Linguistic styles: language use as an individual difference

Traumatic memories of childbirth relate to maternal postpartum posttraumatic stress disorder

The trauma memory is a crucial feature of PTSD etiology and maintenance. Nonetheless, the nature of memories associated with childbirth-related posttraumatic stress disorder (CB-PTSD) requires explication. The present study, as part of a larger project on psychological outcomes of childbirth, utilized a multi-method approach to characterize childbirth memories in relation to CB-PTSD symptoms. We here assessed 413 women who completed self-report measures pertaining to CB-PTSD, postpartum depression, and childbirth memories. Additionally, a subset of 209 women provided written childbirth narratives, analyzed using Linguistic Inquiry and Word Count software. Women endorsing CB-PTSD symptoms on the PTSD-Checklist (PCL)-5 reported more incoherent childbirth memories with more emotional and sensory details, and more frequent involuntary recall and reliving of the memory. They also indicated the childbirth experience was more central to their identity. Written narratives in those with probable CB-PTSD were characterized by less (positive) affective processes, and more cognitive processes. We infer that childbirth memories in women who endorse symptoms of CB-PTSD in the early postpartum period resemble those described in the general PTSD literature. This suggests that childbirth may be experienced as traumatic and evoke a traumatic memory, implicated in symptom endorsement. Opportunities for therapeutic interventions modifying traumatic memories of childbirth in women at risk for CB-PTSD need to be investigated. Future research examining characteristics of traumatic childbirth memories is needed to advance our understanding of this overlooked postpartum condition.

Small Words That Matter: Linguistic Style and Conceptual Disorganization in Untreated First-Episode Schizophrenia

This study aimed to shed light on the linguistic style affecting the communication discourse in first-episode schizophrenia (FES) by investigating the analytic thinking index in relation to clinical scores of conceptual and thought disorganization (Positive and Negative Syndrome Scale, PANSS-P2 and Thought and Language Index, TLI). Using robust Bayesian modeling, we report three major findings: (1) FES subjects showed reduced analytic thinking, exhibiting a less categorical linguistic style than healthy control (HC) subjects (Bayes factor, BF10 > 1000), despite using the same proportion of function and content words as HCs; (2) the lower the analytic thinking score, the higher the symptoms scores of conceptual disorganization (PANSS-P2, BF = 22.66) and global disorganization of thinking (TLI, BF10 = 112.73); (3) the linguistic style is a better predictor of conceptual disorganization than the cognitive measure of processing speed in schizophrenia (SZ). These findings provide an objectively detectable linguistic style with a focus on Natural Language Processing Analytics of transcribed speech samples of patients with SZ that require no clinical judgment. These findings also offer a crucial insight into the primacy of linguistic structural disruption in clinically ascertained disorganized thinking in SZ. Our work contributes to an emerging body of literature on the psychopathology of SZ using a first-order lexeme-level analysis and a hypothesis-driven approach. At a utilitarian level, this has implications for improving educational and social outcomes in patients with SZ.

Does It Matter What I Say? Using Language to Examine Reactions to Ostracism as It Occurs

Most of our knowledge related to how social exclusion affects those who ostracize and those who are being ostracized is based on questionnaires administered after the ostracism situation is over. In this research, we strived to further our understanding of the internal dynamics of an ostracism situation. We therefore examined individuals' language-specifically, function words-as a behavior indicative of psychological processes and emergent states that can be unobtrusively recorded right in the situation. In online chats, 128 participants talked about a personal topic in groups of three. In the experimental group (n = 79), two conversation partners ignored every contribution by the third. We found that, compared to the control group, these targets of ostracism used language indicative of a self-focus and worsened mood, but not of social focus or positivity, although positivity was related to a writer's likeability. Sources of ostracism used language suggesting that they were distancing themselves from the situation, and they further engaged in victim derogation. We discuss how our results highlight the severity and potential self-sustainability of ostracism.

Fronto-temporal brain activity and connectivity track implicit attention to positive and negative social words in a novel socio-emotional Stroop task

Previous inconsistencies on the effects of implicitly processing positively - vs. negatively - connotated emotional words might reflect the influence of uncontrolled psycholinguistic dimensions, and/or social facets inherent in putative "emotional" stimuli. Based on the relevance of social features in semantic cognition, we developed a socio-emotional Stroop task to assess the influence of social vs. individual (non-social) emotional content, besides negative vs. positive valence, on implicit word processing. The effect of these variables was evaluated in terms of performance and RTs, alongside associated brain activity/connectivity. We matched conditions for several psycholinguistic variables, and assessed a modulation of brain activity/connectivity by trial-wise RT, to characterize the maximum of condition- and subject-specific variability. RTs were tracked by insular and anterior cingulate activations likely reflecting implicit attention to stimuli, interfering with task-performance based on condition-specific processing of their subjective salience. Slower performance for negative than neutral/positive words was tracked by left-hemispheric structures processing negative stimuli and emotions, such as fronto-insular cortex, while the lack of specific activations for positively-connotated words supported their marginal facilitatory effect. The speeding/slowing effects of processing positive/negative individual emotional stimuli were enhanced by social words, reflecting in specific activations of the right anterior temporal and orbitofrontal cortex, respectively. RTs to social positive and negative words modulated connectivity from these regions to fronto-striatal and sensorimotor structures, respectively, likely promoting approach vs. avoidance dispositions shaping their facilitatory vs. inhibitory effect. These results might help assessing the neural correlates of impaired social cognition and emotional regulation, and the effects of rehabilitative interventions.

The potential of game- and video-based assessments for social attributes: examples from practice

Purpose

The purpose of this study is to describe the development and psychometric properties of a novel game- and video-based assessment of social attributes. Despite their increasing adaption, little research is available on the suitability of games and video analytics for measuring noncognitive attributes in the selection context.

Design/methodology/approach

The authors describe three novel assessments and their psychometric properties in a sample of 1,300 participants: a game-based adaptation of an Emotion Recognition Task, a chatbot-based situational judgment test for emotion management and a video-based conscientiousness assessment.

Findings

The novel assessments show good to moderate convergent validity for Emotional Recognition (r = 0.42), Emotion Management (r = 0.39) and Conscientiousness (r = 0.21). The video-based assessment demonstrates preliminary predictive validity for self-reported work performance. Novel game-based assessments (GBAs) are perceived as better designed and more immersive than traditional questionnaires. Adverse impact analysis indicates small group differences by age, gender and ethnicity.

Research limitations/implications

Predictive validity findings need to be replicated using objective measures of performance, such as performance ratings by supervisors and extended to the GBAs. Adverse impact should be evaluated using a real-life applicant pool and extended to additional groups.

Practical implications

Evidence for the psychometric validity of novel assessment formats supports their adoption in selection and recruitment. Improved user experience and shortened assessment times open up new areas of application.

Originality/value

This study gives first insights into psychometric properties of video- and game-based assessments of social attributes.

Developing and validating the self-transcendent emotion dictionary for text analysis

Recent years have seen a growing amount of research effort directed toward what positive media psychologists refer to as self-transcendent emotions, such as awe, admiration, elevation, gratitude, inspiration, and hope. While these emotions are invaluable to promote greater human connectedness, prosociality, and human flourishing, researchers are constrained in terms of analyzing self-transcendent emotions as expressed in spoken and written languages. Drawing upon the word-counting approach of the text analysis paradigm, this project aimed at constructing a dictionary tool-Self-Transcendent Emotion Dictionary (STED)-which can be uploaded into mainstream, text analytic software (e.g., LIWC) to identify and analyze self-transcendent emotions in large corpora. This dictionary tool was then refined and validated via three studies, where individual words were first rated with regard to their fitness into the proposed construct (Step 1), and then used to analyze essays written to reflect the corresponding construct (Step 2). Finally, the refined dictionary was applied to examine words used in nearly 4,000 human-coded New York Times articles (Step 3). Results indicated that the final dictionary, consisting of 351 lexicons and phrases, exhibits acceptable face and construct validity, and possesses a reasonable level of external validity and applicability. Despite its shortcoming in accounting for the rhetorical techniques ingrained in natural human language, the STED could be instrumental for social scientific inquiry of positive emotions in textual narratives.

Natural emotion vocabularies as windows on distress and well-being

To date we know little about natural emotion word repertoires, and whether or how they are associated with emotional functioning. Principles from linguistics suggest that the richness or diversity of individuals' actively used emotion vocabularies may correspond with their typical emotion experiences. The current investigation measures active emotion vocabularies in participant-generated natural speech and examined their relationships to individual differences in mood, personality, and physical and emotional well-being. Study 1 analyzes stream-of-consciousness essays by 1,567 college students. Study 2 analyzes public blogs written by over 35,000 individuals. The studies yield consistent findings that emotion vocabulary richness corresponds broadly with experience. Larger negative emotion vocabularies correlate with more psychological distress and poorer physical health. Larger positive emotion vocabularies correlate with higher well-being and better physical health. Findings support theories linking language use and development with lived experience and may have future clinical implications pending further research.

The narrative arc: Revealing core narrative structures through text analysis

Scholars across disciplines have long debated the existence of a common structure that underlies narratives. Using computer-based language analysis methods, several structural and psychological categories of language were measured across ~40,000 traditional narratives (e.g., novels and movie scripts) and ~20,000 nontraditional narratives (science reporting in newspaper articles, TED talks, and Supreme Court opinions). Across traditional narratives, a consistent underlying story structure emerged that revealed three primary processes: staging, plot progression, and cognitive tension. No evidence emerged to indicate that adherence to normative story structures was related to the popularity of the story. Last, analysis of fact-driven texts revealed structures that differed from story-based narratives.

Head vs. Heart: Which Path do you Take? Self-Location Shapes Language Use, Indicating Rational and Experiential Thinking Styles

Self-location is a novel construct that identifies a bodily organ (head vs. heart) to represent self-concept. However, how self-location influences verbal performance is not well understood. This research investigates language use associated with self-location, an individual difference construct based on two different metaphoric concepts (use your head vs. follow your heart). Study 1 established the associations between self-location and verbal performance assessed by language variables in two writing tasks. Findings showed that self-location was related to specific language variables (nouns vs. verbs). In Study 2a, self-location was presented as a manipulation of the head-heart metaphor to predict language variables in a decision-making story recalled by participants. In Study 2b, a manipulation check was added, and the Heinz dilemma was used as a writing topic to control responses in different conditions. Studies 2a and 2b demonstrate that the metaphor could facilitate use of specific language variables. Implications of manipulating self-location are discussed.

Development and Linguistic Cue Analysis of the State-Trait Cheerfulness Inventory-Short Form

The present study derived a short form of the State-Trait Cheerfulness Inventory-Trait Version (STCI-T30) using an item response theory framework. Latent trait test-retest correlations and reliability across the latent continuum in the STCI-T30 remained high. Moreover, the STCI-T30 showed external validity with criterion variables (e.g., playfulness) and a short writing task completed by these participants was rated by unacquainted judges to infer the author's cheerfulness, seriousness, and bad-mood. Results suggested significant self-other and inter-judge agreement of cheerfulness, seriousness, and bad-mood and linguistic cues analysis suggested cheerfulness and bad-mood manifested through writing in tone, social processes, and affect.

Effects of an Advancing Tenure on CEO Cognitive Complexity

We study how the cognitive complexity of chief executive officers (CEOs) changes during their tenures. Drawing from prior theory and research, we argue that CEOs attain gradually greater role-specific knowledge, or expertise, as their tenures advance, which yields more complex thinking. Beyond examining the main effect of CEO tenure on cognitive complexity, we consider three moderators of this relationship, each of which is expected to influence the accumulation of expertise over a CEO’s time in office: industry dynamism, industry jolts, and CEO positional power. We conduct our tests on a sample of 684 CEOs of public corporations. The analytic centerpiece of our study is a novel index of CEO cognitive complexity based on CEOs’ language patterns in the question-and-answer portions of quarterly conference calls. As part of our extensive theory of measurement, we provide evidence of the reliability and validity of our index. Our results indicate that CEOs, in general, experience substantial increases in cognitive complexity over their time in office. Examined moderators somewhat, but modestly, alter this general trajectory, and nonlinearities are not observed. We discuss the implications of our findings.

Reviews Left and Right: The Link Between Reviewers’ Political Ideology and Online Review Language

Online reviews, i.e., evaluations of products and services posted on websites, are ubiquitous. Prior research observed substantial variance in the language of such online reviews and linked it to downstream consequences like perceived helpfulness. However, the understanding of why the language of reviews varies is limited. This is problematic because it might have vital implications for the design of IT systems and user interactions. To improve the understanding of online review language, the paper proposes that consumers’ personality, as reflected in their political ideology, is a predictor of such online review language. Specifically, it is hypothesized that reviewers’ political ideology as measured by degree of conservatism on a liberal–conservative spectrum is negatively related to review depth (the number of words and the number of arguments in a review), cognitively complex language in reviews, diversity of arguments, and positive valence in language. Support for these hypotheses is obtained through the analysis of a unique dataset that links a sample of online reviews to reviewers’ political ideology as inferred from their online news consumption recorded in clickstream data.

An Adjective Selection Personality Assessment Method Using Gradient Boosting Machine Learning

Goldberg’s 100 Unipolar Markers remains one of the most popular ways to measure personality traits, in particular, the Big Five. An important reduction was later preformed by Saucier, using a sub-set of 40 markers. Both assessments are performed by presenting a set of markers, or adjectives, to the subject, requesting him to quantify each marker using a 9-point rating scale. Consequently, the goal of this study is to conduct experiments and propose a shorter alternative where the subject is only required to identify which adjectives describe him the most. Hence, a web platform was developed for data collection, requesting subjects to rate each adjective and select those describing him the most. Based on a Gradient Boosting approach, two distinct Machine Learning architectures were conceived, tuned and evaluated. The first makes use of regressors to provide an exact score of the Big Five while the second uses classifiers to provide a binned output. As input, both receive the one-hot encoded selection of adjectives. Both architectures performed well. The first is able to quantify the Big Five with an approximate error of 5 units of measure, while the second shows a micro-averaged f1-score of 83%. Since all adjectives are used to compute all traits, models are able to harness inter-trait relationships, being possible to further reduce the set of adjectives by removing those that have smaller importance.

Mapping breast cancer survivors' psychosocial coping along disease trajectory: A language approach

This article reports a language analysis of breast cancer patients' posts in an online support group. Adopting web-scraping techniques, the study analyzed 27,078 online posts contributed by 1443 users along multiple linguistic dimensions to investigate the trajectory of the patients' psychosocial adaptation of the disease. The findings suggested that breast cancer patients' emotional experiences and adjustment in the course of illness vary from one stage to another. They reached the peak of emotional expression, struggle and despair, and self-focus at Stage III, whereas wiped out negative emotions and signaled a desire for connections with others at Stage IV.

The Effect of User Psychology on the Content of Social Media Posts: Originality and Transitions Matter

Multiple studies suggest that frequencies of affective words in social media text are associated with the user's personality and mental health. In this study, we re-examine these associations by looking at the transition patterns of affect. We analyzed the content originality and affect polarity of 4,086 posts from 70 adult Facebook users contributed over 2 months. We studied posting behavior, including silent periods when the user does not post any content. Our results show that more extroverted participants tend to post positive content continuously and that more agreeable participants tend to avoid posting negative content. We also observe that participants with stronger depression symptoms posted more non-original content. We recommend that transitions of affect pattern derived from social media text and content originality should be considered in further studies on mental health, personality, and social media.

Multilingual Transformer-Based Personality Traits Estimation

Intelligent agents have the potential to understand personality traits of human beings because of their every day interaction with us. The assessment of our psychological traits is a useful tool when we require them to simulate empathy. Since the creation of social media platforms, numerous studies dealt with measuring personality traits by gathering users’ information from their social media profiles. Real world applications showed how natural language processing combined with supervised machine learning algorithms are effective in this field. These applications have some limitations such as focusing on English text only and not considering polysemy in text. In this paper, we propose a multilingual model that handles polysemy by analyzing sentences as a semantic ensemble of interconnected words. The proposed approach processes Facebook posts from the myPersonality dataset and it turns them into a high-dimensional array of features, which are then exploited by a deep neural network architecture based on transformer to perform regression. We prove the effectiveness of our work by comparing the mean squared error of our model with existing baselines and the Kullback–Leibler divergence between the relative data distributions. We obtained state-of-the-art results in personality traits estimation from social media posts for all five personality traits.

Linguistic Predictors of Problematic Drinking in Alcohol-related Facebook Posts

Emerging adults often increase problematic drinking during college. Although they generally do not seek help for problematic drinking, college students discuss their drinking on social media. This study followed college students' Facebook profiles from the inception of their attendance at a university and identified alcohol-related posts. Within 28 days of their first alcohol-related Facebook post, participants were interviewed to assess problematic drinking (binge drinking episodes and number of drinks). Linguistic analysis of alcohol-related Facebook posts found that use of negative emotion language and swear words were related to problematic drinking, in support of proposed hypotheses. Results are situated within alcohol use disorder and health research examining the link between problematic drinking and anxiety, deviant behavior, and negative emotions.

Personality Traits and Emotional Word Recognition: An ERP Study

Recent research has investigated how personality trait differences influence the processing of emotion conveyed by pictures, but limited research has examined the emotion conveyed by words. The present study investigated whether extraversion (extroverts vs. introverts) and neuroticism (high neurotics vs. low neurotics) influence the processing of positive, neutral, and negative words that were matched for arousal. Electroencephalography (EEG) was recorded from healthy participants while they performed a lexical decision task. We found that personality traits influenced emotional word recognition at N400 (300-450 ms) and LPC (450-800 ms). At the earlier (N400) stage, the more extraverted and neurotic a participant was, the more reduced the N400s for the positive words relative to neutral words were. This suggests that the extroverts and high neurotics (i.e., high impulsivity) identified positive content in words during lexical feature retrieval, which facilitated such retrieval. At the later (LPC) stage, both the introverts and high neurotics (i.e., high anxiety) showed greater LPCs to negative than neutral words, indicating their sustained attention and elaborative processing of negative information. These results suggest that extraversion and neuroticism collectively influence different stages of emotional word recognition in a way that is consistent with Gray's biopsychological theory of personality.

Can linguistic analysis be used to identify whether adolescents with a chronic illness are depressed?

Comorbid depression is common in adolescents with chronic illness. We aimed to design and test a linguistic coding scheme for identifying depression in adolescents with chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME), by exploring features of e-consultations within online cognitive behavioural therapy treatment. E-consultations of 16 adolescents (aged 11-17) receiving FITNET-NHS (Fatigue in teenagers on the interNET in the National Health Service) treatment in a national randomized controlled trial were examined. A theoretically driven linguistic coding scheme was developed and used to categorize comorbid depression in e-consultations using computerized content analysis. Linguistic coding scheme categorization was subsequently compared with classification of depression using the Revised Children's Anxiety and Depression Scale published cut-offs (t-scores ≥65, ≥70). Extra linguistic elements identified deductively and inductively were compared with self-reported depressive symptoms after unblinding. The linguistic coding scheme categorized three (19%) of our sample consistently with self-report assessment. Of all 12 identified linguistic features, differences in language use by categorization of self-report assessment were found for "past focus" words (mean rank frequencies: 1.50 for no depression, 5.50 for possible depression, and 10.70 for probable depression; p < .05) and "discrepancy" words (mean rank frequencies: 16.00 for no depression, 11.20 for possible depression, and 6.40 for probable depression; p < .05). The linguistic coding profile developed as a potential tool to support clinicians in identifying comorbid depression in e-consultations showed poor value in this sample of adolescents with CFS/ME. Some promising linguistic features were identified, warranting further research with larger samples.

Gender-based differences in letters of recommendation written for ophthalmology residency applicants

BACKGROUND

To determine whether gender-based differences may be present in letters of recommendation written for ophthalmology residency applicants.

METHODS

All applications submitted through SF Match to the UCLA Stein Eye Institute Residency Training Program from the 2017-2018 application cycle were analyzed using validated text analysis software (Linguistic Inquiry and Word Count (Austin, TX)). The main outcome measures were differences in language use in letters of recommendation by gender of applicant.

RESULTS

Of 440 applicants, 254 (58%) were male and 186 (42%) were female. The two gender groups had similar United States Medical Licensing Exam (USMLE) Step 1 scores, undergraduate grade point averages (uGPA's), proportions of underrepresented minority (URM) applicants and Gold Humanism Honor Society members, numbers of academic and service activities listed, and gender distributions of their letter writers (all P values > 0.05). However, letters written for male applicants were determined to use more "authentic" words than those written for female applicants (mean difference, 0.800; 95% CI, 0.001-1.590; P = 0.047). Letters written for male applicants also contained more "leisure" words (mean difference, 0.056; 95% CI, 0.008-0.104; P = 0.023) and fewer "feel" words (mean difference, 0.033; 95% CI, 0.001-0.065; P = 0.041) and "biological processes" words (mean difference, 0.157; 95% CI, 0.017-0.297; P = 0.028).

CONCLUSIONS

There were gender differences detected in recommendation letters in ophthalmology consistent with prior studies from other fields. Awareness of these differences may improve residency selection processes.

Language-Style Similarity and Social Networks

This research demonstrates that linguistic similarity predicts network-tie formation and that friends exhibit linguistic convergence over time. In Study 1, we analyzed the linguistic styles and the emerging social network of a complete cohort of 285 students. In Study 2, we analyzed a large-scale data set of online reviews. In both studies, we collected data in two waves to examine changes in both social networks and linguistic styles. Using the Linguistic Inquiry and Word Count (LIWC) framework, we analyzed the text of students' essays and of 1.7 million reviews by 159,651 Yelp reviewers. Consistent with our theory, results showed that similarity in linguistic style corresponded to a higher likelihood of friendship formation and persistence and that friendship ties, in turn, corresponded to a convergence in linguistic style. We discuss the implications of the coevolution of linguistic styles and social networks, which contribute to the formation of relational echo chambers.

Individual predictors of tactical planning performance in junior military officers

This exploratory study examined the influence of personal and psycholinguistic factors on tactical planning performance in a convenience sample of Australian Army personnel. Sixty male junior officers undertaking a tactical training course completed self-report measures of decision-making style, problem-solving appraisal, personality, and tactical planning expertise. A measure of general cognitive ability was sourced from participants' military records. During the course, officers completed several tactical planning exercises that were assessed by course instructors, and an overall measure of tactical planning performance was derived. Psycholinguistic data was derived from participants' thought processes elicited during an analogous planning exercise using a written think-aloud protocol. Automatic linear modeling analyses identified the combination of greater expertise and intuitive decision-making style, less spontaneous decision-making style and problem-solving confidence, younger age, and greater use of words relating to risk and past-focused concepts in their thought records, as significant predictors of better tactical planning, accounting for 55% of the variance in performance. Additionally, general cognitive ability, conscientiousness, and rational decision-making style shared moderate positive bivariate correlations with planning performance. This research has identified individual factors that contribute to tactical mission planning, along with future research avenues, which may inform the training of military commanders at the tactical level.

Why health professions education needs functional linguistics: the power of 'stealth words'

CONTEXT

Language is one of the primary modalities for teaching and learning in the health professions in contexts ranging from the more formal teaching relationships of medical school to the guided practice of trainees through continuing education and the deliberate practice of lifelong learning. Yet linguistic analysis, with the possible exception of discourse analysis, has not become a core methodological tool in the field of health professions education (HPE). The purpose of this paper is to argue for the more widespread adoption of one particular approach to linguistics, one that examines less of what learners and instructors say and looks more at how they say it: functional linguistics.

FUNCTIONAL LINGUISTICS

THE POWER OF 'STEALTH WORDS': This approach theorises and structures the functions of language, regularly focusing attention on 'stealth words' such as I, but and was. Drawing on a rich body of literature in linguistics, psychology, the learning sciences and some early work in HPE, we demonstrate how functional linguistic tools can be applied to better understand learners' and instructors' beliefs, reasoning processes, values and emotions.

FUNCTIONAL LINGUISTICS AND REFLECTION

AN APPLICATION OF STEALTH WORDS: A brief qualitative analysis of one tool - analysis of the generic use of 'you' to mean 'one' or 'anyone' - demonstrates how functional linguistics can offer insight into physicians' bids for credibility and alignment as they think aloud about their clinical reasoning.

FUNCTIONAL LINGUISTICS AND HPE

FUTURE DIRECTIONS: Finally, we offer suggestions for how functional linguistic tools might address questions and gaps in four active research areas in HPE: reflection; emotion and reasoning; learning in simulated contexts, and self-regulated learning.

CONCLUSIONS

We argue that the words used by learners, instructors and practitioners in the health professions as they move through undergraduate and graduate training into practice can offer clues that will help researchers, instructors and colleagues to better support them.

The Development of the Chinese Sentiment Lexicon for Internet

This paper examines the development of the Chinese Sentiment Lexicon for Internet (CSLI), a sentiment lexicon for capturing the valence and arousal in Chinese online social media texts. We first review the current sentiment lexicons and their building process, including the collection of words, judging the emotionality of words, and testing reliability and validity. In Study 1, we develop CSLI and test its initial reliability and validity. In Study 2, we further test the convergent validity of CSLI by examining its correlations with human judgment in 429 aggregated Weibo comments. In Study 3, the predictive validity of CSLI is examined by linking its results to personality traits among 52 undergraduates. Two replication studies are also conducted to verify the findings in Study 2 and 3. The results have generally supported the reliability and validity of CSLI. Therefore, CSLI can be used as a research tool to capture the degree of valence and arousal in Chinese online social media texts. Its potential to promote human well-being is also discussed.

Predicting future mental illness from social media: A big-data approach

In the present research, we investigated whether people's everyday language contains sufficient signal to predict the future occurrence of mental illness. Language samples were collected from the social media website Reddit, drawing on posts to discussion groups focusing on different kinds of mental illness (clinical subreddits), as well as on posts to discussion groups focusing on nonmental health topics (nonclinical subreddits). As expected, words drawn from the clinical subreddits could be used to distinguish several kinds of mental illness (ADHD, anxiety, bipolar disorder, and depression). Interestingly, words drawn from the nonclinical subreddits (e.g., travel, cooking, cars) could also be used to distinguish different categories of mental illness, implying that the impact of mental illness spills over into topics unrelated to mental illness. Most importantly, words derived from the nonclinical subreddits predicted future postings to clinical subreddits, implying that everyday language contains signal about the likelihood of future mental illness, possibly before people are aware of their mental health condition. Finally, whereas models trained on clinical subreddits learned to focus on words indicating disorder-specific symptoms, models trained to predict future mental illness learned to focus on words indicating life stress, suggesting that kinds of features that are predictive of mental illness may change over time. Implications for the underlying causes of mental illness are discussed.

A linguistic study of Chinese Weibo users who lost their only child

The loss of an only child is one of the most painful life events and creates tremendous change in its parents' lives. Analyzing parents' online language use may help to better understand their loss, especially their psychological process. This study compared the online word use of 228 lost-only-child (LOC) parents to that of their peers. We also tracked the change in word use for a subset of these parents (n = 36) quarterly during the first 2 years following their bereavement. The implications of the word use of Chinese LOC parents for mood, parent-child bond, and lifestyle are then discussed.

How to tell a happy person: Accuracy of subjective well-being perception from texts

Although perceptions of subjective well-being (SWB) in unacquainted others have been shown to play a major role in impression formation, little is known about how accurate such perceptions are. In two original studies and one pre-registered replication, we explored the accuracy of life satisfaction and happiness judgments from texts and its underlying mechanisms (use of linguistic cues). Participants filled in life satisfaction and happiness measures and completed a brief writing task. Another sample of participants judged the targets’ life satisfaction and happiness from the obtained texts. All three studies demonstrated a small to moderate self-other agreement. A linguistic analysis showed that targets with higher (vs. lower) scores on SWB were less likely to use negation words in their texts, which allowed observers to make accurate judgment of their SWB level. Two studies pointed at negative emotion words as valid and positive emotion words as invalid (but often used) cues to happiness, yet these effects did not replicate in Study 3.

Applying Natural Language Processing to Evaluate News Media Coverage of Bullying and Cyberbullying

Bullying events have frequently been the focus of coverage by news media, including news stories about teens whose death from suicide was attributed to cyberbullying. Previous work has shown that news media coverage is influential to readers in areas such as suicide, infectious disease outbreaks, and tobacco use. News media may be an untapped resource to promote bullying prevention messages, though current news media approaches to describing bullying and cyberbullying remain unexplored. The purpose of this study was to evaluate the current state of news media coverage of bullying and cyberbullying. A sample of newspaper articles covering bullying or cyberbullying across regional and national US newspapers from 6 recent years was identified. A content analysis using natural language processing was conducted with the Linguistic Inquiry and Word Count (LIWC) software program for key variables including affective, social, and cognitive processes. Evaluation included the percentage of words that represented Fear-based reporting such as alarmist words (e.g., epidemic, tragic), as well as words that represent Public Health-oriented messages such as prevention. A total of 463 newspaper articles met inclusion criteria, including 140 cyberbullying articles and 323 bullying articles. Findings indicated that cyberbullying articles scored higher on affective processes such as measures of anxiety (Mdn = 0.34) compared to bullying articles (Mdn = 0.22). A greater number of cyberbullying articles were Fear-based (41.4%) than were bullying articles (19.5%). An equivalent number of cyberbullying articles (50.0%) and bullying articles (49.8%) were Public Health-oriented. Findings may be used to collaborate with journalists toward optimizing prevention-oriented reporting.

Catering to Your Concerns

Android users are increasingly concerned with the privacy of their data and security of their devices. To improve the security awareness of users, recent automatic techniques produce security-centric descriptions by performing program analysis. However, the generated text does not always address users’ concerns as they are generally too technical to be understood by ordinary users. Moreover, different users have varied linguistic preferences that do not match the text. Motivated by this challenge, we develop an innovative scheme to help users avoid malware and privacy-breaching apps by generating security descriptions that explain the privacy and security related aspects of an Android app in clear and understandable terms. We implement a prototype system, PERSCRIPTION, to generate personalised security-centric descriptions that automatically learn users’ security concerns and linguistic preferences to produce user-oriented descriptions. We evaluate our scheme through experiments and user studies. The results clearly demonstrate the improvement on readability and users’ security awareness of PERSCRIPTION’s descriptions compared to existing description generators.