Pattern recognition (psychology)


In psychology and cognitive neuroscience, pattern recognition is a cognitive process that matches information from a stimulus with information retrieved from memory.
Pattern recognition occurs when information from the environment is received and entered into short-term memory, causing automatic activation of a specific content of long-term memory. An example of this is learning the alphabet in order. When a carer repeats "A, B, C" multiple times to a child, the child, using pattern recognition, says "C" after hearing "A, B" in order. Recognizing patterns allows anticipation and prediction of what is to come. Making the connection between memories and information perceived is a step in pattern recognition called identification. Pattern recognition requires repetition of experience. Semantic memory, which is used implicitly and subconsciously, is the main type of memory involved in recognition.
Pattern recognition is crucial not only to humans, but also to other animals. Even koalas, which possess less-developed thinking abilities, use pattern recognition to find and consume eucalyptus leaves. The human brain has developed more, but holds similarities to the brains of birds and lower mammals. The development of neural networks in the outer layer of the brain in humans has allowed for better processing of visual and auditory patterns. Spatial positioning in the environment, remembering findings, and detecting hazards and resources to increase chances of survival are examples of the application of pattern recognition for humans and animals.
There are six main theories of pattern recognition: template matching, prototype-matching, feature analysis, recognition-by-components theory, bottom-up and top-down processing, and Fourier analysis. The application of these theories in everyday life is not mutually exclusive. Pattern recognition allows us to read words, understand language, recognize friends, and even appreciate music. Each of the theories applies to various activities and domains where pattern recognition is observed. Facial, music and language recognition, and seriation are a few of such domains. Facial recognition and seriation occur through encoding visual patterns, while music and language recognition use the encoding of auditory patterns.

Theories

Template matching

Template matching theory describes the most basic approach to human pattern recognition. It is a theory that assumes every perceived object is stored as a "template" into long-term memory. Incoming information is compared to these templates to find an exact match. In other words, all sensory input is compared to multiple representations of an object to form one single conceptual understanding. The theory defines perception as a fundamentally recognition-based process. It assumes that everything one sees, they understand only through past exposure, which then informs their future perception of the external world. For example, A, A, and A are all recognized as the letter A, but not B. This viewpoint is limited, however, in explaining how new experiences can be understood without being compared to an internal memory template.

Prototype matching

Unlike the exact, one-to-one, template matching theory, prototype matching instead compares incoming sensory input to one average prototype. This theory proposes that exposure to a series of related stimuli leads to the creation of a "typical" prototype based on their shared features. It reduces the number of stored templates by standardizing them into a single representation. The prototype supports perceptual flexibility, because unlike in template matching, it allows for variability in the recognition of novel stimuli. For instance, if a child had never seen a lawn chair before, they would still be able to recognize it as a chair because of their understanding of its essential characteristics as having four legs and a seat. This idea, however, limits the conceptualization of objects that cannot necessarily be "averaged" into one, like types of canines, for instance. Even though dogs, wolves, and foxes are all typically furry, four-legged, moderately sized animals with ears and a tail, they are not all the same, and thus cannot be strictly perceived with respect to the prototype matching theory.

Multiple discrimination scaling

Template and feature analysis approaches to recognition of objects have been merged / reconciled / overtaken by multiple discrimination theory. This states that the amounts in a test stimulus of each salient feature of a template are recognized in any perceptual judgment as being at a distance in the universal unit of 50% discrimination from the amount of that feature in the template.

Recognition–by–components theory

Similar to feature–detection theory, recognition by components focuses on the bottom-up features of the stimuli being processed. First proposed by Irving Biederman, this theory states that humans recognize objects by breaking them down into their basic 3D geometric shapes called geons. An example is how one break downs a common item like a coffee cup: one recognizes the hollow cylinder that holds the liquid and a curved handle off the side that allows one to hold it. Even though not every coffee cup is exactly the same, these basic components help to recognize the consistency across examples. RBC suggests that there are fewer than 36 unique geons that when combined can form a virtually unlimited number of objects. To parse and dissect an object, RBC proposes that people attend to two specific features: edges and concavities. Edges enable the observer to maintain a consistent representation of the object regardless of the viewing angle and lighting conditions. Concavities are where two edges meet and enable the observer to perceive where one geon ends and another begins.
The RBC principles of visual object recognition can be applied to auditory language recognition as well. In place of geons, language researchers propose that spoken language can be broken down into basic components called phonemes. For example, there are 44 phonemes in the English language.

Top-down and bottom-up processing

Top-down processing

Top-down processing refers to the use of background information in pattern recognition. It always begins with a person's previous knowledge, and makes predictions due to this already acquired knowledge. Psychologist Richard Gregory estimated that about 90% of the information is lost between the time it takes to go from the eye to the brain, which is why the brain must guess what the person sees based on past experiences. In other words, we construct our perception of reality, and these perceptions are hypotheses or propositions based on past experiences and stored information. The formation of incorrect propositions will lead to errors of perception such as visual illusions. Given a paragraph written with difficult handwriting, it is easier to understand what the writer wants to convey if one reads the whole paragraph rather than reading the words in separate terms. The brain may be able to perceive and understand the gist of the paragraph due to the context supplied by the surrounding words.

Bottom-up processing

Bottom-up processing is also known as data-driven processing, because it originates with the stimulation of the sensory receptors. Psychologist James Gibson opposed the top-down model and argued that perception is direct, and not subject to hypothesis testing as Gregory proposed. He stated that sensation is perception and there is no need for extra interpretation, as there is enough information in our environment to make sense of the world in a direct way. His theory is sometimes known as the "ecological theory" because of the claim that perception can be explained solely in terms of the environment. An example of bottom up-processing involves presenting a flower at the center of a person's field. The sight of the flower and all the information about the stimulus are carried from the retina to the visual cortex in the brain. The signal travels in one direction.

Seriation

In psychologist Jean Piaget's theory of cognitive development, the third stage is called the Concrete Operational State. It is during this stage that the abstract principle of thinking called "seriation" is naturally developed in a child. Seriation is the ability to arrange items in a logical order along a quantitative dimension such as length, weight, age, etc. It is a general cognitive skill which is not fully mastered until after the nursery years. To seriate means to understand that objects can be ordered along a dimension, and to effectively do so, the child needs to be able to answer the question "What comes next?" Seriation skills also help to develop problem-solving skills, which are useful in recognizing and completing patterning tasks.

Piaget's work on seriation

Piaget studied the development of seriation along with Szeminska in an experiment where they used rods of varying lengths to test children's skills. They found that there were three distinct stages of development of the skill. In the first stage, children around the age of 4 could not arrange the first ten rods in order. They could make smaller groups of 2–4, but could not put all the elements together. In the second stage where the children were 5–6 years of age, they could succeed in the seriation task with the first ten rods through the process of trial and error. They could insert the other set of rods into order through trial and error. In the third stage, the 7-8-year-old children could arrange all the rods in order without much trial and error. The children used the systematic method of first looking for the smallest rod first and the smallest among the rest.

Development of problem-solving skills

To develop the skill of seriation, which then helps advance problem-solving skills, children should be provided with opportunities to arrange things in order using the appropriate language, such as "big" and "bigger" when working with size relationships. They should also be given the chance to arrange objects in order based on the texture, sound, flavor and color. Along with specific tasks of seriation, children should be given the chance to compare the different materials and toys they use during play. Through activities like these, the true understanding of characteristics of objects will develop. To aid them at a young age, the differences between the objects should be obvious. Lastly, a more complicated task of arranging two different sets of objects and seeing the relationship between the two different sets should also be provided. A common example of this is having children attempt to fit saucepan lids to saucepans of different sizes, or fitting together different sizes of nuts and bolts.