网范文：“Objectification Theory：Basic terms and principles” 探讨令人信服地表明,概念结构在很大程度上取决于经验。虽然声称被认为是有争议的,断言心理表征的结构可以通过语言进行了探讨。需要更多的理论来源的证据，来证明自己的指控。然而这是可能的,概念隐喻理论需要兼容现代心理学和神经学，英语论文网站，为其提供一个框架。这篇心理范文将寻求建立与现代认知科学的架构,关注抽象和具体概念的关系。
我将讨论最近的事态发展，在理论上的心理表征,并描述经典模型以及新的或更少的措施。最后,我将评估适用性探讨。由于本文的跨学科性质，我会努力证明措施和理论选择，在模型的发展上提出构想。下面的范文继续详述。
Introduction
In the previous chapter I introduced the notions of conceptual metaphor and tried to establish the background in which the theory operates, its main tenets, and goals for further development. Research has convincingly shown that much of conceptual structure depends on physical experience. Although the claim has been considered controversial (cf. Gibbs 2017), CMT asserts the structure of mental representation can be studied through language. In the previous chapter I have shown that language data alone is no longer considered sufficient to support CMT, and that the theory needs more sources of evidence to substantiate its claims regarding conceptualisation processes. However, for this to be possible, conceptual metaphor theory needs to provide a framework compatible with modern psychological and neurological research. This chapter will seek to establish the role of conceptual metaphor in modern cognitive science, focusing on the creation of abstract and concrete concepts.
I will discuss recent developments in theory of mental representation, and describe classical models (Markman 1999) as well as newer or less established approaches (Barsalou 1999; Semin and Smith 2017) to representation. Finally, I will assess the applicability of CMT to empirical conceptualisation research, both with and without the amendments introduced in Objectification Theory (Szwedek 2017). Due to the interdisciplinary nature of this thesis I will strive to justify the methodological and theoretical choices made in the development of the proposed conceptualisation model.
What may be a natural choice of framework for a linguist need not necessarily feel justified for someone outside the field. I am aware that the approach taken in this and the following theoretical chapters of this thesis may feel too detailed or redundant to some, while not explicit enough to others. However, if the aim is to produce a model that is potentially useful to researchers from a variety of fields within cognitive sciences, including psychology, linguistics, computational modelling, and AI compromises need to be made. I hope that this chapter is informative without dwelling too much on details, and useful without sounding authoritative. Both methodological and theoret- 43 ical frameworks of cognitive linguistics are in constant development, which makes it impossible to proclaim one approach as correct. Nevertheless, I hope that the methodological choices made for this model will to some degree hold up to the pressure of time.
Mental representation “Thinking can best be understood in terms of representational structures in the mind and computational procedures that operate on those structures” (Thagard 2017: 12). Accounting for knowledge representation is an important part of most cognitive models. Theories of mental representation serve as frameworks within which research approaches are developed and studies conducted. It is important to scrutinise not only the assumptions and predictions of the chosen theory of representation, but also its compatibility with the increasingly empirically-focused domain of cognitive science.
The debate whether conceptual structure is embodied or symbolic has a long history, and it is not for the first time that perception has been identified as a source of representation. Prior to the twentieth century theories of cognition relied on perception to a great extent (Barsalou 1999). The subsequent developments in logic, statistics and computational modelling inspired theories that divorced perception and meaning. Stemming from these developments was the classical representation theory, which postulates that mental representations take the form of amodal symbols (Fodor 1975, 1983).
According to amodal theories of cognition, perception and knowledge are separate. While they do not contest the claim that many mental representations stem from perception, action or introspection, these types of theories are based on the belief that concepts are stored in the semantic knowledge system in the form of amodal symbols (Markman 1999; Markman and Dietrich 2017a). More recently, theories of mental representation have undergone a paradigm shift again from the classical to a more embodied approach to conceptualisation. This change may be helpful in reconciling the psychological and cognitive linguistic research on mental representation and semantic storage or the so-called mental lexicon.
The classical theory of representation assumes that perceptual and conceptual systems are separate (Goldstone and Barsalou 1998) and that concepts, in particular 44 abstract concepts, are arbitrary and amodal in nature (Lakens 2017). This belief has been challenged by research demonstrating that the sensorimotor cortex is involved in semantic processing (Rohrer 2017, 2017; Rizzolatti and Craighero 2017). Areas in the brain formerly thought to be responsible for sensorimotor functions appear be involved with higher cognitive processes including conceptualisation and language (Hauk and Pulvermüller 2017; Pulvermüller et al. 2017, 2017). In other words, when we are asked to imagine playing tennis, not only does this activate the brain areas responsible for language comprehension and memory, but also the motor cortex areas normally involved in playing. In fact, this discovery has already been used to detect consciousness of patients in vegetative state (Cruse et al. 2017; Owen et al. 2017).
It is slowly becoming clear that Lakoff and Johnson (1999) who hypothesised that conceptual structure reflected bodily experience may have been right in many respects, including the relation between conceptual structure, language, and perception. Although their claims were based on a set of correspondences between linguistic expressions, the beliefs expressed in CMT are increasingly corroborated by evidence from empirical studies. Amodal representation is among many principles of the classical theory of representation that have been questioned by research in the vein of embodied cognition, including perceptual symbol systems, dynamic systems, and situated cognition based accounts of thought and language (Markman and Dietrich 2017a). The ongoing debate is of importance not only to researchers in the broad field of cognitive science, but also directly to cognitive linguistics which is largely based on traditional theories of representation. The change in framework may be slow but is significant on many levels. In particular, the shift towards a more empirically based framework could become mortar that connects all the heretofore separate bricks within cognitive science including prototype theory, image schemata, cognitive linguistics and conceptual metaphor.
Language, memory and representation structure
Traditional theories of representations are grounded in language (Collins and Loftus 1975; Fodor 1975). Although they treat language and symbols as different, the adopted approach to knowledge representation mirrors linguistic structure (Barsalou et al. 2017). For example, predicates for objects, events, and properties often correspond to the words that denote them. For instance, there is a substantial overlap between the concept of a bird and the meaning of the word “bird”. Amodal symbols in relevant literature are typically represented by words, under the assumption that the word is close to what constitutes the content of the symbol. Furthermore, symbolic thought is assumed to be analogous to language in many ways.
Understanding language is based on sequential processing of words in a sentence, just as conceptual processing is assumed to be sequential, with symbols processed as lists or in sentence-like structures (Fodor and Pylyshyn 1988). Although the classical account of representation seems compatible with structural linguistics, researchers in the field of cognitive linguistics tend to adopt non classical assumptions approaches to representation. This is because amodality of representations cannot be easily reconciled with the belief that cognition is embodied, a key assumption of cognitive semantics (Lakoff 1987; Lakoff and Johnson 1999). Cognitive linguistics is far from rejecting any kind of connection between words and concepts. Quite on the opposite, concepts and words are considered closely related. “It is nearly impossible to talk about a child learning the concept of sheep without her learning the word because the evidence that the child knows the concept comes from her applying the word correctly” (Murphy 2017: 386).
This assumption is reflected not only in the theoretical framework of the discipline, but also in study design as many cognitive lin- 48 guistic studies rely solely on linguistic evidence to draw inferences about conceptual structure, a fact that has been vigorously criticised from outside and within the domain (Murphy 1997; Gibbs 2017, 2017). Another classical assumption is that all representations need to be enduring and, therefore, stored in long-term semantic memory (Markman and Dietrich 2017a). However, grounded cognition approaches including situated cognition and dynamic systems views present possible alternatives. The former states that many aspects of the world remain stable so that we do not need to remember them, or code in an enduring form.
The phenomenon of change blindness, when we do not notice changes in these elements of a scene that are out of our conscious focus, could be seen as evidence in support of this hypothesis. Alternatively, the dynamic systems approach defines representations as dynamic states of a neural network. Such states by definition are neither amodal nor enduring because information is supplied from the sensorimotor systems, and dynamically influences the state of the network. With each new piece of information the pattern of activation in a network changes, and the mental representation is slightly adjusted. Both grounded cognition approaches make a compelling argument that representation does not necessarily need to be enduring or amodal. It is entirely possible that representations change in relation to experience, and depend on the original sensory information.
What is more, studies show that linguistic forms in the brain language systems and situated simulations in the brain modal systems are related (cf. Barsalou et al. 2017). There seems to be no need for amodal enduring representations. Nevertheless, we are far from conclusively stating which of the two approaches serves as a better foundation for a mental representation model. It has been mentioned in the previous section that a felicitous account of mental representation needs to account for both abstract and concrete concepts.
Concreteness and the symbol grounding problem It has been stated already that amodal representations are by definition not directly grounded in experience. While the classical account allows for perceptual involvement 49 in categorisation, the process by which percepts become divorced from modality-specific data to become concepts is unclear. Classical theories of mental representation need to account for what is called the symbol grounding problem, or the nature of the relationship between the symbol and its reference to its specific instances (Harnad 1990). In contrast, this issue does not arise in embodiment- and simulation-based accounts of representation used by cognitive linguistics. The hypothesis of grounded cognition states that representations are grounded in sensory modalities.
This assumption forms the basis of a substantial amount of research within cognitive linguistics in general and CMT in particular. Often studies do not focus on searching for the origin of representation, but rather on showing that conceptual structure is grounded in experience. Concrete concepts are assumed to be directly or indirectly embodied, or based on bodily experience and sensory data. There are many contrasting accounts of abstract conceptualisation. It is the intention of the author of the present thesis to show how concept creation processes can be modelled through conceptual metaphorization. Proponents of the classical approach claim that it is impossible to fully explain the emergence of abstract concepts without reference to amodal symbols. However, studies suggest that representations grounded in specific sensory modalities are flexible, and that both concrete and abstract concepts can be accounted for in this manner (Barsalou 1999; Goldstone and Barsalou 1998). Because of the close relation between the two processes, research on the nature of concepts is important for categorisation studies.
Categorisation Categorisation is one of the most basic cognitive and indispensable functions of living organisms. Even amoebas distinguish between two categories of things they encounter: “food” and “not food” (Lakoff 1987). For the purpose of this study we will follow the definition of category as a class of objects in the world (Murphy 2017). Categories are useful cognitive devices. Once established, they serve to reduce the mental load. In fact, Rosch (1999: 252) identifies providing maximum information with minimum cognitive 50 effort as one of the principles of categorisation. Newly encountered entities can be classified as category members rather than analysed separately.
For instance, if we have two categories for animals, domestic (friendly) and wild (potentially hostile) sorting unfamiliar animals we encounter into one of those groups relieves us from having to decide whether we should expect any animal we encounter to be friendly or not on an individual basis. Furthermore, categories permit generalisation and inferencing. Once we decide that the goat we met is a domesticated animal we may also determine that all goats are domesticated and, as a result, infer that goats are non-hostile. Naturally, category based inferences and generalisations are not error-proof. The goat we classified as domesticated may prove to be wild, or it may be a domesticated animal with a nasty temper. The fact that categorisation judgements are vulnerable to error is illustrated by such phenomena as stereotyping (Andersen and Klatzky 1987; Hamilton 1981; Zarate and Smith 1990). Nevertheless, categorisation remains a ubiquitous cognitive phenomenon. Whether it is recognising that the person sitting across the table is our spouse, or pronouncing a joke to be racist, most cognitive acts can be seen as acts of categorisation (Goldstone and Kersten 2017). Classical and embodied accounts of mental representation take different approaches to categorisation and, as a consequence, the nature of concepts.()
网站原创范文除特殊说明外一切图文作品权归所有；未经官方授权谢绝任何用途转载或刊发于媒体。如发生侵犯作品权现象，保留一切法学追诉权。
更多范文欢迎访问我们主页 当然有需求可以和我们 联系交流。-X()
，英语论文范文 |
