Dissertations / Theses on the topic 'Letter processing'
To see the other types of publications on this topic, follow the link: Letter processing.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 18 dissertations / theses for your research on the topic 'Letter processing.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Schmitt, Andreas [Verfasser], and Thomas [Akademischer Betreuer] Lachmann. "What makes a letter a letter? – New evidence for letter-specific processing strategies / Andreas Schmitt ; Betreuer: Thomas Lachmann." Kaiserslautern : Technische Universität Kaiserslautern, 2016. http://d-nb.info/1122110987/34.
Full text
2
Wagstaffe, Joanna Katherine. "The encoding of letter position in orthographic processing." Thesis, University of Nottingham, 2006. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.438427.
Full text
3
Mycroft, Rachel Helen. "Causes and mechanisms of letter-by-letter reading : the roles of visual and language processing." Thesis, University of Exeter, 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.391887.
Full text
4
Bolger, Patrick Anthony. "Alphabetic processing in English and Spanish." Diss., Tucson, Arizona : University of Arizona, 2005. http://etd.library.arizona.edu/etd/GetFileServlet?file=file:///data1/pdf/etd/azu%5Fetd%5F1289%5F1%5Fm.pdf&type=application/pdf.
Full text
5
Leach, Corinne. "MANIPULATING TEMPORAL COMPONENTS DURING SINGLE-WORD PROCESSING TO FACILITATE ACCESS TO STORED ORTHOGRAPHIC REPRESENTATIONS IN LETTER-BY-LETTER READERS." Master's thesis, Temple University Libraries, 2019. http://cdm16002.contentdm.oclc.org/cdm/ref/collection/p245801coll10/id/574233.
Full textAbstract:
Public HealthM.A.
This study investigated the benefits of rapid presentation of written words as a treatment strategy to enhance reading speed and accuracy in two participants with acquired alexia who are letter-by-letter readers. Previous studies of pure alexia have shown that when words are rapidly presented, participants can accurately perform lexical decision and category judgment tasks, yet they are unable to read words aloud. These studies suggest that rapid presentation of words could be used as a treatment technique to promote whole-word reading. It was predicted that treatment utilizing rapid presentation (250/500 ms) will increase reading speed and accuracy of both trained and untrained words compared to the words trained in standard presentation (5000 ms). A single-subject ABACA/ACABA multiple baseline treatment design was used. Treatment was provided twice per week for four weeks for both rapid and standard presentation treatment. Each session comprised a spoken-to-written word decision task and semantic category judgment task. Stimuli included 80 trained words divided between the two treatments and 20 untrained controls. Weekly probes to assess reading accuracy were administered after every two treatment sessions. Based on effect sizes, results showed no consistent unambiguous benefit for rapid or standard presentation treatment. However, possible generalization to untrained words due to rapid presentation treatment was observed. Future research is warranted to investigate the effectiveness of rapid presentation treatment in letter-by-letter readers.
Temple University--Theses
6
Munnecom, Lorenna, and Miguel Chaves de Lemos Pacheco. "Exploration of an Automated Motivation Letter Scoring System to Emulate Human Judgement." Thesis, Högskolan Dalarna, Mikrodataanalys, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:du-34563.
Full textAbstract:
As the popularity of the master’s in data science at Dalarna University increases, so does the number of applicants. The aim of this thesis was to explore different approaches to provide an automated motivation letter scoring system which could emulate the human judgement and automate the process of candidate selection. Several steps such as image processing and text processing were required to enable the authors to retrieve numerous features which could lead to the identification of the factors graded by the program managers. Grammatical based features and Advanced textual features were extracted from the motivation letters followed by the application of Topic Modelling methods to extract the probability of each topics occurring within a motivation letter. Furthermore, correlation analysis was applied to quantify the association between the features and the different factors graded by the program managers, followed by Ordinal Logistic Regression and Random Forest to build models with the most impactful variables. Finally, Naïve Bayes Algorithm, Random Forest and Support Vector Machine were used, first for classification and then for prediction purposes. These results were not promising as the factors were not accurately identified. Nevertheless, the authors suspected that the factors may be strongly related to the highlight of specific topics within a motivation letter which can lead to further research.
7
O'Brien, Marita A. "Effects of Shape, Letter Arrangements, and Practice on Text Entry on a Virtual Keyboard." Thesis, Georgia Institute of Technology, 2006. http://hdl.handle.net/1853/11499.
Full textAbstract:
This research study examined the design of a virtual keyboard that can be used for text entry with a rotary controller, particularly when users may differ in age and experience with a particular system. I specifically examined the shape and letter arrangement on the virtual keyboard to help determine the best features to use in a design. Two keyboard shapes, an Oval and a Plus, were selected to represent different aspects of the shape. Two keyboard arrangements, Alphabetic and a Standard QWERTY-based ordering, were selected to represent a well-known and less familiar arrangement. In the experiment, older and younger adults entered words over two consecutive days. Most of the time, they used either the Oval or the Plus, but they also used the alternate shape at specific points during their practice session to allow assessment of their ability to transfer what they had learned. At the end of the second day, they also used a variation of the practiced arrangement to examine how well they had learned the letter arrangement.
Text entry performance on both shapes improved as a function of practice, demonstrating that participants could learn even unfamiliar devices and virtual keyboards to complete a word entry task. No overall shape effects were found for any level of performance, but shape did affect how participants learned and performed the word entry task. In particular, unique visual features on a shape may facilitate memorization of letter/visual cue mappings. These shape features are particularly important for older adults, as younger adults seem to develop a mental model that helps them memorize letter locations on either shape. With practice, older adults could achieve optimal performance levels with an Alphabetic keyboard on the Plus shape that has the more visually unique corners. In general, alphabetic ordering is best not only because it helped visual search, but also because it facilitated better movement planning. Overall, designers should consider creating unique visual features on a virtual keyboard that will blend with the compatibility and allowed movements for the selected device to create an effective virtual keyboard.
8
Chang, Ya-Ning. "The role of visual processing in computational models of reading." Thesis, University of Manchester, 2012. https://www.research.manchester.ac.uk/portal/en/theses/the-role-of-visual-processing-in-computational-models-of-reading(86ca64ef-b471-4932-a891-53728301ff1c).html.
Full textAbstract:
Visual processing is the earliest core process required to support a normal reading system. However, little attention has been given to its role in any of the existing cognitive/computational models of reading. The ultimate goal of this thesis is to create a large-scale model of reading, which can generate phonology and semantics from print. Building such a model will allow for the exploration of a number of theoretically important cognitive phenomena in both normal and impaired reading including: font and size invariance; letter confusability; length effects; and pure alexic reading patterns. To achieve this goal, there are a number of important sub-goals that need to be achieved: (1) to develop a visual processing component which is capable of recognising letters in different fonts and sizes; (2) to produce a model that can develop useful intermediate (orthographic) representations as a consequence of learning; (3) to develop a set of semantic representations compact enough to allow efficient learning but that can still capture realistic semantic similarity relationships; (4) to integrate all the components together into a large-scale recurrent reading model; and (5) to extend the model to support picture naming, and to explore whether damage to the visual system can produce symptoms similar to those found in PA patients. Chapter 2 started by developing a simple feedforward network for letter recognition. The model was trained with letters in various transformations, which allowed the model to learn to deal with size and shape invariance problems as well as accounting for letter confusability effects and generalising to previously unseen letters. The model achieved this by extracting key features from visual input which could be used to support accurate letter recognition. Chapter 3 incorporated the letter recognition component developed in Chapter 2 into a word reading model. The reading model was trained on the mappings between print and phonology, with the orthographic representations which learn to emerge over training. The model could support accurate nonword naming and simulated the length by lexicality interaction observed in normal reading. A system of semantic representations was developed in Chapter 4 by using co-occurrence statistics to generate semantic codes that preserved realistic similarity relationships. Chapter 5 integrated all the components developed in the previous chapters together into a large-scale recurrent reading model. Finally, Chapter 6 extended the reading model to perform object recognition along with the reading task. When the model's visual system was damaged it was able to simulate the abnormal length effect typically seen in PA patients. The damaged model also showed impaired reaction times in object naming and preserved sensitivity to lexical/semantic variables in reading. The picture naming performance was modulated by visual complexity. In summary, the results highlight the importance of incorporating visual information into computational models of single word reading, and suggest that doing so will enable the exploration of a wide range of effects that were previously inaccessible to these types of connectionist models.
9
Mercado, Salazar Jorge Anibal, and S. M. Masud Rana. "A Confirmatory Analysis for Automating the Evaluation of Motivation Letters to Emulate Human Judgment." Thesis, Högskolan Dalarna, Institutionen för information och teknik, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:du-37469.
Full textAbstract:
Manually reading, evaluating, and scoring motivation letters as part of the admissions process is a time-consuming and tedious task for Dalarna University's program managers. An automated scoring system would provide them with relief as well as the ability to make much faster decisions when selecting applicants for admission. The aim of this thesis was to analyse current human judgment and attempt to emulate it using machine learning techniques. We used various topic modelling methods, such as Latent Dirichlet Allocation and Non-Negative Matrix Factorization, to find the most interpretable topics, build a bridge between topics and human-defined factors, and finally evaluate model performance by predicting scoring values and finding accuracy using logistic regression, discriminant analysis, and other classification algorithms. Despite the fact that we were able to discover the meaning of almost all human factors on our own, the topic models' accuracy in predicting overall score was unexpectedly low. Setting a threshold on overall score to select applicants for admission yielded a good overall accuracy result, but did not yield a good consistent precision or recall score. During our investigation, we attempted to determine the possible causes of these unexpected results and discovered that not only is topic modelling limitation to blame, but human bias also plays a role.
10
Jansche, Martin. "Inference of string mappings for speech technology." Columbus, Ohio : Ohio State University, 2003. http://rave.ohiolink.edu/etdc/view?acc%5Fnum=osu1061209163.
Full textAbstract:
Thesis (Ph. D.)--Ohio State University, 2003.Title from first page of PDF file. Document formatted into pages; contains xv, 268 p.; also includes graphics. Includes abstract and vita. Advisor: Chris Brew, Dept. of Linguistics. Includes bibliographical references (p. 252-266) and index.
11
Ramberg, Linnea, and Emmy Lindberg. "Analoga och digitala skrivverktyg : en jämförande studie av elevers texter." Thesis, Linnéuniversitetet, Institutionen för svenska språket (SV), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-101162.
Full textAbstract:
Studiens syfte är att jämföra hur elevers narrativa texter i årskurs 6 skiljer sig åt beroende på om de skrivs med hjälp av analoga skrivverktyg eller digitala skrivverktyg med ordbehandlingsprogram. Studien omfattar åtta elevtexter som analyserats utifrån aspekterna stavning, särskrivning samt felanvändning av versaler och gemener. En teoretisk utgångspunkt som studien utgått från är det sociokulturella perspektivets tankesätt om medierande redskap, där dessa redskap i vår studie är penna och tangentbord. Den andra teoretiska utgångspunkten är kognitivismen med fokus på kognitiva och sensomotoriska processers betydelse för elevers textskrivande. Resultatet visade att eleverna hade betydligt fler fel i sina analoga texter jämfört med de digitala gällande stavfel och felanvändning av versaler och gemener men däremot utmärkte sig resultatet för särskrivning eftersom inga skillnader blev framträdande oavsett vilket skrivverktyg som användes. Som förväntat presterade eleverna bättre i sina digitala texter där mycket går att koppla till det ordbehandlingsprogram som eleverna fick hjälp av. En annan viktig iakttagelse är att de analoga texterna innehöll oroväckande mycket fel vilket kan innebära att analoga texter skrivs allt mer sällan och att elever därför går miste om viktiga förmågor kopplat till de kognitiva och sensomotoriska processerna. Med andra ord visar resultatet att valet av medierande redskap har betydelse för elevernas skrivande.
12
Ghidotti, Piovan Alessandro. "Sviluppo di tecniche di signal processing real-time per la lettura di codici a barre." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2012. http://amslaurea.unibo.it/4252/.
Full text
13
DE, LUCA LETIZIA. "LETTURA PSICODINAMICA DEI PROCESSI DI TRANSIZIONE ENTRO IL SISTEMA PENITENZIARIO ITALIANO." Doctoral thesis, Università Cattolica del Sacro Cuore, 2017. http://hdl.handle.net/10280/19979.
Full textAbstract:
La ricerca si è posta l’obiettivo di delineare, esplorare ed approfondire le rappresentazioni, i vissuti e le strategie operative messe in atto dagli operatori penitenziari italiani in risposta al cambiamento organizzativo in atto nel contesto nazionale. Tale scelta ha previsto la partizione del progetto in due studi.
Il primo studio ha fatto riferimento alla metodologia Grounded Theory (Glaser & Strauss, 1967) con riferimento al contributo di Charmaz (2006). Sono stati coinvolti 123 operatori penitenziari di 4 regioni italiane. I risultati hanno evidenziato importanti elementi di similarità tra le figure professionali, nonostante la differenza di ruolo. Il più importante elemento di trasversalità emerge rispetto ai problemi percepiti come essenziali nell’attuale pratica di lavoro. Infine, elementi di omogeneità caratterizzano anche i vissuti riferiti, a prevalenza emozionale negativa.
Il secondo studio si è ispirato alla tradizione dell’analisi di processo. I dati raccolti i tramite osservazione partecipata sono stati riletti secondo due metodologie: analisi del contenuto tramite T-LAB (Lancia, 2004) ed analisi delle modalità interattive tramite applicazione della griglia IPA (Bales, 1950; 1970). I risultati mostrano una generale tendenza a soffermarsi su aspetti legati al compito, tralasciando però la dimensione relazionale e la centralità di momenti di auto-riflessione rispetto al processo in atto.The research had the aim of outlining, exploring and deepening representations, experiences and operational strategies implemented by Italian prison operators in response to the organizational change taking place in the national context. This choice has foreseen the partition of the project in two studies. The first study referred to Grounded Theory methodology (Glaser & Strauss, 1967; Charmaz, 2006). There were involved 123 operators coming from 4 Italian regions. The results highlight important elements of similarity among professional figures, despite the difference of role. The most important element of transversality emerges about perceived problems. Finally, elements of homogeneity also characterize the reported experiences, with negative emotional prevalence. The second study was inspired by the tradition of process analysis. Data, collected through participatory observations, were interpreted according to two methodologies: content analysis via T-LAB (Lancia, 2004) and analysis of interactive modes through the application of IPA grid (Bales, 1950, 1970). The results show a general tendency to linger on aspects related to the task, but to omit the relational dimension and the centrality of moments of self-reflection about the ongoing process.
14
Zani, Roberto <1974>. "Collegi universitari e formazione del capitale sociale: una lettura relazionale." Doctoral thesis, Alma Mater Studiorum - Università di Bologna, 2007. http://amsdottorato.unibo.it/238/.
Full text
15
Coelho, Rui Pedro Gomes. "The role of deviant-letter position in cognate word processing." Master's thesis, 2015. http://hdl.handle.net/1822/38637.
Full textAbstract:
Dissertação de mestrado integrado em PsicologiaThe Bilingual Interactive Activation Plus Model (BIA+; Dijkstra & van Heuven, 2002) is perhaps the most relevant computational model on bilingual word recognition. Although interesting it fails to explain modulations on cognate word processing as a function of deviantletter position (Font, 2001). Thus, the present research aimed to further explore the role of deviant-letter position on cognate word processing by using a masked priming lexical decision task. To that purpose 288 stimuli (144 Portuguese-English translation words [72 cognates and 72 noncognates] and 144 pseudowords) were selected. Cognates were assigned to three experimental conditions according to deviant-letter position: 1) at end of the word (matriz- MATRIX); 2) at the beginning of the word (coala-KOALA); and 3) at the end and at the beginning (escala-SCALE). Twenty-eight proficient Portuguese-English bilinguals took part in the study. The results revealed faster responses for cognates with greater degree of crosslanguage overlap (Conditions 1 and 2). More important, priming effects were not modulated by deviant-letter position and thus no amendments seems to be needed in the “front-end” of the coding scheme of the BIA+ model, at least regarding cognate processing. Future studies should be developed in order to explore if these results are restricted to outer deviant-letters.
O Bilingual Interactive Activation Plus Model (BIA+; Dijkstra & van Heuven, 2002) é, possivelmente, o modelo bilingue mais relevante acerca do reconhecimento de palavras. Porém, o modelo não é capaz de explicar modulações no processamento de palavras cognatas em função da posição da letra desviante (Font, 2001). A presente investigação procurou explorar o papel desta variável no processamento de palavras cognatas, usando uma tarefa de decisão lexical com um paradigma de priming mascarado. Para tal, foram selecionados 288 estímulos (144 palavras [72 cognatas e 72 nãocognatas] e 144 pseudopalavras). As palavras cognatas foram distribuídas por condições em função da posição da letra desviante: final (e.g., matriz- MATRIX); inicial (e.g., coala-KOALA); inicial e final das palavras (e.g., escala-SCALE). Participaram no estudo 28 bilingues proficientes de Português Europeu-Inglês. Os resultados revelaram respostas mais rápidas para cognatas com maior grau de sobreposição entre línguas (Condição 1 e 2). Adicionalmente, os efeitos de priming não foram modulados pela posição da letra desviante. Portanto, conclui-se que o sistema de codificação do modelo BIA+ encontra-se bem formulado, pelo menos no que diz respeito ao processamento de palavras cognatas. Futuras investigações devem ser conduzidas para explorar se estes resultados são restritos a letras desviantes em posições externas.
16
Al, Dahhan Noor. "Saccadic eye movements and pause/articulation components during a letter naming speed task: Children with and without dyslexia." Thesis, 2013. http://hdl.handle.net/1974/8332.
Full textAbstract:
Naming speed (NS) tasks that measure how quickly and accurately participants can name visual stimuli (e.g., letters) are commonly used to predict reading ability. However, the link between NS and reading is poorly understood. Three methods were used to investigate how NS relates to reading and what cognitive processes are involved: (a) changing stimulus composition to emphasize phonological and/or visual aspects (Compton, 2003); (b) decomposing NS times into pause and articulation components; and (c) analyzing eye movements during a NS task. Participants were in three groups: dyslexics (aged 9, 10), chronological-age (CA) controls (age 9, 10), and reading-level (RL) controls (aged 6, 7). We used a letter NS task and three variants that were either phonologically and/or visually confusing while subjects’ eye movements and articulations were recorded, and examined how these manipulations influenced NS performance and eye movements.
For all groups, NS manipulations were associated with specific patterns of behaviour and saccadic performance, reflecting differential contributions of NS to reading. RL controls were less efficient, made more errors, saccades and regressions, and made longer fixation durations, articulation times, and pause times than CA controls. Dyslexics consistently scored in between controls, except for the number of saccades and regressions in which they made more than both control groups. Overall there were clear developmental changes in NS performance, NS components, and eye movements in controls from ages 6 to 10 that appear to occur more slowly for dyslexics.
Furthermore, pause time and fixation duration were key features in the NS-reading relationship, and increasing visual similarity of the letter matrix had the greatest effect on performance for all subjects. This latter result was demonstrated by the decrease in efficiency and eye-voice span, increase in naming errors, saccades, and regressions, and longer pause times and fixation durations found for all subjects. We conclude that NS is related to reading via fixation durations and pause times; longer fixation durations reflect the greater amount of time needed to acquire visual/orthographic information from stimuli, and longer pause times in children with dyslexia reflect the greater amount of time needed to prepare to respond to stimuli.Thesis (Master, Neuroscience Studies) -- Queen's University, 2013-09-26 12:24:53.951
17
Milivojevic, Branka. "About turn:neural mechanisms underlying visual processing of rotated letters and digits." 2007. http://hdl.handle.net/2292/2433.
Full textAbstract:
This thesis explores neural activity associated with processing of rotated alphanumeric characters, focusing particularly on linear and quadratic trend components of orientation-dependent activity. Choice of these components was driven by results of reaction-time (RT) studies; judging whether characters are normal or backward (parity task) typically elicit RTs that are linearly related to character disorientation, implying mental rotation of the characters to the upright, while judging whether they are letters or digits (categorisation task) elicits RTs related nonlinearly to disorientation, combining both linear and quadratic component, but not indicative of mental rotation. In Experiment 1 neural activity was monitored using fMRI while participants performed these tasks. In the next two experiments, neural processing was monitored with high-density EEG. In Experiment 2 participants performed the same two tasks, while in Experiment 3 they performed the category task and red-blue colour judgements. In Experiment 1, linear increases in fMRI activation were elicited only by the parity task and were observed in the posterior portion of the dorsal intraparietal sulcus and lateral and medial pre-supplementary motor areas, suggesting a fronto-parietal network underlying mental rotation. Experiment 2 showed that linear increases in parietal negativity between 350 and 710 ms only evident in the parity task, again indicating that mental rotation is only elicited by that task. Contrary to previous evidence, Experiment 2 indicated that both hemispheres may be involved in mental rotation, but rotation is faster in the right hemisphere than in the left hemisphere. Experiment 1 also showed that effects of orientation common to both tasks were best characterised by a quadratic trend, and were restricted to the supramarginal gyrus. This activation was interpreted as representing orientation-dependent shape recognition. Experiments 2 and 3 also revealed orientation-dependent neural activity at three distinct stages prior to mental rotation. First, on the P1 component, there was a difference between oblique and vertical orientations, suggesting the extraction of orientation based on axis of elongation. Next, orientation affected the N1 component, with longer latencies and larger amplitudes with misorientation, and smaller effects for inversion than for intermediate angular rotations. Finally, changes in orientation affected the P2 component differently for the parity and category tasks, probably signalling the perception of orientation relative to a parity-defined memory representation, and serving as a preparation for mental rotation. These experiments identify both the orientation-specific neural processing that occurs prior to the onset of mental rotation, and the subsequent neural correlates of mental rotation itself.Top Achiever Doctoral Scholarship, University of Auckland Doctoral Scholarship, The New Zealand Neurological Foundation, University of Auckland Research Fund (Project numbers: 3607199, 3605876 3604420)
18
Ferreira, Filipa Freire. "How tone is used across CEO letters : the impact of financial crisis evidence from the UK." Master's thesis, 2021. http://hdl.handle.net/10400.14/36870.
Full textAbstract:
We study the tone present in CEO Letters of firms listed in London Stock Exchange (LSE), more specifically, by defining two alternative approaches to assess tone: the CEO Letter as a whole and the weighted approach, where the percentages chosen for Sections 1, 2 and 3 are, 10%, 80% and 10%, respectively. By differentiating these two models, we intend to analyse if differences exist between the whole text tone and the tone present in each section. Secondly, we aim to understand if there are differences in the tone means within sections for the same CEO Letter. Additionally, we study the impact of financial crisis by analysing the changes in some textual features of CEO Letters, more specifically, tone, uncertainty, and causality. Furthermore, we also analyse if there is a significant change in the relationship between firm’s current performance and its reported tone in crisis versus non crisis settings that might help us conclude which scenario is more likely for firms to engage in impression management. By using several linear regression models, we conduct an analysis of the textual features in CEO Letters as a whole and in its sections, individually. There are indeed differences between the whole text tone and the tone of each section and, differences in the tone means within sections. Tone is explained by the financial crisis. While uncertainty words increase with Financial Crisis, we do not find any relevant link in the use of causal words. Regarding current firm’s performance and tone relationship, we do not observe any significant change in this relationship from crisis to non-crisis years.Estudamos o tom presente nas cartas dos CEOs de empresas cotadas na London Stock Exchange (LSE), mais especificamente, definindo dois modelos alternativos para medir o tom: o tom da carta como um todo e o tom da carta dividida em secções, sendo que a percentagem escolhida para as secções 1, 2 e 3 da carta são, respetivamente, 10%, 80% e 10%. Com esta diferenciação de modelos, pretendemos analisar se existem diferenças entre o tom do texto como um todo e o tom presente em cada uma das secções. Numa segunda instância, pretendemos analisar se existem diferenças nas médias do tom entre as três secções da carta. Adicionalmente, estudamos o impacto da crise financeira de 2008 através da análise de mudanças de certas caraterísticas textuais das cartas dos CEOs, nomeadamente, o tom, a incerteza e a causalidade. Para além disto, analisamos ainda se existe uma diferença na relação entre a performance atual de uma empresa e o tom reportado pela mesma em anos de crise face a anos de não crise que possa indiciar qual o tipo de contexto económico mais propício à adoção de técnicas de impression management. Usando vários modelos de regressão linear, realizamos uma análise das caraterísticas textuais na carta como um todo e em cada uma das suas secções. Existem diferenças entre as secções entre si e entre o tom das secções e o tom da carta como um todo. O tom é explicado pela crise. Embora as palavras de incerteza aumentem durante a crise, o mesmo não acontece para as palavras causais. Quanto à relação entre a performance atual da empresa e o tom reportado pela mesma, não existe nenhuma alteração significante entre anos de crise e não crise.