User:JSchaef/sandbox

There seems to be a ton of information about sequence learning, so I think by the end of this project I will hopefully have made at least two sections and thoroughly explained those. With the rest of my heading ideas on the talk page.

Sequence Learning

- introduction with definition and summary of article

- the different approaches to sequence learning

- types of sequence learning

    - temporal sequence learning
    - motor sequence learning
    - implicit sequence learning

- Sequence learning problems: four basic categories

    - sequence prediction
    - sequence recognition
    - sequence generation
    - sequential decision making

- Sequence learning models

    - learning paradigms
    - implementation paradigms
    - whether the world is deterministic or probabilistic
    - whether the world is Markovian or non- Markovian;
    - whether the task is closed- or open-loop
    - whether action is involved
    - whether an action policy is deterministic or stochastic 
    - applicable domains—for example, software agents, speech and language processing, navigation learning, or motor sequence learning.

- Experiments

- Role of different parts of the brain in sequence learning

Sequence learning

Sequence learning is inherent to human ability because it is an integrated part of conscious and nonconscious learning as well as activities. Sequences of information or sequences of actions are used in various everyday tasks: "from sequencing sounds in speech, to sequencing movements in typing or playing instruments, to sequencing actions in driving an automobile." Sequence learning can be used to study skill acquisition and used in studies of various groups ranging from neuropsychological patients to infants. According to Ritter and Nerb, “The order in which material is presented can strongly influence what is learned, how fast performance increases, and sometimes even whether the material is learned at all.” Sequence learning, more known and understood as a form of explicit learning, is now also being studied as a form of implicit learning as well as other forms of learning. Sequence learning can also be referred to by sequential behavior, behavior sequencing, and serial order in behavior.

History

In first half of the 20th century, Margaret Floy Washburn, John B. Watson, and other behaviorists believed behavioral sequencing to be governed by the reflex chain, which states that stimulation cause by an initial movement triggers an additional movement, which triggers another additional movement, and so on. In 1951, Karl Lashley, a neurophysiologist at Harvard University, published, “The Problem of Serial Order in Behavior,” addressing the current beliefs about sequence learning and introducing his hypothesis. He criticized the previous view on the basis of six lines of evidence:

“(1) Movements can occur even when sensory feedback is interrupted. (2) Some movement sequences occur too quickly for elements of the sequences to be triggered by feedback from the preceding elements. (3) Errors in behavior suggest internal plans for what will be done later. (4) The time to initiate a movement sequence can increase with the length or complexity of the sequence. (5) The properties of movements occurring early in a sequence can anticipate later features. (6) Neural activity can indicate preparation of upcoming behavior events, including upcoming behavior events in the relatively long-term future.”

Hierarchical organization of plans

Karl Lashley negated that sequence learning, or behavioral sequencing or serial order in behavior, is not attributed from sensory feedback. Rather, he proposed that there are plans for behavior since the nervous system prepares for some behaviors but not others. He said that there was a hierarchical organization of plans. He came up with several lines of evidence.

The first evidence states that the context changes functional interpretations of the same behaviors. Such as the way “wright, right, right, rite, and write” are interpreted based on the context of the sentence. “Right” can be interpreted as a direction or as something good depending on the context. A second line of evidence says that errors are involved in human behavior as hierarchical organization. In addition, “hierarchical organization of plans comes from the timing of behavioral sequences.” The larger the phrase, the longer response time, which factors into “’decoding’” or “’unpacking’” hierarchical plans. Another source of evidence is derived from how easy or hard it is to learn a sequence. The mind can create a “memory for what is about to happen” as well as a “memory for what has happened.” The final source of evidence for the hierarchical organization of plans is characterized by chunking (psychology). This skill puts single units into a unit comprised of multiple units.

Types of sequence learning

There are two broad categories of sequence learning – explicit and implicit – with subcategories. Explicit sequence learning was known and studied from the discovering of sequence learning. However recently, implicit sequence learning has gained more attention and research. A form of implicit learning, implicit sequence learning refers to the underlying methods of learning that people are unaware of; in other words, learning without knowing. Hsiang-Ling Jennifer Lin states that the exact properties and number of mechanisms of implicit learning are debated by researchers. Other forms of implicit sequence learning include motor sequence learning, temporal sequence learning, and Associative Sequence Learning.

Sequence learning problems

Sequence learning problems are used to better understand the different types of sequence learning. There are four basic sequence learning problems: sequence prediction, sequence generation, sequence recognition, and sequential decision making. These “problems” show how sequences are formulated. They show the patterns sequences follow and how these different sequence learning problems are related to each other.

Sequence prediction attempts to predict the next immediate element of a sequence based on all of the preceding elements. Sequence generation is basically the same as sequence prediction where a sequence attempts to be pieced together one by one the way it naturally occurs. Sequence recognition takes certain criteria and determines whether or not the sequence is legitimate. Sequential decision making or sequence generation through actions breaks down into three variations: goal-oriented, trajectory-oriented, and reinforcement-maximizing. These three variations all want to pick an action(s) or step(s) that will lead to the goal in the future.

These sequence learning problems reflect hierarchical organization of plans because each element in the sequences build on the previous elements.