memory). Pictures were presented to participants with frontal lobe damage (caused by a ruptured aneurysm or stroke) during the recognition task. These participants displayed an abnormal increase in the number of false recognition errors, also referred to as a false alarm (identifying a "distracter" item on the second list as a "target" item on the first list) and were very confident in their (erroneous) answer provided as compared to those adults without frontal lobe damage. The false recognition of distracter items of the same category represents a defective search process. In the processing of items within similar categories, participants are relying on gist memory for the identification of a specific item (recall that verbatim memory encodes for specific details, whereas gist memory encodes only general information) as opposed to specific item matching utilizing verbatim memory. In attempts to lower the number of false alarms, researchers employed Tulving's classic remember vs. know judgements paradigm. In this experiment, the distracter items are from a different category than the target items. This way participants identified if they could consciously recall (remember) the picture by providing detailed information about when and how the information was learned. If the participant could only provide evidence that they had an unconscious feeling of familiarity that they had previously seen the picture, then they provided a 'know' response, indicating that they knew they had seen the photo before but could not be certain it was a target item. With regards to the FTT, this would be equivalent to two distinct gist memory encodings: one gist encoding for the initial target item, and another for the distracter item. Due to the separate occurrences of encoding, the distracter should provide less interference and thus the ability to distinguish the distracter from the target is stronger. A PET scan was used during the recognition task to determine the activation of certain areas in the brain. The PET scan was indicative the right frontal region was the brain was consistently more active during retrieval of a memory trace rather than during the encoding of a memory trace.

Children's neural activation on false recognition tasks
In comparison, the same results have been identified for normal developing children. The same as with patients with frontal lobe damage, children who have a higher frontal lobe activation during a false recognition task showed lower incidences of false recognition. Younger, normal developing children are more likely than older children, who are more likely than healthy adults to make false recognition errors. Young children have underdeveloped frontal lobes. This under activity of the frontal lobes area is comparable to an adult with frontal lobe damage who would display a similar inactivity in the frontal lobe. Frontal lobe development peaks in grey matter and white matter around the age of 10 and continues to develop into young adulthood. Activation is greater in the fully developed adult frontal lobe than it is in the immature frontal lobe of a child. Accurate recollection rejection is significantly related to frontal lobe activation, where activation indicates a lower probability of a false recognition response. Recollection rejection is just the opposite of false recognition. It is the ability to retrieve item-specific information from memory (using verbatim memory) when gist memory was used to encode semantically related target items (those of the same or related categories).

Age-related differences for neural activation
Age is a relevant factor in the activation of specific brain regions. Where younger children tend to show a decreased activation of the frontal lobe during the recall of memory related tests, older adults show reduced activation of both the medial temporal gyrus (see notable gyri) and the medial parietal cortex (regions implemented in memory formation). This means that there are cognitive memory differences between the young, the middle-aged, and the old based on the activation of different brain structures. Identifying active cognitive structures involved in memory process can also be used as a predictor of the type of memory encoding the individual is using: gist or verbatim. Testing the neural activity during encoding can be accomplished using an event-related fMRI. Correctly identifying an item as the same item shown in the first trial period is associated with activation of the right fusiform cortex during encoding. False memories (identifying a distractor item as a target item) and partial memories (identifying an identical target item as only similar to what they remember previously seeing) are associated with the activation of the left fusiform cortex during encoding. The incorrect (false memory) or not completely correct (partial memory) response can be attributed to the difference in activation. The left fusiform cortex processing is concerned with general object encoding, such as naming an object whereas the right fusiform gyrus processes information associated with feature specific encoding, hence the correct identification of a previously seen item.

Childhood
The fuzzy trace theory explains why children are able to infer relationships between two objects from a known relationship with a third object, known as transitive reasoning. According to Piaget's theory of cognitive development, these kinds of relationships can only be drawn once the child has mature logic and reasoning skills. But it has been shown that these skills can be acquired by children before logic and reasoning skills are intact, by using their verbatim and gist memory traces.

Verbatim memory development
Unsurprisingly, as an individual transitions from infancy to childhood, verbatim memories show significant improvements (see infantile amnesia). The most noticeable improvements come between the preschool and early elementary-school years. Evidence of these developmental shifts in verbatim memory comes directly from studies using nonsense words, where it has been shown that older children are more adept at distinguishing between presented nonsense words and their distractors than the younger children in the sample (ability increases with age from age 7-11).

Initial improvements in verbatim memory during preschool and early elementary-school years are the most dramatic. This was discovered in studies where children must distinguish between a learned target and a new one which is very similar but has distinct differences, such as a sentence that has one different word in it.

Gist memory development
Development of gist memories tends to be more complex. Preschoolers can categorize everyday objects in accordance to their meanings, such as placing pictures of shirts in the ‘clothing’ category and hamburgers in the ‘food’ category. However they have trouble linking unfamiliar objects or pictures to familiar meanings. For example, if you showed you them a picture of an animal they’ve never seen, they may have difficulty placing it into the ‘animal’ category.

Adulthood, aging, and Alzheimer’s disease
Direct access to verbatim memory traces tends to decline with old age but, generally, gist memory and reconstructive processes accommodate for the loss so that, in adulthood, memory recall doesn’t decrease as much. In contrast, patients with Alzheimer’s disease experience interference in the reconstruction aspect of memory link to the gist traces and so memory recall is drastically reduced and can only rely on verbatim memories which tend have limited storage for about 4 words in memory.

False memories
Testing for false memories is often measured using the Deese-Roediger-McDermott paradigm. This paradigm presents a list of words to the participants (ex. table, sit, legs) which are based on a semantically related, but non-presented word (ex. chair). False memories are exhibited when the non-presented word is recognized as being on the initial list of semantically related words. When comparing false memory incidences between children and adult, children show a lower incidence of false memory than do adults when the task involves divided attention. There also appears to be age-related differences regarding the prevalence of false memories in 11- compared to 7-year olds, where 11-year olds show a higher incidence of false memory errors. When tested, both children and adults show increased false memory for neutral rather than negative non-presented traces. That is, non-presented, but semantically related words that have a negative emotionality are less susceptible to be recalled as a false memory. Similarly, the emotional state of the participant also influences the rate of false memories. Negative mood states show reduced false memory errors. Those in a negative mood are more likely to use verbatim (item-specific) processing rather than gist processing. The utilization of verbatim processing is more susceptible to decay, but is more detailed and thus less susceptible to incidences of false memories. The type of mood (negative or neutral) is temporally crucial surrounding the presentation of the semantically related initial list items. If the negative mood is apparent before learning the list of semantically related items, then both encoding and retrieval are affected. Retrieval is influenced in this case based on the cognitive mechanisms required to retrieve the negatively encoded information. Whereas if the negative mood is only present at the time of testing, then only retrieval is influenced. False memories are less apparent when the negative mood is present before learning occurs and show no effect when only present at retrieval. From this, memory is dependent on mood state at the time of encoding a trace.

Implications with Fuzzy-Trace Theory
FTT states that traces are encoded using verbatim or gist. When there is not enough item-specific (verbatim) information, a fuzzy-trace must be used and reconstructed as best as possible. When an error arises due to a lack of verbatim information, then a false error has likely occurred. Accurate identification of semantically related presented and non-presented words are based on verbatim encoding. With false memories, verbatim traces become weaker over time and reliance is more weighted toward gist traces relevant to the memory. Although gist memories are more durable, they lack the detailed information of a verbatim memory and so memories are forced to be reconstructed consequentially giving rise to more false memories. FTT suggests children are poorer at gist encoding because they lack the ability to identify and overall theme. Therefore, there is a lesser emphasis on gist encoding for children compared to adults. This also provides an explanation of frequency of false memories based on developmental differences between different age cohorts. As children mature developmentally, they are more skilled at encoding gist traces but also become susceptible to false memory errors. The nature of memory encoding changes as a child developmentally matures, this has a great influence on cognitive skills that bring about life decisions.

Decision making
The concept of the FTT explains that individuals make decisions based on the gist of information alternatively to using word for word verbatim. This theory explains why individuals make errors in decision making. Gist is based on one’s interpretation and value of the information and is a form of complex reasoning,  it is the “fuzzy traces” in FTT, where as verbatim is the literal translation of the information presented. FTT is based on a dual- process theory, this is the idea that both verbatim and gist based processing are used together when encoding information. Here gist and verbatim are the two processes which counter one another. The use of verbatim in decision making can be futile and ineffective in altering an individuals’ judgement, as most individuals are dependent on using the gist of the information to make a decision. Decisions are subject to gist based thinking because they are made based on emotional meaning, personal values, and or personal interpretation of the information. When individuals encode information, they encode verbatim and gist of the information equally. They then summarize the verbatim numbers and rate the gist of the information qualitatively in the order and value to them. Individuals prefer to use the least specific interpretation of the information to make a final decision. Gist based thinking causes framing effects to arise because numbers are translated semantically in situations where the focus is not on precise information but on the overall idea of the information. For example; good versus bad, or more versus less. Through gained experience individuals depend on gist based thinking as they age. This is known as fuzzy-processing preference.

Medical decision making
Fuzzy trace theory (FTT) in medical decision making explains how healthcare professionals and patients interpret, communicate, and relate information. The theory has found that individuals encode medical information equally using verbatim and gist based processing (dual processing theory) however, they depend on gist based representations to make formal decisions. Gist based thinking is dependent on how the information is emotionally significant to the individual. Once the patient or health professional is given medical information the patient will take into account their own personal values and knowledge. Understanding of medical health information among healthcare providers and patients is inadequately represented by gist representation, leaving both healthcare professionals and patients alike unable to fully appreciate possible prevention of diseases and treatment. Fuzzy trace theory clarifies why giving health professionals and patients verbatim information (numerical or otherwise) is ineffective. Doctors and health professional may have correct memory of the medical information (verbatim), however they will use qualitative gist memory to communicate facts. Researchers in this field are finding new effective ways to present medical information. Having people pay better attention to verbatim facts will not help them better understand medical information or to help better convey it. There are three concerns in FTT concerning health information and communications.

1) Verbatim information is in ineffective in conveying health information and risks,

2) A stronger connection between health information and operation is needed, and lastly

3) There is a need to use theory based interventions to help doctors, healthcare professionals and patients with communicated accurate information.

In medical decision making, identifying numerical values regarding health risks can be a question of life and death. The problem with using FTT is that relying on gist based thinking may compromise an individual’s health as gist is not adequate enough to make judgements on medical health. People use many degrees of gist information, however, they may use only one interpretation at a given time for decision making. These degrees of gist are thought to be parallel to the levels of measurement which include nomial, ordinal, interval and ratio scales. For information to be accurate in the minds of people, the information needs to be of interest it needs to apply to gist- based intuition.

Risk perception
Risk perception in FTT is the effect of making decisions based on an individuals tolerance and acuity of risk Risk taking is often seen in youth and young adults and is described by the decision to take a chance in a given situation.

According to the FTT those with the most knowledge in decision making on a particular subject should be able to better discriminate among risky categories. The more well-informed and experienced a person becomes in a task the more likely that their thought processing becomes more gist based. This leads to risky decision making as their perception of risk becomes diluted. This is referred to as risk discrimination. With time and experience, one’s knowledge about a given field becomes more reliant in intuitive gist based thinking. For example, a physician, who has a great deal of knowledge about heart health and knows much about heart attacks, is treating a patient with chest pain. If the patient is not in imminent danger of experiencing a heart attack they will more likely to dismiss the patient. The physicians perception of the situation is not risky. Risk perception varies from individual to individual in that one’s risk tolerance maybe higher or lower that the next. For example, doctors who encountered loss of patients more frequently ordered more laboratory tests  than other doctors who experiences significantly fewer losses