User:AngeAri/Chunking (psychology)

Chunking and Memory in Chess Revisited

Previous research has shown that chunking is an effective tool for enhancing memory capacity due to the nature of grouping individual pieces into larger, more meaningful groups that are easier to remember. Chunking is a popular tool for people who play chess, specifically a master. Chase and Simon (1973a) discovered that the skill levels of chess players are attributed to long-term memory storage and the ability to copy and recollect thousands of chunks. The process helps acquire knowledge at a faster pace. Since it is an excellent tool for enhancing memory, a chess player who utilizes chunking has a higher chance of success. According to Chase and Simon, while re-examining (1973b), an expert chess master is able to access information in long-term memory storage quickly due to the ability to recall chunks. Chunks stored in long-term memory are related to the decision of the movement of board pieces due to obvious patterns.

Chunking Models for Education

Many years of research has concluded that chunking is a reliable process for gaining knowledge and organization of information. Chunking provides explanation to the behavior of experts, such as a teacher. A teacher can utilize chunking in their classroom as a way to teach the curriculum. Anderson (2000) proposed that teachers can use chunking as a method to segment the curriculum into natural components. A student learns better when focusing on key features of material, so it is important to create the segments to highlight the important information. By understanding the process of how an expert is formed, it is possible to find general mechanisms for learning that can be implemented into classrooms.

Article Draft

Chunking as the learning of long-term memory structures

This usage derives from Miller's (1956) idea of chunking as grouping, but the emphasis is now on long-term memory rather than only on short-term memory. A chunk can then be defined as "a collection of elements having strong associations with one another, but weak associations with elements within other chunks". The emphasis of chunking on long-term memory is supported by the idea that chunking only exists in long-term memory, but it assists with redintegration, which is involved in the recall of information in short-term memory. It may be easier to recall information in short-term memory if the information has been represented through chunking in long-term memory. Norris and Kalm (2021) argued that “redintegration can be achieved by treating recall from memory as a process of Bayesian inference whereby representations of chunks in LTM (long-term memory) provide the priors that can be used to interpret a degraded representation in STM (short-term memory)”. In Bayesian inference, priors refer to the initial beliefs regarding the relative frequency of an event occurring instead of other plausible events occurring. When one who holds the initial beliefs receives more information, one will determine the likelihood of each of the plausible events that could happen and thus predict the specific event that will occur. Chunks in long-term memory are involved in forming the priors, and they assist with determining the likelihood and prediction of the recall of information in short-term memory. For example, if an acronym and its full meaning already exist in long-term memory, the recall of information regarding that acronym will be easier in short-term memory.

Chase and Simon in 1973 and later Gobet, Retschitzki, and de Voogt in 2004 showed that chunking could explain several phenomena linked to expertise in chess. Following a brief exposure to pieces on a chessboard, skilled chess players were able to encode and recall much larger chunks than novice chess players. However, this effect is mediated by specific knowledge of the rules of chess; when pieces were distributed randomly (including scenarios that were not common or allowed in real games), the difference in chunk size between skilled and novice chess players was significantly reduced. Several successful computational models of learning and expertise have been developed using this idea, such as EPAM (Elementary Perceiver and Memorizer) and CHREST (Chunk Hierarchy and Retrieval Structures). Chunking may be demonstrated in the acquisition of a memory skill, which was demonstrated by S. F., an undergraduate student with average memory and intelligence, who increased his digit span from seven to almost 80 within 20 months or after at least 230 hours. S. F. was able to improve his digit span partly through mnemonic associations, which is a form of chunking. S. F. associated digits, which were unfamiliar information to him, with running times, ages, and dates, which were familiar information to him. Ericsson et al. (1980) initially hypothesized that S. F. increased digit span was due to an increase in his short-term memory capacity. However, they rejected this hypothesis when they found that his short-memory capacity was always the same, considering that he “chunked” only three to four digits at once. Furthermore, he never rehearsed more than six digits at once nor rehearsed more than four groups in a supergroup. Lastly, if his short-term memory capacity increased, then he would have shown a greater capacity for the alphabets; he did not. Based on these contradictions, Ericsson et al. (1980) later concluded that S. F. was able to increase his digit span due to “the use of mnemonic associations in long-term memory,” which further supports that chunking may exist in short-term memory rather than long-term memory.

Chunking has also been used with models of language acquisition. The use of chunk-based learning in language has been shown to be helpful. Understanding a group of basic words and then giving different categories of associated words to build on comprehension has shown to be an effective way to teach reading and language to children. Research studies have found that adults and infants were able to parse the words of a made-up language when they were exposed to a continuous auditory sequence of words arranged in random order. One of the explanations was that they may parse the words using small chunks that correspond to the made-up language. Subsequent studies have supported that when learning involves statistical probabilities (e.g., transitional probabilities in language), it may be better explained via chunking models. Franco and Destrebecqz (2012) further studied chunking in language acquisition and found that the presentation of a temporal cue was associated with a reliable prediction of the chunking model regarding learning, but the absence of the cue was associated with increased sensitivity to the strength of transitional probabilities. Their findings suggest that the chunking model can only explain certain aspects of learning, specifically language acquisition.

Chunking in motor learning

Chunking is a method of learning that can be applied in a number of contexts and is not limited to learning verbal material. Information is grouped into pieces called chunks which can be processed and stored more effectively than individual items. Chunking has been demonstrated to be critical in the development of motor learning. Chunking can also make it easier for students to acquire and apply skills in many settings by assisting them to recognize the underlying structures of a task at hand . Karl Lashley, in his classic paper on serial order, argued that the sequential responses that appear to be organized in a linear and flat fashion concealed an underlying hierarchical structure. This was then demonstrated in motor control by Rosenbaum et al. in 1983. Thus sequences can consist of sub-sequences and these can, in turn, consist of sub-sub-sequences. Hierarchical representations of sequences have an advantage over linear representations: They combine efficient local action at low hierarchical levels while maintaining the guidance of an overall structure. While the representation of a linear sequence is simple from a storage point of view, there can be potential problems during retrieval. For instance, if there is a break in the sequence chain, subsequent elements will become inaccessible. On the other hand, a hierarchical representation would have multiple levels of representation. A break in the link between lower-level nodes does not render any part of the sequence inaccessible, since the control nodes (chunk nodes) at the higher level would still be able to facilitate access to the lower-level nodes.

Schematic of a hierarchical sequential structure with three levels. The lowest level could be a linear representation, while intermediate levels denote chunk nodes. The highest level is the entire sequence.

Chunks in motor learning are identified by pauses between successive actions in Terrace (2001). It is also suggested that during the sequence performance stage (after learning), participants download list items as chunks during pauses. He also argued for an operational definition of chunks suggesting a distinction between the notions of input and output chunks from the ideas of short-term and long-term memory. Input chunks reflect the limitation of working memory during the encoding of new information (how new information is stored in long-term memory), and how it is retrieved during subsequent recall. Output chunks reflect the organization of over-learned motor programs that are generated on-line in working memory. Sakai et al. (2003) showed that participants spontaneously organize a sequence into a number of chunks across a few sets and that these chunks were distinct among participants tested on the same sequence. They also demonstrated that the performance of a shuffled sequence was poorer when the chunk patterns were disrupted than when the chunk patterns were preserved. Chunking patterns also seem to depend on the effectors used.

Perlman found in his series of experiments that tasks that are larger in size and broken down into smaller sections had faster respondents than the task as a large whole. The study suggests that chunking a larger task into a smaller more manageable task can produce a better outcome. The research also found that completing the task in a coherent order rather than swapping from one task to another can also produce a better outcome.

Chunking in Infants

As we know chunking is used in adults in different ways which can include low-level perceptual features, category membership, semantic relatedness, and statistical co-occurrences between items they view. Although due to recent studies we are starting to realize that infants also use chunking. Chunking helps them when they have to revisit an object later on and will be easier for them to recognize. It will help them have an idea of what it is because they take into consideration the features on the object. This will help them see how the objects are different. They also use different types of knowledges to help them with chunking like conceptual knowledge, spatiotemporal cue knowledge, and knowledge of their social domain.

There have been studies that use different chunking models like PARSER and the Bayesian model. PARSER is a is a chunking model designed to account for human behavior by implementing psychologically plausible processes of attention, memory, and associative learning. In a recent study, it was determined that these chunking models like PARSER are seen in infants more than chunking models like Bayesian. PARSER is seen more because it is typically endowed with the ability to process up to three chunks simultaneously.

Why is it impressive for Infants to be Chunking

Many individuals believed that infants didn't have the ability to use chunking since they thought their minds weren't able to process. They instead are able to use chunking by using their social knowledge, abstract knowledge and subtle cues since they can not create a perception of their social group on their own. It is also impressive because infants can form chunks using shared features or spatial proximity between objects. In 6 months after birth, infants’ short-term memory capacity encompasses only a single item, and expands to encompass two items by 8-10 months of age, and possibly up to four items. Later on, around 8-months they are capable of holding only up to two items in memory, computing pairwise statistics might be a more efficient strategy than trying to build up representations of chunks that could consist of many more than two items.

Chunking learning style and short term memory

Chunking can be form of data suppression that allows for more information to be stored in short-term memory. Rather than verbal short-memory measured by the number of items stored, Miller (1956) suggested that verbal short-term memory are stored as chunks. Later studies were done to determine if chunking was a form data compression when there is limited space for memory. Chunking works as data compression when it comes to redundant information and it allows for more information to be stored in short-term memory. However, memory capacity may vary.

Chunking and Working Memory

A experiment was done to see how chunking could beneficial to patients who had Alzheimer’s disease. This study was based on the how chunking was used to improve working memory in normal young people. Working memory is impaired in the early stages of Alzheimer's disease affects the ability to everyday task. It also affects executive control of working memory. It was found that participants who has mild Alzheimer's disease were able to use working memory strategies to enhance verbal and spatial working memory performance.

It has been long thought that chunking can improve working memory. A study was done to see how chunking can improve working memory when it came to symbolic sequences and gating mechanisms. This was done by having 25 participants learn 16 sequences through trial and error. The target was presented alongside a distractor and participants were to able to identify the target by using right or left buttons on a computer mouse. The final analysis was done on only 19 participants. The results showed that chunking does improve symbolic sequence performance through decreasing cognitive load and real-time strategy. Chunking has provided to be effective in reducing the load on adding items into the working memory. Having chunking allows for more items to be encoded into working memory with more availably to transfer into the long-term memory.