Coh-Metrix: Analysis of text on cohesion and language

Advances in computational linguistics and discourse processing have made it possible to automate many language- and text-processing mechanisms. We have developed a computer tool called Coh-Metrix, which analyzes texts on over 200 measures of cohesion, language, and readability. Its modules use lexicons, part-of-speech classifiers, syntactic parsers, templates, corpora, latent semantic analysis, and other components that are widely used in computational linguistics. After the user enters an English text, Coh-Metrix returns measures requested by the user. In addition, a facility allows the user to store the results of these analyses in data files (such as Text, Excel, and SPSS). Standard text readability formulas scale texts on difficulty by relying on word length and sentence length, whereas Coh-Metrix is sensitive to cohesion relations, world knowledge, and language and discourse characteristics.     

Coh-metrix: analysis of text on cohesion and language.

Advances in computational linguistics and discourse processing have made it possible to automate many language- and text-processing mechanisms. We have developed a computer tool called Coh-Metrix, which analyzes texts on over 200 measures of cohesion, language, and readability. Its modules use lexicons, part-of-speech classifiers, syntactic parsers, templates, corpora, latent semantic analysis, and other components that are widely used in computational linguistics. After the user enters an English text, CohMetrix returns measures requested by the user. In addition, a facility allows the user to store the results of these analyses in data files (such as Text, Excel, and SPSS). Standard text readability formulas scale texts on difficulty by relying on word length and sentence length, whereas Coh-Metrix is sensitive to cohesion relations, world knowledge, and language and discourse characteristics.     

Linguistic correlates of self in deceptive oral autobiographical narratives

The current study collected orally-delivered autobiographical narratives from a sample of 44 undergraduate students. Participants were asked to produce both deceptive and non-deceptive versions of their narrative to two specific autobiographical question prompts while standing in front of a video camera. Narratives were then analyzed with Coh-Metrix software on 33 indices of linguistic cohesion. Following a Bonferroni correction for the large number of linguistic variables (p < .002), results indicated that the deceptive narratives contained more explicit action verbs, less linguistic complexity, and less referential coherence (sentences being cohesive with each other). The results support a theory that, in deceptive narratives, there is greater narrative distance between the self that narrates and the self that is narrated about. This suggests that narrative selves are constituted not as autonomous selves, but are subject to processes (e.g., psychological, linguistic, social) that are likely operating on a subconscious level.     

Processing of temporal information: Evidence from eye movements

In two experiments, we recorded eye movements to study how readers monitor temporal order information contained in narrative texts. Participants read short texts containing critical temporal information in the sixth sentence, which could be either consistent or inconsistent with temporal order information given in the second sentence. In Experiment 1, inconsistent sentences yielded more regressions to the second sentence and longer refixations of it. In Experiment 2, this pattern of eye movements was shown only by readers who noticed the inconsistency and were able to report it. Theoretical and methodological implications of the results for research on text comprehension are discussed.     

Learning, thinking, and emoting with discourse technologies

This is an unusual moment in the history of psychology because of landmark advances in digital information technologies, computational linguistics, and other fields that use the computer to analyze language, discourse, and behavior. The technologies developed from this interdisciplinary fusion are helping students learn and think in ways that are sensitive to their cognitive and emotional states. Recent projects have developed computer technologies that help us understand the nature of conversational discourse and text comprehension in addition to improving learning. AutoTutor and other systems with conversational agents (i.e., talking heads) help students learn by holding conversations in natural language. One version of AutoTutor is sensitive to the emotions of students in addition to their cognitive states. Coh-Metrix analyzes texts on multiple levels of language and discourse, such as text genre, cohesion, syntax, and word characteristics. Coh-Metrix can assist students, teachers, principals, and policymakers when they make decisions on the right text to assign to the right student at the right time. Computers are not perfect conversation partners and comprehenders of text, but the current systems are undeniably useful. (PsycINFO Database Record (c) 2011 APA, all rights reserved).     

Not all narrative shifts function equally

Readers of narratives keep track of narrative events and the information associated with these events. Does some of this associated information help structure the processing of and memory for the narrative? In three experiments, we examined the role of basic event building blocks (character, time, and location) in event indexing during text comprehension. These three experiments dealt with perceived coherence, perceived cohesion, and on-line processing, respectively. The results indicated that characters are more likely to serve as event indexes. Although the findings with respect to indexing were similar in all three experiments, interesting differences emerged as a function of the level of text comprehension examined (coherence, cohesion, or on-line processing).     

Referential cohesion and logical coherence of narration after closed head injury

A group with closed head injury was compared to neurologically intact controls regarding the referential cohesion and logical coherence of narrative production. A sample of six stories was obtained with tasks of cartoon-elicited story-telling and auditory-oral retelling. We found deficits in the clinical group with respect to referential cohesion, logical coherence, and accuracy of narration. The occurrence of deficits depended on the condition of narrative production and, to some extent, on the particular story used. The primary implications of this study pertain to the attention given by researchers to the feature of discourse production being studied and processing demands of the task.     

Nursing Home Adaptation in Visually Impaired Older Adults Using a Narrative Approach1,2

This study used a narrative approach to examine the difficulties that older visually impaired people experience when adapting to life in a nursing home. The interview data were collected from one blind aged person (n = 1), focusing on the adaptation process from moving into the nursing home to the present and maintaining context of narrative that included the life history. Subcategories were extracted based on a framework of temporal coherence, causal coherence, thematic coherence, and situational coherence. This study revealed three factors related to adaptation to life in a nursing home for a visually impaired person, including psychological adaptation: becoming accustomed to life in the facility after changes in the living environment, having gratitude and maintaining “moderating human relationships,” and participating in activities connected to one's purpose in life. He was able to psychologically adapt and came to accept his death through an awareness of aging.     

The epistemic implications of engineering rhetoric

The texts (and talk) of engineers take different forms. In this essay, I present and critique several texts written for different purposes and audiences but all intended to convey to the reader the technical details of whatever they are about—whether a textbook passage describing the fundamental behavior of an electrical component, a journal article about a mathematical technique intended for use in design optimization, a memo to co-workers within a firm about a heat transfer analysis of a remotely sited building, or a general introduction to the field of? ergonomics? My aim is to explore how the ways in which engineers describe and document their problems and projects frame what they accept, display and profess as useful knowledge. In this I am particularly interested in how engineers envision the ‘users’ of, or participants in, their productions. Like science, engineering texts are written as if they were timeless and untainted by socio-cultural features. However, a technical treatise is not devoid of metaphor or creative rendering of events; there is always a narrative within which worldly data and instrumental logic is embedded—but it is a story in which the passive voice prevails, history is irrelevant, and the human actor or agent is painted in quantitative parameters fitting the occasion. Whether this rhetoric can be sustained in the face of challenges to traditional ways of doing engineering is an open question.     

The Spirit(s) of Science: Paradoxical Positivism as Religious Discourse among Spiritualists

This article examines a narrative dilemma that popular texts on evolution face. On the one hand, popular science tends to privilege linear and culturally familiar narrative structures, as previous studies of popularization have often emphasized. On the other hand, however, the Darwinian idea of natural selection resists linear narration, as narrative theorist H. Porter Abbott has argued. This resistance arises from the fact that evolution by natural selection lacks proper narrative entities and narrative events and that it relies on two parallel narrative levels, the levels of species and organism. This paper explores how two popular science books on evolution negotiate this narrative dilemma by introducing a third narrative level. Both texts appropriate characteristics from the narrative levels of species and organism and project them on molecular and minute scales by portraying evolution as a micro-narrative that takes place in chromosomes, genes, cells and microscopic details of human organs. While this textual strategy produces a coherent and compelling narrative that for the most part succeeds in masking the structural gap between the narrative levels of species and organism, it also risks naturalizing cultural imagery. In particular, this micro-narrative tends to represent popular gender ideologies as biological truths embedded in molecular processes within our bodies.     

Effects of models and demonstrations of personal narratives on the use of lexical ties by writers with mental retardation

A single subject reversal design was used to investigate the effects of two methods for teaching writing to eight adolescents with mild or moderate mental retardation during a weekly process writing workshop. During each workshop, each student wrote a personal narrative while the teacher circulated among the group and provided positive verbal feedback and spelled words on request. Students' texts were later typed with corrected spelling and punctuation and returned to them with verbal praise before the next writing workshop. During the first experimental condition, the subjects listened to an exemplar text selected by the teacher. In the second phase, the teacher read a narrative she had written as the exemplar text. During the third experimental condition, the teacher orally self-instructed herself through steps in writing a narrative. The subjects' texts were analyzed for number of words written and number and type of lexical ties used to produce cohesion between sentences in a text. The model narrative produced the best gains in the number of words written by the poorer readers, while the demonstration improved the cohesiveness of their texts. The better readers and writers wrote more cohesively during the model narrative and produced longer text during the demonstration condition.     

Coherence formation during narrative text processing: a comparison between auditory and audiovisual text presentation in 9- to 12-year-old children

Cognitive Processing - In an experiment with 114 children aged 9–12&nbsp;years, we compared the ability to establish local and global coherence of narrative texts between auditory and...     

Contextual influences on narrative discourse in normal young adults

In an effort to provide some expectation regarding the influences of task and rehearsal, the narrative ability of 20 normal young adults was examined under the conditions of story retelling and story generation. Each task was repeated after a 1-week interval in order to investigate a possible practice effect. Sentence production, intersentential cohesion, and story grammar organization were analyzed. Results indicated that the subjects' performance varied as a function of task presentation as well as the measure used to describe narrative production. The only across-trials difference noted was an increase in cohesive adequacy from trial 1 to trial 2 for one of the story generation tasks. It may be concluded that (a) context does influence the manner and competence of narrative presentation, and (b) that rehearsal of a complex narrative improves narrative coherence.     

Monitoring of comprehension: The role of text difficulty in metamemory for narrative and expository text

The effect of text difficulty on metamemory for narrative and expository text was investigated. In Experiment 1, we found an interaction between type of text and type of question (thematic or detailed). For readers of narrative texts, correlations between predicted and actual performance were highest for detailed questions, but this pattern was reversed for readers of expository texts. Next, text difficulty was explored as a possible factor affecting metamemory accuracy. In Experiments 2 and 3, metamemory accuracy was a nonmonotonic function of text difficulty. Subjects made remarkably accurate predictions of future performance (meanG > .6) for both narrative and expository texts that were of intermediate difficulty (approximately a 12th-grade reading level). We propose anoptimum effort hypothesis, predicting greatest metamemory accuracy when the texts are of intermediate difficulty.     

The impact of children’s lexical and morphosyntactic knowledge on narrative competence development: A prospective cohort study

The authors investigate the contribution of children's early comprehension of relational terms and morphosyntactic knowledge to the development of narrative competence in kindergarten and Grade 1. Narrative competence was assessed through the cohesion, coherence, and structure of children’s productions. The participants in this study were 714 Italian children. The authors measured their oral narrative competence through a storytelling task at the beginning and end of the kindergarten year. A total of 115 children were randomly selected and followed through Grade 1, and their narrative competence was measured again. According to the path analysis model, early morphosyntactic knowledge contributes to explain narrative competence in Grade 1. Early comprehension of relational terms contributes to narrative competence at the end of the school year. These findings confirm the importance of exploring the influence of early language skills on narrative competence development and suggests early intervention at the level of language antecedents of narrative competence.     

Joint Reading of Informational Science Text Versus Narrative Stories: How Does each Affect Language and Literacy Abilities Among Kindergarteners?

The effectiveness of a literacy intervention program based on a joint interactive reading of informational science texts in increasing children’s science vocabulary, language and literacy skills, delivered by the kindergarten teacher, was examined in 34 Hebrew-speaking kindergarten children exhibiting different levels of emergent literacy skills. The impact of the informational science text intervention was compared to that of a similar program based on narrative stories. Post-intervention, both groups improved their domain-specific vocabulary – science or narrative – but the change in domain-specific vocabulary learning was lower in the informational science group as opposed to the narrative group. Improvement was also found with regard to general vocabulary, print concepts, and listening comprehension following both interventions. Change in scientific vocabulary was positively related with change in morphological awareness, change in print concepts, and listening comprehension. No such findings were found with regard to narrative vocabulary. The study suggests that a short informational science intervention program can enhance science vocabulary and literacy skills in kindergarten children and should be used more often in kindergartens.     

Typing versus thinking aloud when reading: Implications for computer-based assessment and training tools

The goal of this study was to assess the impact of modality of production of think-aloud protocols on reading strategies. Readers in two studies spoke or typed protocols for narrative or science texts and completed comprehension tests for each text. Human judges identified the presence of paraphrasing, bridging inferences, and elaborating within the protocols. Reading comprehension skill was assessed with the Nelson-Denny test. With respect to narrative texts, paraphrasing and bridging were less frequent when readers were typing than when they were thinking aloud. With respect to science texts, less-skilled readers made bridging inferences more frequently when typing than when speaking. Conversely, skilled readers generated more paraphrases than bridges when typing thoughts but not when speaking. These results have implications for computer-based tools for reading assessment and intervention.     

From syntactical to textual strategies of argumentation

The paper focuses on tevelopmental relations between syntactical complexity, cohesion — especially conjuctional connection — and textual coherence in a sample of 150 argumentative texts written by school children (grades 4, 7, 10 and 12) and young adults (university students). In common sense and even in linguistics and psychology written text and especially written argument has been taken to be the prototype of syntactically complex, self-contained and explicit text over a long period of research on the topic. Thus it has been argued that syntactic connectedness and explicit cohesion of propositions were the most valuable indicators for a well-done contextualization and abstraction of content space and rhetorical space in writing. Empirical data show this common-sense position to be questionable. The discussion emphazises the role of argumentative implicitness as a necessary condition for getting the reader involved in the discourse and thus for the emergence of coherence in argumentative texts.     

Relations Between Narrative Coherence,Identity, and Psychological Well‐Being in Emerging Adulthood

The hypothesis that the ability to construct a coherent account of personal experience is reflective, or predictive, of psychological adjustment cuts across numerous domains of psychological science. It has been argued that coherent accounts of identity are especially adaptive. We tested these hypotheses by examining relations between narrative coherence of personally significant autobiographical memories and three psychological well‐being components (i.e., purpose and meaning, positive self‐view, positive relationships). We also examined the potential moderation of the relations between coherence and well‐being by assessing the identity content of each narrative. We collected two autobiographical narratives of personally significant events from 103 undergraduate students and coded them for coherence and identity content. Two additional narratives about generic/recurring events were also collected and coded for coherence. We confirmed the prediction that constructing coherent autobiographical narratives is related to psychological well‐being. Further, we found that this relation was moderated by the narratives' relevance to identity and that this moderation held after controlling for narrative ability more generally (i.e., coherence of generic/recurring events). These data lend strong support to the coherent narrative identity hypothesis and the prediction that unique events are a critical feature of identity construction in emerging adulthood.     

What is General Philosophy of Science?

The very idea of a general philosophy of science relies on the assumption that there is this thing called science??as opposed to the various individual sciences. In this programmatic piece I make a case for the claim that general philosophy of science is the philosophy of science in general or science as such. Part of my narrative makes use of history, for two reasons. First, general philosophy of science is itself characterised by an intellectual tradition which aimed to develop a coherent philosophical view of science, qua a part of culture with distinctive epistemic features and a distinctive relation to reality. But, second, this tradition went through some important conceptual shifts which re-oriented it and made it more sensitive to the actual development of science itself. The historical narrative focuses on three such moments: the defining moment, associated with Aristotle, and two major conceptual turns, related to Kant and Duhem. The pressures on the very idea of a general philosophy of science that followed the collapse of the macro-models of science that became popular in the 1960s, the pressures that lay all of the emphasis on fragmentation and not on integration, can be dealt with by a new synthesis within general philosophy of science of the constitutive and the historical, in light of the intellectual tradition that has defined it.