Education of the republic of uzbekistan state university of world languages

Category learning
Category learning tries to understand how categories are acquired in the first place, whereas categorization study studies how categories are retained and used. To achieve this, researchers frequently use unique categories of arbitrary items (e.g., dot matrices) to ensure that the stimuli are completely unfamiliar to the participants. Researchers in category learning have mainly concentrated on two distinct types of category learning. Participants in classification learning are tasked with predicting category labels for a stimulus based on its presented features²⁶. Learning between-category information and diagnostic properties of categories are crucial to classification learning. In contrast, inference learning requires participants to infer the presence/value of a category feature based on the presence of other category features and/or a specified category label. Learning within-category information and the category's archetypal traits is crucial to inference learning²⁷.
There are two types of category learning tasks: supervised learning and unsupervised learning. Learners are given category labels in supervised learning assignments. The information retrieved from labeled example categories is subsequently used by learners to categorize stimuli into the proper category, which may require the abstraction of a rule or idea connecting observed object features to category labels. Learners are not given category labels in unsupervised learning assignments. Learners must therefore recognize intrinsic structures in a data collection and classify stimuli based on similarity. As a result, unsupervised learning is a process of developing a categorization system. Category learning tasks can take several forms:

• 
  Participants can learn categories in rule-based activities through explicit reasoning processes. The classification of stimuli in these tasks is performed through the application of a learnt rule (i.e., if stimulus is large on dimension x, respond A).
  
• 
  Prior to making categorization decisions, learners must synthesize perceptual information from various stimulus dimensions in information-integration tasks. Unlike rule-based jobs, information-integration tasks do not allow for simply articulated rules. Viewing an X-ray and attempting to determine the presence of a tumor can be viewed as a real-world instance of an information-integration challenge.
  
• 
  Prototype distortion challenges require students to create a prototype for a specific category. The properties of the prototype are then randomly manipulated to produce candidate exemplars for the category, which learners must classify as belonging to the category or not.
  

Researchers in category learning have offered numerous explanations for how humans learn categories. The prototype theory, the exemplar theory, and the decision bound theory are the dominant theories of category learning. According to the prototype theory, in order to understand a category, one must first learn the category's prototype. Following that, fresh stimuli are classified by selecting the category with the most similar prototype²⁸. According to the exemplar theory, in order to learn a category, one must first learn about the exemplars that belong to that category. Following that, a novel stimulus is classified by computing its similarity to known exemplars of potentially relevant categories and picking the category containing the most comparable exemplars²⁹. According to choice bound theory, in order to learn a category, one must first learn the parts of a stimulus space associated with specific responses or the borders (the decision bounds) that separate these response regions. A novel stimulus is then classified by determining which response zone it is contained within³⁰.
Categorization is critical for cognition, apprehension, and representation of the world around us. "Issues with conceptualization and categorization of the world are essential topics of cognitive science, subsequently - cognitive linguistics," writes E. Kubryakova³¹. "Many challenges that cognitive linguistic tackles at the modern stage of its evolution are associated with the study of linguistic methods and mechanisms for knowledge representation," says N. Boldyrev. In this context, language categorical systems are given considerable consideration”³². While discussing the significance and relevance of categorization as a cognitive activity, we should also consider another part of the problem. Various people categorize reality into categories in slightly different ways, even within the same country with everyone sharing the same linguistic and cultural foundation. On the other hand, everyone thinks in their own unique way, which results in diverse individual linguistic world representations and different ways of objectifying seemingly identical concepts and categories. When we compare two different civilizations, this feature stands out even more. One example could be the discrepancy between English and Russian age periodization words. Bearing this in mind, we might add F. de Saussure's well-known thesis that language is not a nomenclature; there are also a number of comments by modern scientists that fall into the same category. D. Chandler, for example, observes that "language categories are not merely a result of some preconceived organization of the world... These are not 'natural' concepts merely reflected' in language"³³. V. Demyankov mentions the same concept: “By means of language we construct what take for reality, so we can’t say that language provides reference to the real world as a really existing independent matter”. Furthermore, one of the most important aspects of categorization as a cognitive process is how categories (which are nonverbal in the first place) are objectified through spoken language. Regardless matter how crucial its tasks are in terms of cognitive processes; language is necessarily regarded as secondary to brainwork. Individual approaches to categorizing the world can be observed among people from the same linguistic and cultural setting. Additionally, it is visible not just in ordinary life but also on a professional level. One example is splitting the human pelvis as an anatomical entity into fictitious pieces. In fact, the pelvis lacks any clear external characteristics that would allow it to be exactly and unequivocally divided into segments and classified. As a result, there are several categories of the same anatomical entity based on different dividing criteria. A traumatologist and an obstetrician-gynecologist, for example, might divide the pelvis into wholly separate imaginary portions, each according to the special character of his subject matter. This could be one of the reasons for the lack of a unified classification of pelvic bones.