Sample Propositional Representations In Psychology Research Paper. Browse other research paper examples and check the list of research paper topics for more inspiration. If you need a research paper written according to all the academic standards, you can always turn to our experienced writers for help. This is how your paper can get an A! Feel free to contact our research paper writing service for professional assistance. We offer high-quality assignments for reasonable rates.
In cognitive psychology the term ‘proposition’ refers to a truth-valued unit of meaning that is represented symbolically in the human cognitive system. Propositions have two important components. First, they consist of semantic information structures that represent aspects of experience (e.g., events, states), real and hypothetical situations in the world, and abstract conceptual knowledge. Second, they include truth value and modality information that represents an attitude or belief on the part of a speaker, writer, or any holder of the proposition, with respect to the truth or falsity of the semantic relationships that are represented by the proposition. Propositions function as a semantic base for natural language where they represent types of semantic information that are distinguished in natural languages; and they function as a base for reasoning, serving as truth-valued and quantiﬁed predicates to which logical reasoning operations may be applied. This research paper examines propositional representations from a psychological point of view, the functions of propositions in the human cognitive system, models of propositional representations, and a closely related discourse analysis and model-building technique in cognitive science known as ‘propositional analysis.’
Academic Writing, Editing, Proofreading, And Problem Solving Services
Get 10% OFF with 24START discount code
1. Propositional Representation In Cognitive Systems
1.1 Propositions As Cognitive Representations
Like all cognitive representations, propositions are a form of semantic representation. As such, they may be used to represent concrete situations of cognition and action in the world, or they may represent more general or abstract conceptual knowledge that is a part of one’s long-term store of knowledge about the world and how to act in it. However, in contrast to propositions, systems of cognitive representation are usually thought of as large networks of concepts and relations that reﬂect a person’s comprehension and knowledge. Such networks are often referred to as ‘mental models,’ ‘situation models,’ ‘conceptual or semantic networks,’ ‘schemata,’ ‘knowledge frames,’ or in terms of other similar constructs (see Rumelhart and Norman 1988 for a review).
For example, when a visual scene is interpreted, it is represented semantically in the cognitive system as a network in which objects are identiﬁed and represented (as nodes), their properties are represented by links to attributes, and their relationships to each other are represented by particular types of semantic links. To talk about a scene requires the construction of propositions. These consist of truth-valued assertions about information from this network. The validity of these assertions can be evaluated and they can consequently serve as a basis for discussion, inferences, and reasoning. A major task for a theory of cognitive representation is to identify the types of concepts and relations that are used by humans to interpret and represent their physical world as it is interpreted and described through natural language.
An essential property of systems of cognitive representation is their relationship to the particular situation or situations in the world that they represent. Thus, systems of cognitive representation must support evaluations of their validity as representations of (real or hypothetical) situations in the world, or in relation to bodies of shared abstract or concrete knowledge (e.g., in a scientiﬁc ﬁeld). This semiotic property of cognitive representations underlies the truth value and modality information present in propositions, since it is information about asserted truth conditions that permits discussion and reasoning about perceived situations in the world. Consequently, the study of propositional representations in natural language processing, and of reasoning and inference through natural language, has provided psychologists with models of semantic representation that have had a large impact on theories of cognitive representation in general (see Frederiksen 1975, 1986, Kintsch 1974, Rumelhart and Norman 1975, van Dijk and Kintsch 1983).
1.2 The Inﬂuence Of Other Disciplines On Cognitive Theories Of Propositional Representation
Propositions are closely associated with the production and comprehension of natural language, where they represent semantic information that is asserted by a speaker or writer and explicitly encoded in language structures and discourse. Researchers studying language comprehension have shown that in addition to interpreting the propositions encoded in lexical and syntactic structures, understanding discourse also involves construction of inferred propositions. Inferences are used to complete, elaborate, and connect local propositional information into a coherent text representation, and relate a text’s meaning to prior knowledge and to its situational context (Denhiere and Rossi 1991, Graesser and Bower 1990, Kintsch 1998). The construction of propositional representations, whether through interpretation of natural language or through reasoning about situations in the world, is a fundamental aspect of the cognitive processes that enable us to understand and interpret the world. Thus, propositions are a fundamental aspect of the human cognitive system where they function in a variety of ways by providing the basic representational units of meaning in cognition and language.
Cognitive theories of propositions as units of semantic representation have been inﬂuenced by work from many ﬁelds including: linguistics (case grammar, linguistic semantics, text linguistics, pragmatics, systemic linguistics); logic; model theory; situation semantics; philosophy of language (speech act theory, semiotics); and computer science (artiﬁcial intelligence and language understanding, conceptual graph and semantic network representations, frame theory, scene analysis, and computational linguistics). Other relevant disciplines include ethnography of communication (conversational inference); cognitive sociology (conversational analysis, ethnomethodology); and cognitive linguistics. For further reading on work in computer science, see Brachman and Levesque 1985; for a multidisciplinary overview, see van Dijk 1997. Fauconnier (1997) discusses recent work in cognitive linguistics and related ﬁelds.
1.3 The Role Of Propositions In The Architecture Of The Cognitive System
The architecture of the cognitive system consists of its general properties as an information-processing system (e.g., Simon 1996 and other papers in the same volume). These include systems for: processing input information; constructing and storing internal symbolic codes; interpreting input information; accessing, using, and modifying stored prior knowledge; operating on internal representations (e.g., to understand, reason, and solve problems); and producing appropriate actions and results (including language) as output. The cognitive architecture that supports natural language processing capabilities consists of a stratiﬁed semantic processing system in which propositions are intermediate semantic units that are involved in producing and comprehending natural language in relation to its situational context and to the participants’ long-term store of conceptual and relational knowledge. Figure 1 identiﬁes the major elements of this stratiﬁed processing system and the central role that propositions play in comprehending and producing natural language discourse. The major levels of representation and processing are:
Natural language discourse: phonetic or lexical patterns; word morpheme sequences; syntactic structures; linguistic forms signaling discourse cohesion; linguistic and extralinguistic markers (e.g., gestures, stress, intonation) that relate discourse constituents to events, states, and representations in the spatiotemporal environment and signal the situational cohesion of the discourse; and (if the discourse is interactive) speech acts and conversational structures. Propositional representations: representations of concepts and relational semantic information explicitly encoded in natural language expressions (the text base); inferred propositions that elaborate and connect local propositions in the text and resolve anaphoric references; coherence inferences; macropropositions that summarize content; propositions generated by logical rules to represent the logical structure of arguments; and propositions that resolve extralinguistic references to the situational context.
Active mental representations of situations and knowledge: conceptual networks constructed to link propositional information into a contextually appropriate connected and coherent cognitive representation—a currently active mental model, situation model, or knowledge structure that is constructed based on the text propositions, situational information, and prior knowledge.
In a stratiﬁed semantic processing system, these multiple representations are constructed in a situational context that includes a physical environment (of things, states, actions, events, processes, sequences, etc.) and a representational environment (of symbols, language, icons, images, texts, etc.). They also reﬂect an individual’s use of prior conceptual and relational knowledge that is potentially relevant to the situation and content of the discourse.
2. Cognitive And Communicative Functions Of Propositions
2.1 The Semantic Content Of Linguistic Expressions
Propositions represent the semantic content of linguistic expressions. For example, in ‘The airplane was rising swiftly,’ the proposition would represent the structure of an event involving an action (rising), an object undergoing change as a result of the action (the airplane), and an attribute of the action (swiftly). It would also specify how the airplane is determined (deﬁnite) and quantiﬁed (singular), and the truth assertion about the proposition (positive truth value). To see how this can become more complex, in ‘Laura felt the airplane rising swiftly,’ the previous proposition would have to be embedded so as to reﬂect its semantic role as the thematic content of a process (feel) of a patient or experiencer (Laura), and this proposition would have its truth value speciﬁed and objects (Laura) quantiﬁed. As can be seen in Fig. 1, producing utterances in a language involves the speciﬁcation of propositions that ‘chunk’ conceptual and semantic information taken from situation or knowledge models into propositional units so that the information can be encoded morphologically and syntactically in a particular language. conceptual and relational knowledge
2.2 Units Of Meaning For Comprehension, Inference, Reasoning, Thinking, And Learning Processes
In comprehension, the end product is a mental representation of a situation and knowledge. The direction of processing can be data-driven (when the ﬂow of processing is from semantic interpretation of language units, to propositional inferences, to the construction of a situation model), or it can be knowledge-driven (when prior conceptual knowledge or knowledge frames are applied to guide and constrain the construction of a situation model that incorporates propositional information from a text base). Processes of inference and logical reasoning can operate on propositional predicates, or they may be based on mental models that are constructed to facilitate a more inductive style of reasoning based on concrete examples (Johnson-Laird 1996). Thinking processes involve the construction, activation, and modiﬁcation of situation models and propositions, and learning involves the incorporation of propositions and information from active situation and knowledge models into a more permanent knowledge structure.
2.3 Building Blocks For Systems Of Cognitive Representation
Propositions provide the building blocks for systems of cognitive representation. If knowledge is to be communicated through language or treated as an object of logical reasoning, then it must at some point be propositionalized. Consequently, there should be a close relationship between propositions and systems of cognitive representation. Thus, cognitive psychologists have adopted the strategy of building models of knowledge structure to incorporate semantic relationships that are present in the propositional semantic structures that are expressed in natural language (Frederiksen and Breuleux 1990, Rumelhart and Norman 1988).
2.4 The Semantic Representation Of Situations And Knowledge
The propositional information that we construct to describe situations through language is similar to the semantic structures we use to interpret and construct cognitive representations of situations and knowledge. Thus, a primary function of propositions and situation models is to represent and continually update our representations of knowledge and of the currently experienced situational contexts in which we are participating.
2.5 Communication, Conversational Inference, And Situated Cognition
In social environments of collaborative action, reasoning, and problem solving, the main channel of communication is conversational dialog. Participation in such task-oriented dialogs demands conversational inferences by the participants, and these are based on comprehension of the propositional content of the speech as it occurs within the situation and with reference to relevant domain knowledge. Communication and collaborative action in such environments depends on comprehension of the propositions asserted by the participants in the conversation. Thus, propositional representations play a central role in such situations of conversational interaction and socially situated cognition (see Koschmann 1999 for an example).
3. Models Of Propositional Representations
3.1 Types Of Propositions
Propositions represent a variety of types of semantic information that may be expressed in language. Types of propositions have been deﬁned in terms of the particular semantic relations and concepts that are involved in a particular type of situation that is being represented. A listing of some of the types of propositions that have been identiﬁed can help convey the richness of propositional representations.
Events: Events are structures involving an action that produces a change, its causes, results, goals, instruments, and other information specifying the participants and objects eﬀected, the internal structure of an event, and properties of the action, e.g., ‘The man walked from the top of the hill to the cemetery gate.’
Processes: Processes identify a process and the objects or persons who experience or are involved in the process; they include cognitive processes having thematic content (e.g., know, believe) and represent the structure of a complex process and its properties, e.g., ‘The war raged.’ ‘John knows Sally.’ ‘Laura felt the airplane rising gradually.’
States: States identify objects and specify properties of the objects (e.g., attributes, parts, set relations) as well as their determination and quantiﬁcation, e.g., ‘The painting is beautiful.’ ‘All of the visitors were tourists.’ ‘The painting is on the wall.’
Abstract relations: Abstract relations identify abstract concepts (corresponding to propositions) and their properties, e.g., ‘This theory is complex.’ ‘The diagnosis was not obvious.’
Identities: Identity relations link objects or propositions into identity sets, e.g., ‘The American President is George W. Bush.’
Algebraic relations: Order and equivalence relations are involved in comparative statements and apply to variables that represent values identiﬁed in other propositions, e.g., ‘John is taller than Mary.’ ‘His temperature is normal.’
Functions: Functions represent operations deﬁned on variables that return values, e.g., ‘Ottawa is close to Montreal.’ ‘Montreal is closer to Toronto than to New York.’ ‘The mean height of the team members is six feet.’
Dependency relations: Dependency relations make one proposition depend on another, such as by causal, conditional, and logical implication relations, e.g., ‘He won because he had trained intensively.’ ‘If you ﬁnish your supper, then you can have dessert.’
Conjoint relations: Conjoint relations, such as ‘and,’ alternating ‘or’ and exclusive ‘or,’ specify intersecting, alternating, or mutually exclusive sets of objects or propositions.
Tense and aspect components: These specify temporal properties of propositions, e.g., ‘He was attending the opera.’
Modality and truth value components: These specify the asserted truth conditions of propositions, e.g., ‘He may not have been at the Opera.’
Determination and quantiﬁcation components: These specify the determination and quantiﬁcation of concepts, e.g., objects, and of abstract concepts, e.g., ‘Six of the men volunteered,’ ‘All people are mortal.’
3.2 Predicate-Argument vs. Conceptual Graph Structures
Two sorts of formalisms have been used to represent propositions: predicate-argument notation, and conceptual graphs (Sowa 1984). Predicate-argument formalisms use a list notation in which a proposition is represented as a predicate (e.g., event) followed by a list of its arguments (the components of the event: the action, the object of the action, the result, etc.). Information about the determination and quantiﬁcation of objects may be embedded in its argument slot. For example, consider the following sentence taken from a procedural text in chemistry: ‘Collect the samples in an airtight bottle.’ An event proposition for this sentence might be as follows (in predicate notation):
7.1 Event: Act (collect), Object (samples: Def, Plural), Result (7.2, 7.3), Truth.Value (Positive);
The result consists of pointers to two embedded state propositions representing the location of the samples after the action has taken place and a property of the bottle:
7.2 State: Object (samples: Def, Plural), Locative. Relation (Object (bottle: Token, Singular)), Truth. Value (Positive);
7.3 State: Object (bottle: Token, Singular), Attribute.Relation (airtight), Truth.Value (Positive)
The alternative conceptual graph notation for this example is given in Fig. 2. Conceptual graphs are equivalent but have the advantage of helping to visualize the semantic relationships being represented by the propositions. This notation is particularly useful in developing models of knowledge structures in particular domains being studied.
3.3 Propositions As Canonical Frames Or Semantic Grammars
If propositions are constructed in interpreting and producing language, and in making inferences, how can we explain how they are generated? Two types of explanation have been considered: the canonical frame explanation, and the semantic grammar explanation. It is possible to think of propositions as a vast collection of frame structures which contain ‘slots’ (or variables) that can be ‘bound to’ particular conceptual information.
If we replace the arguments in the event proposition by variables, we have an event frame for the action collect which can be applied to many diﬀerent situations by ‘binding’ the variables to particular ‘values’: ‘Collect the butterﬂies in the jar’ (Y = butterﬂies, Z = a state representing the butterﬂies’ location in the jar).
Event: Act (collect), Object (Y: Determiner, Quantiﬁer), Result (Z), Truth.Value (Positive);
A conceptual frame theory requires a very large store of such frames in memory and processes to apply them to new linguistic or situational data.
Others have argued that representing propositions as a set of frames is insuﬃciently powerful to represent the full range of propositional representations that can be constructed. It is argued that like natural language structures, propositional structures reﬂect rules of formation, and propositional representations need to be deﬁned in terms of knowledge representation languages or semantic grammars. Semantic grammars provide a way of precisely deﬁning a language of propositional representation, and provide rules to guide the generation of propositions and semantic parsing of linguistic expressions. Both canonical frames and semantic grammars appear to be plausible as cognitive mechanisms for the construction of propositional representations (see Brachman and Levesque 1985, Frederiksen 1986, Rumelhart and Norman 1988 for discussions of both perspectives).
4. Propositional Analysis
Theories and models of propositional representation have provided a principled basis for the development of precise methods for analyzing the semantic content of texts as representations of knowledge, developing models of expert domain knowledge, and analyzing the cognitive representations and processes underlying cognitive processes in discourse comprehension and production, inference, reasoning, and other cognitive processes that are reﬂected in verbal protocols. Propositional analysis is essentially a method for analyzing the propositional content of texts, verbal protocols such as text recalls, responses to questions, reasoning, or talk-aloud protocols, and even task-oriented dialog in collaborative situations of learning or performance (see Frederiksen 1986 and Kintsch 1998).
Propositional analysis normally begins with transcription and segmentation of textual material. Linguistic analysis of the text segments may be undertaken in terms of the linguistic structures previously cited. Semantic parsing rules or propositional frames are applied to specify the propositions explicitly encoded in the sentences. The generation of propositions can be guided by computer tools that enable the application of semantic parsing rules or apply frames to analyze the propositional content of text segments. Inferred propositions may be added to the model to ﬁll recognized gaps in the text information or argument structure, relate the text to aspects of the context or task, summarize the macrostructure of the text, provide responses to questions, etc. Inference analysis can be based on data from subjects who verbally recall, discuss, or analyze a text. Models of situation and knowledge models may be constructed applying conceptual network formalisms to the propositions. Finally, texts can be analyzed in terms of high-level conceptual frame structures such as underlie procedural texts, problem-solving or reasoning protocols, argument structures, narratives, and story plot structures.
- Brachman R J, Levesque H J 1985 Readings in Knowledge Representation. Morgan Kaufmann, Los Altos, CA
- Denhiere G, Rossi J-P 1991 Text and Text Processing. NorthHolland, Amsterdam
- Fauconnier G 1997 Mappings in Thought and Language. Cambridge University Press, Cambridge, UK
- Frederiksen C H 1975 Representing logical and semantic structure of knowledge acquired from discourse. Cognitive Psychology 7: 371–458
- Frederiksen C H 1986 Cognitive models and discourse analysis. In: Cooper C R, Greenbaum S G (eds.) Studying Writing: Linguistic Approaches. Sage, Beverly Hills, CA, pp. 227–67
- Frederiksen C H, Breuleux A 1990 Monitoring cognitive processing in semantically complex domains. In: Frederiksen N, Glaser R, Lesgold A, Shafto M G (eds.) Diagnostic Monitoring of Skill and Knowledge Acquisition. Erlbaum, Hillsdale, NJ, pp. 351–91
- Graesser A C, Bower G H (eds.) 1990 The Psychology of Learning and Motivation: Inferences and Text Comprehension. Academic Press, New York, Vol. 25
- Johnson-Laird P N 1996 The process of deduction. In: Steier D S, Mitchell T M (eds.) Mind Matters: A Tribute to Allen Newell. Erlbaum, Mahwah, NJ, pp. 363–99
- Kintsch W 1974 The Representation of Meaning in Memory. Erlbaum, Mahwah, NJ
- Kintsch W 1998 Comprehension: A Paradigm for Cognition. Cambridge University Press, Cambridge, UK
- Koschmann T (ed.) 1999 Edge of many circles: making meaning of. Discourse Processes 27(2): 103–17
- Rumelhart D E, Norman D A 1975 Explorations in Cognition. W H Freeman, San Francisco
- Rumelhart D E, Norman D A 1988 Representation in memory.
- In: Atkinson R C, Herrnstein R J, Lindzey G, Luce R D (eds.) Stevens’ Handbook of Experimental Psychology, 2nd edn. Wiley, New York, pp. 511–87
- Simon H A 1996 The patterned matter that is mind. In: Steier D S, Mitchell T M (eds.) Mind Matters: A Tribute to Allen Newell. Erlbaum, Mahwah, NJ, pp. 407–31
- Sowa J F 1984 Conceptual Structures: Information Processing in Mind and Machine. Addison-Wesley, Reading, MA
- Van Dijk T A (ed.) 1997 Discourse as Structure and Process. Sage, London
- Van Dijk T A, Kintsch W 1983 Strategies of Discourse Comprehension. Academic Press, New York