Linguistic Inquiry Word Count Research Articles

Dear Program Director: An Evaluation of Implicit Bias in Letters of Recommendation for Neurosurgery Residency.

Despite comprising half of medical students, women represent only 29.6% of neurosurgery applicants and 17% of residents, suggesting a "leak" in the career pipeline for women neurosurgeons. Surveys persistently show that neurosurgery programs identify United States Medical Licensing Examination (USMLE®) Step 1 score and letters of recommendation (LORs) as the most important factors in selecting applicants to interview. A previous study in neurosurgery found no differences in LORs. However, multiple studies in other specialties have demonstrated implicit gender bias in LORs, which may influence resident selection. Our objective is to evaluate neurosurgery residency LORs for evidence of implicit gender bias. Retrospective analysis of LORs for interviewed neurosurgery applicants at a single institution during the 2014 to 2020 National Residency Matching Program (NRMP®) match cycles. Letters were evaluated using Linguistic Inquiry & Word Count (LIWC) software (Pennebaker Conglomerates), and additional applicant data were obtained from candidate applications. LIWC (Pennebaker Conglomerates) output data included custom dictionary categories and terms that were analyzed using Prism 10 and Rstudio. Two hundred eighteen applications were reviewed for a total of 827 letters. LIWC (Pennebaker Conglomerates) analysis showed significant differences in word count (331 vs 297, difference = 34, 95% CI: 9-61, P = .008). LORs for applicants who were men were more likely to mention Alpha Omega Alpha Honor Medical Society (1.17 vs 0.778, difference = 0.4, 95% CI: 0.13-0.67, P = .023). USMLE® Step 1 scores were significantly lower for women (241 vs 247, difference = 6, 95% CI: 2-10, P = .004). There was no significant difference between letters for men and women for all categories evaluated in the linguistic evaluation. LORs are vital to the neurosurgical residency application process. The data exhibit some differences between the men and women applicants but few differences in their LORs, consistent with the results of the previous neurosurgical study.

Linguistic dissection of nursing handoffs: Implications for patient safety in varied-acuity hospital settings.

This study examines the intricate language and communication patterns of nurse-to-nurse handoffs across three units with varying patient acuity levels and nurse-patient ratios, seeking to identify linguistic factors that may affect the quality of information transfer and patient outcomes. A mixed-methods cross-sectional design. This study used the Nurse-to-Nurse Transition of Care Communication Model to explore the content and meaning of language in nursing handoffs within a large academic medical centre. Data were collected on three units through digital audio recordings of 20 handoffs between June and September 2022, which were transcribed and analysed using the Linguistic Inquiry Word Count programme. Trustworthiness was established by adhering to COREQ and STROBE guidelines for qualitative and quantitative research, respectively. Analysis revealed a preference for casual, narrative language across all units, with ICU nurses demonstrating a higher confidence and leadership in communication. Cognitive processes such as insight and causation were found to be underrepresented, indicating a potential area for miscommunication. Communication motives driven by affiliation were more pronounced in ICU settings, suggesting a strong collaborative nature. No significant differences were observed among the units post multiple testing adjustments. Speech dysfluencies were most pronounced in ICU handoffs, reflecting possible stress and cognitive overload. The study highlights the need for improved communication strategies such as interventions to enhance language clarity and incorporating technological tools into handoff processes to mitigate potential miscommunications and errors. The findings advance nursing science by highlighting the critical role of nuanced language in varied-acuity hospital settings and the necessity for structured nurse education in handoff communication and standardized handoff procedures. This study underscores the critical role of language in nurse-to-nurse handoffs. It calls for enhanced communication strategies, technology integration and training to reduce medical errors, improving patient outcomes in high-acuity hospital settings. Nurses only.

Emotion language use in narratives of the 9/11 attacks predicts long-term memory.

Despite considerable cognitive neuroscience research demonstrating that emotions can influence the encoding and consolidation of memory, research has failed to demonstrate a relationship between self-reported ratings of emotions collected soon after a traumatic event and memory for the event over time. This secondary analysis of data from a multisite longitudinal study of memories of the September 11, 2001 terrorist attacks, asked the question of whether emotional language use could predict memory over time. In the 2 weeks following the 9/11 attacks, participants (N = 691; Mage = 36.8; 72% identifying as male; 76% identifying as white) wrote narratives about how they learned of the attacks and the impact of the attacks on them. Language features of these narratives were extracted using the Linguistic Inquiry Word Count program and used to predict three types of memory: (a) event memory accuracy, (b) flashbulb memory consistency, and (c) emotion memory consistency. These outcomes were assessed at the time of writing, 1, 3, and 10 years after the 9/11 attacks. Results of linear mixed-effects models indicate that greater use of negative emotion words in narratives predicts better event memory accuracy 3 and 10 years after the attacks and worse flashbulb memory consistency 10 years after the attacks. However, emotion word use does not predict emotion memory consistency across time. We also examine whether other exploratory linguistic predictors are associated with memory over time. These findings suggest that written language may serve as a potential early indicator of memory over time. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

“There’s no one as honest as those in pain”: The language of Tom Petty’s song lyrics

Tom Petty’s songs are among the most recognized and influential in rock music. One aspect of Petty’s songs that contributes to his popularity is his use of language in his lyrics. The current study performed two sets of analyses: (a) to examine linguistic features of Petty’s song lyrics over the course of his songwriting career; and (b) to investigate similarities and differences between Petty’s lyrics and the earlier and contemporaneous lyrics of an artist widely assumed to have influenced him, Bob Dylan. Results of analyses using a standard textual program, Linguistic Inquiry Word Count (LIWC), showed that Petty reduced his references to social relations and reduced his focus on future time orientations over his songwriting career. There were also rises and declines in expressions of authenticity as well as increases in analytical thinking. In analyses comparing Petty to Dylan using a time-lagged approach, Petty’s songs remained shorter in number of words than Dylan’s songs and there was an increase in indicators of analytical thinking in Petty’s later work reaching levels that are characteristic of Dylan’s songs. Further research is needed to trace the influences on prolific songwriting and other aspects of music beyond lyrical expression.

Emoji use in social media posts: relationships with personality traits and word usage.

Prior research has demonstrated relationships between personality traits of social media users and the language used in their posts. Few studies have examined whether there are relationships between personality traits of users and how they use emojis in their social media posts. Emojis are digital pictographs used to express ideas and emotions. There are thousands of emojis, which depict faces with expressions, objects, animals, and activities. We conducted a study with two samples (n = 76 and n = 245) in which we examined how emoji use on X (formerly Twitter) related to users' personality traits and language use in posts. Personality traits were assessed from participants in an online survey. With participants' consent, we analyzed word usage in posts. Word frequencies were calculated using the Linguistic Inquiry Word Count (LIWC). In both samples, the results showed that those who used the most emojis had the lowest levels of openness to experience. Emoji use was unrelated to the other personality traits. In sample 1, emoji use was also related to use of words related to family, positive emotion, and sadness and less frequent use of articles and words related to insight. In sample 2, more frequent use of emojis in posts was related to more frequent use of you pronouns, I pronouns, and more frequent use of negative function words and words related to time. The results support the view that social media users' characteristics may be gleaned from the content of their social media posts.

Alexithymia in the Narratization of Romantic Relationships: The Mediating Role of Fear of Intimacy.

The purpose of the study was to verify the hypothesis concerning the relationship between alexithymia and selected indicators used to describe emotional events, specifically romantic relationships. Alexithymia, due to significant distortions in cognitive processing of emotional content, is demonstrated by poor recognition of emotions in oneself and others and, as a result, by deficits in empathy, avoidance of social relationships, and deficits in the ability to mentalize. Differences in narrations were tested by alexithymia levels (high vs. low) and the relation between specific narration features and individual alexithymia factors, i.e., difficulties in identifying emotions, difficulties in verbalising emotions, and externally oriented thinking. A total of 356 people who had been in a romantic relationship for at least six months participated in the study. The TAS-20 was applied to measure alexithymia, and the FIS questionnaire was used to investigate anxiety in close relationships. Participants were asked to freely describe the romantic relationship they were in at that moment. The Linguistic Inquiry Word Count (LIWCLIWC2015 v1.6-unlimited duration academic licence) software was used for content analysis. The study was conducted online. On the basis of the analyses conducted, high levels of alexithymia were found to be significantly associated with a lower total number of words used in narrative, a lower number relating to positive emotions, a lower number relating to causation and insight, and a higher number relating to negative emotions. Various results were obtained for individual dimensions of alexithymia in relation to the LIWC categories and the mediating role of fear of intimacy. For the difficulty identifying feelings (DIF), a significant mediating effect was observed only for words associated with negative emotions, whereas for the difficulty describing feelings (DDF), significant mediating effects were found for words relating to negative emotions and causality. In the case of externally oriented thinking (EOT), significant mediating effects were obtained for all analysed categories from LIWC.

LINGUISTIC PROFILES OF PARAGRAPH RECALL PREDICT DEPRESSION IN OLDER ADULTS

Abstract Our previous work identified linguistic markers from paragraph recall that predicted cognitive impairment. However, many features were related to psychological constructs, and mood disorders, such as depression, are also associated with cognitive impairment. Therefore, we aimed to identify linguistic markers that are associated with depression and examine their overlap with those for cognitive impairment. We used Linguistic Inquiry Word Count, a natural language processing tool, on Logical Memory transcriptions from the Long Life Family Study. Depression status was positive for participants with a self-reported depression diagnosis or high depressive symptoms (i.e., a CESD-10 score≥10) at testing. We used logistic regression models with Generalized Estimating Equations adjusted by age, sex, education, and within-family relatedness to identify linguistic features associated with depression status (Bonferroni-corrected threshold of p<.001). The sample included 61 participants with evidence of depression and 539 participants without (mean age=72.1±10.9 years, female=63%). A lower percentage of verb use in delayed recall was significantly associated with depression (Odds Ratio [OR]=0.71, CI 0.55-0.93, p=.0001). Additionally, 3 immediate recall features (verbs, OR=0.71; prepositions, OR=1.30; differentiation words, OR=1.30) and 5 delayed recall features (function words, OR=0.68; prepositions, OR=1.50; future tense, OR=0.48; cognitive process words, OR=0.78; past tense, OR=0.74) were significant at a nominal level (p=.009 to p=.035). We identified linguistic features from paragraph recall responses that predict depression and were largely distinct from those associated with cognitive impairment as only two features overlapped. Linguistic analyses of spoken responses may help us understand contributors to test performance and distinguish between cognitive impairment and depression.

‘I am tired, sad and kind’: self-evaluation and symptoms of depression in adolescents

IntroductionAlthough self-evaluation i.e., negative perceptions of the self is a common depression symptom in adolescents, little is known about how this population spontaneously describe their self and available data on adolescent self-evaluation is limited. This study aimed to generate and report on a list of words used by healthy adolescents and those with elevated depression symptoms to describe their self-evaluation. Linguistic analysis (LIWC) was then used to compare self-evaluation between the two groups.MethodsAdolescents aged 13–18 years (n = 549) completed a measure of depression symptoms (the Mood and Feelings Questionnaire) and a measure of self-evaluation (the Twenty Statements Test). Responses were then collated and presented in a freely accessible resource and coded using Linguistic Inquiry Word Count (LIWC) analysis.ResultsSelf-evaluation words generated by adolescents were uploaded to a publicly accessible site for future research: https://doi.org/10.15125/BATH-01234. Adolescents with elevated depression symptoms described themselves as ‘Tired’ and ‘Sad’ more than healthy adolescents. However, there was no difference between groups in respect to their use of specific positive, prosocial self-evaluation ‘words’ (i.e., ‘Caring’ and ‘Kind). Following Linguistic Inquiry Word Count (LIWC) analysis, adolescents with elevated depression symptoms generated significantly more words than healthy adolescents, generated more words classified as negative emotion, anxiety and sadness and generated fewer words classified positive emotion than healthy adolescents.ConclusionsAs predicted by the cognitive model of depression, our findings suggest that adolescents with elevated symptoms of depression generated more negative self-evaluation words than healthy adolescents; however they also generated prosocial positive self-evaluation words at the same rate as non-depressed adolescents. These novel data therefore identify an ‘island’ of resilience that could be targeted and amplified by psychological treatments for adolescent depression, and thus provide an additional technique of change.

Identifying self-disclosed anxiety on Twitter: A natural language processing approach

BackgroundText analyses of social media posts are a promising source of mental health information. This study used natural language processing to explore distinct language patterns on Twitter related to self-reported anxiety diagnosis. MethodsA total of 233.000 tweets made by 605 users (300 reporting anxiety diagnosis and 305 not) over six months were comparatively analysed, considering user behavior, Linguistic Inquiry Word Count (LIWC), and sentiment analysis. Twitter users with a self-disclosed diagnosis of anxiety were classified as ‘anxious’ to facilitate group comparisons. ResultsSupervised machine learning models showed a high prediction accuracy (Naïve Bayes 81.1 %, Random Forests 79.8 %, and LASSO-regression 79.4 %) in identifying Twitter users’ self-disclosed diagnosis of anxiety. Additionally, a Latent Profile Analysis (LPA) identified four profiles characterized by high sentiment (31 % anxious participants), low sentiment (68 % anxious), self-immersed (80 % anxious), and normative behavior (38 % anxious). ConclusionThe digital footprint of self-disclosed anxiety on Twitter posts presented a high frequency of words conveying either negative sentiment, a low frequency of positive sentiment, a reduced frequency of posting, and lengthier texts. These distinct patterns enabled highly accurate prediction of anxiety diagnosis. On this basis, appropriately resourced, awareness raising, online mental health campaigns are advocated.

95 Delving Beyond the Test Score: Linguistic Markers of Cognitive Impairment on Paragraph Recall

Objective:Cognitive tests requiring spoken responses, such as paragraph recall, are rich in cognitive-related information that is not captured using traditional scoring methods. This study aimed to determine if linguistic features embedded in spoken responses may differentiate between individuals who are and are not cognitively impaired.Participants and Methods:Participants in the Long Life Family Study completed a neuropsychological assessment which included the WMS-R Logical Memory I paragraph recall. For a subset of participants (N=709), test responses were digitally recorded and manually transcribed. We used Linguistic Inquiry Word Count, a text analysis program, to quantify word counts, grammatical features (e.g, prepositions, verb tenses), and the use of content words related to specific semantic categories (e.g., work-related, numbers) for immediate (IR) and delayed recall (DR). We used regression models with Generalized Estimating Equations adjusted by age, sex, education, and within-family correlation to select features associated with cognitive status (normal cognition [NC] versus cognitive impairment [CI]; Bonferroni-corrected threshold p<0.001). Next, we developed a “polyfeature score” (PFS) for both immediate and delayed recall, each calculated as a weighted sum of the selected linguistic features. We then built a logistic regression model to evaluate the predictive value of each PFS for identifying cognitively impaired individuals. In secondary analyses, we used regression models as above to identify features associated with mild cognitive impairment subtype (amnestic [aMCI] versus nonamnestic [naMCI]; threshold p< .05).Results:The sample included 599 participants with NC and 110 with CI (mean age = 72.3 ± 11.0 years, 54% female). The regression identified 8 linguistic features for IR and 7 for DR that significantly predicted cognitive status. Decreased use of content words related to work (e.g., employed, school, police) and biological processes (e.g., cook, cafeteria, eat) and the use of negations (e.g., no, not, can’t) were predictive of cognitive impairment in both recall conditions. In contrast, the use of other content word categories were predictive of cognitive status in only one recall condition (IR: leisure, cognitive processes, space; DR: drives, number). The use of fewer prepositions in IR, more first-person pronouns in DR, and fewer words in the past tense in DR were each associated with cognitive impairment. Word count was not predictive of cognitive status. Both PFSs were highly associated with cognitive status (PFS_IR ß= 0.74, p< 0.001; PFS_DR ß= 0.86, p= 0.001) with high discriminative value (PFS_IR AUC= 0.93, sensitivity = 0.81, specificity= 0.91; PFS_DR AUC= 0.95, sensitivity= 0.77, specificity= 0.88). In the CI subset, linguistic features differed between those classified as aMCI (n= 24) and naMCI (n= 40). Two function word categories predicted aMCI in IR whereas decreased word count, two function word categories, and two content word categories predicted aMCI in DR (all p< .05)Conclusions:Linguistic features from paragraph recall provide high predictive value for classifying cognitive status increasing its potential as a cognitive screener in clinical settings. Additionally, each recall condition identified unique linguistic features associated with cognitive impairment which may aid differentiation of cognitive impairment subtypes and elucidate processes underlying deficits in learning and recall.

Letters of recommendation for gynecologic oncology fellowship differ by gender and race

IntroductionLetters of recommendation (LOR) play a significant role in applicant selection for gynecologic oncology (GO) fellowship. LORs can be agentic, associated with competence, independence and strength, or communal, associated with being accommodating, cooperative and empathetic. Agentic LORs have been shown to reflect favorably on applicants, making them more likely to be selected for interview. The primary goal of this study was to determine linguistic differences in LORs for applicants applying to GO fellowship based on applicant gender and race. MethodsAll applications to a single academic institution from the years 2018–2020 were analyzed. LORs were extracted from all applicants who self-reported their gender and race/ethnicity. Analysis was performed using Linguistic Inquiry Word Count (LIWC) in terms of agentic and communal language. All analysis was done using SAS version 9.4. ResultsFrom 2018 to 2020, there were 239 GO fellowship applications to a single academic institution. There were 186 (78.1%) applicants who identified as female and 52 (21.8%) who identified as male. Female applicants were more likely to have more research (p = 0.047) and volunteer activities (p = 0.02) than male applicants. There were no differences between female and male applicants in terms of age, geographic location of residency, USMLE scores, AOA status, number of publications or attending a Doximitiy-rated top 20 residency program. There were 139 (60.7%) applicants who identified as White, 46 (20.1%) as Asian, 16 (7.0%) as Black, 19 (8.3%) as Latinx and 9 (3.9%) as other. On multivariable analysis, LORs written for male applicants had higher agentic scores compared to females (p < 0.05), and LORs of female applicants received higher communal scores (p < 0.05). LORs for applicants who identified as Black or Latinx were found to have lower agentic scores compared to individuals who identified as White(p < 0.05 for both). ConclusionThere were demonstrated linguistic differences in LOR for GO fellowship based on applicant gender and race/ethnicity. LORs written for white, male applicants were more likely to have higher agentic scores compared to minority, female applicants. LORs written for female applicants were more likely to have higher communal scores. As LORs have been shown to be instrumental in GO fellowship interview selection and linguistic differences exist based on applicant gender and race/ethnicity, this study highlights the need for a more objective GO fellowship application process. DisclosuresPortion of this paper was presented at the Society for Gynecologic Oncology as part of a focus plenary.

Cancer caregivers at the end-of-life: How much me vs. how much we?

Objective: This study explored cancer caregivers' individual and communal coping through their use of personal and communal pronouns during naturally occurring conversations. Methods: Nurse-home hospice visits involving cancer patients and their partner caregivers were audio recorded and then transcribed. Pronoun use was analyzed using Linguistic Inquiry Word Count (LIWC) software and descriptive statistics compared patient and partner caregivers' pronoun use. Personal and communal pronoun use was examined within six identified topics of caregiver speech: patient medical care, daily life, emotion, criticism/disagreement, relationships with family/friends, and asserting needs. Results: Dyads (N = 76) had an average of 35.8 years in their relationship. Caregivers used proportionately more first-person singular (I-talk) than first-person plural (we-talk). However, they used significantly less I-talk than patients and less I-talk than LIWC measures in naturally occurring speech. Caregivers were most likely to discuss patient medical care (41.9%) and least likely to discuss their own needs (3.8%). Conclusion: Partner caregivers may find it easier to express emotions related to communal stressors, rather than their individual ability to cope with end-of-life caregiving. Innovation: Examining personal and communal pronoun use by partner caregivers during nurse-home hospice visits may provide a more objective measure of caregiver coping than standard self-report measures.

A woman's tradition? Quantifying gender difference in the Child ballads

AbstractWhile the major ballad‐collecting efforts of the eighteenth and nineteenth centuries were largely carried out by men, the academic discourse surrounding ballads gendered traditional balladry—orally transmitted narrative poetry—as a woman's tradition. Subsequent ballad scholarship of the twentieth century perpetuated the antiquarian notion of a specifically female ballad tradition, and yet it has remained unclear whether and how this women's tradition should be considered distinct from, for example, a male ballad tradition. This study suggests that a quantitative approach can be used to investigate the question of difference in women's and men's ballad repertoires in the period spanning the mid‐eighteenth century to the close of the nineteenth. Employing Linguistic Inquiry Word Count (LIWC) software to examine the definitive ballad anthology—the Child collection—this study finds that the notion of a woman's tradition goes beyond stereotypical notions of repertoires gendered by genre; women's ballads are instead found to be characterised by distinctive linguistic patterns that convey a strong focus on female subjectivity.

The New Testament vs. The Quran: Americans’ Beliefs About the Content of Muslim and Christian Holy Texts

The stereotypes and biases that people have about various religions may not be applied to just the individuals who belong to those religions, but to the belief systems themselves. We hypothesized that non-Muslim and Muslim Americans would demonstrate biased estimates about the language content of the Quran and New Testament holding positive views towards their relative ingroup and negative views towards the outgroup text. We used two samples ( N1 = 163, N2 = 204) of Americans and the Linguistic Inquiry Word Count Software (LIWC) to test our hypotheses. We determined the differences between the texts across language categories. Participants then rated the categories in terms of their relative frequency across the texts. We compared these ratings to the actual differences between the texts. As hypothesized, participants perceived the ingroup text as positive and the outgroup text as negative. We discuss whether biased beliefs about religious teachings may be separate from but aligned with biases against believers and further contribute to religious stereotypes.

Evaluation of the Brazilian Portuguese version of linguistic inquiry and word count 2015 (BP-LIWC2015)

Text psycholinguistic features are a valuable source for various research topics since they are used to obtain psychological, social, and linguistic aspects from written texts using dictionary files. These files are structured in categories, which are defined as groups of dictionary words that tap a particular domain (e.g., negative emotion words). The Linguistic Inquiry Word Count (LIWC) is a vastly used and versatile computer-based language analysis tool designed for text psycholinguistic analysis. The most recent version of the default English dictionary is LIWC2015, as it was released with the 2015 version of the LIWC software. The literature has recently introduced the latest Brazilian Portuguese LIWC dictionary (BP-LIWC2015), developed with the same categories as the LIWC 2015 English dictionary. However, the literature has also reported the need to evaluate BP-LIWC2015. In this scenario, this work investigates three questions: (i) Since LIWC2015 shows consistent improvements over the English dictionary developed in 2007 (LIWC2007), does BP-LIWC2015 achieves better text classification results than the older Brazilian Portuguese dictionary (BP-LIWC2007)? (ii) What is the equivalence between BP-LIWC2015 and BP-LIWC2007 with LIWC2015? (iii) Are there significant differences between Brazilian Portuguese dictionaries? To answer these questions, we conducted text classification experiments with four datasets and seven classification algorithms to compare the two Brazilian Portuguese LIWC dictionaries reported in the literature (i.e., 2007 and 2015). Second, we used a bilingual Portuguese-English scientific news collection to analyze the correlation between LIWC2015 and Brazilian Portuguese LIWC dictionaries. The results indicate that BP-LIWC2015 outperforms the older version in Brazilian Portuguese text classification. Finally, we found a more significant correlation between BP-LIWC2015 and the original English dictionary than the older version.

Determining psycholinguistic features of deception in phishing messages

PurposeDistinguishing phishing emails from legitimate emails continues to be a difficult task for most individuals. This study aims to investigate the psycholinguistic factors associated with deception in phishing email text and their effect on end-user ability to discriminate phishing emails from legitimate emails.Design/methodology/approachEmail messages and end-user decisions collected from a laboratory phishing study were validated and analyzed using natural language processing methods (Linguistic Inquiry Word Count) and penalized regression models (LASSO and Elastic Net) to determine the linguistic dimensions that attackers may use in phishing emails to deceive end-users and measure the impact of such choices on end-user susceptibility to phishing.FindingsWe found that most participants, who played the role of a phisher in the study, chose to deceive their end-user targets by pretending to be a familiar individual and presenting time pressure or deadlines. Results show that use of words conveying certainty (e.g. always, never) and work-related features in the phishing messages predicted higher end-user vulnerability. On the contrary, use of words that convey achievement (e.g. earn, win) or reward (cash, money) in the phishing messages predicted lower end-user vulnerability because such features are usually observed in scam-like messages.Practical implicationsInsights from this research show that analyzing emails for psycholinguistic features associated with computer-mediated deception could be used to fine-tune and improve spam and phishing detection technologies. This research also informs the kinds of phishing attacks that must be prioritized in antiphishing training programs.Originality/valueApplying natural language processing and statistical modeling methods to analyze results from a laboratory phishing experiment to understand deception from both attacker and end-user is novel. Furthermore, results from this work advance our understanding of the linguistic factors associated with deception in phishing email text and its impact on end-user susceptibility.

Detecting depression of Chinese microblog users via text analysis: Combining Linguistic Inquiry Word Count (LIWC) with culture and suicide related lexicons.

In recent years, research has used psycholinguistic features in public discourse, networking behaviors on social media and profile information to train models for depression detection. However, the most widely adopted approach for the extraction of psycholinguistic features is to use the Linguistic Inquiry Word Count (LIWC) dictionary and various affective lexicons. Other features related to cultural factors and suicide risk have not been explored. Moreover, the use of social networking behavioral features and profile features would limit the generalizability of the model. Therefore, our study aimed at building a prediction model of depression for text-only social media data through a wider range of possible linguistic features related to depression, and illuminate the relationship between linguistic expression and depression. We collected 789 users' depression scores as well as their past posts on Weibo, and extracted a total of 117 lexical features via Simplified Chinese Linguistic Inquiry Word Count, Chinese Suicide Dictionary, Chinese Version of Moral Foundations Dictionary, Chinese Version of Moral Motivation Dictionary, and Chinese Individualism/Collectivism Dictionary. Results showed that all the dictionaries contributed to the prediction. The best performing model occurred with linear regression, with the Pearson correlation coefficient between predicted values and self-reported values was 0.33, the R-squared was 0.10, and the split-half reliability was 0.75. This study did not only develop a predictive model applicable to text-only social media data, but also demonstrated the importance taking cultural psychological factors and suicide related expressions into consideration in the calculation of word frequency. Our research provided a more comprehensive understanding of how lexicons related to cultural psychology and suicide risk were associated with depression, and could contribute to the recognition of depression.

Automated measures of speech content and speech organization in schizophrenia: Test-retest reliability and generalizability across demographic variables

Technological advances in artificial intelligence and natural language processing have increased efficiency of assessing speech content and speech organization in schizophrenia. Despite these developments, there has been little focus on the psychometrics of these approaches. Using two common assessments, the current study addressed this gap by: 1) measuring test-retest reliability; and 2) assessing whether speech content and/or speech organization generalize across demographics. To test these aims, we examined psychometric properties of the Linguistic Inquiry Word Count (LIWC), a speech content measure, and the Coh-Metrix, a speech organization measure. Across baseline to six month (n = 101) and baseline to one year (n = 47) narrative speech samples, we generally observed fair reliability for speech content measures and fair to good reliability for speech organization measures. Regarding demographics, multiple speech indices varied by race, income, and education. The lack of excellent reliability scores for speech indices holds important implications for examining speech variables in clinical trials and highlights the dynamic nature of speech. This work illustrates the importance of designing speech content and speech organization measures with external validity across demographic factors. Future studies examining speech in schizophrenia should account for potential biases against demographic groups introduced by linguistic analysis tools.

Naïve Bayes-Support Vector Machine Combined BERT to Classified Big Five Personality on Twitter

Twitter is one of the most popular social media used to interact online. Through Twitter, a person's personality can be determined based on that person's thoughts, feelings, and behavior patterns. A person has five main personalities likes Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. This study will make five personality predictions using the Naïve Bayes method – Support Vector Machine, Synthetic Minority Over Sampling Technique (SMOTE), Linguistic Inquiry Word Count (LIWC), and Bidirectional Encoder from Transformers Representations (BERT). A questionnaire was distributed to people who used Twitter to collect and become a dataset in this research. The dataset obtained will be processed into SMOTE to balance the data. Linguistic Inquiry Word Count is used as a linguistic feature and BERT will be used as a semantic approach. The Naïve Bayes method is used to perform the weighting and the Support Vector Machine is used to classify Big Five Personalities. To help improve accuracy, the Optuna Hyperparameter Tuning method will be added to the Naïve Bayes Support Vector Machine model. This study has an accuracy of 87.82% from the results of combining SMOTE, BERT, LIWC, and Tuning where the accuracy increases from the baseline.

The mediating effect of cognitive and emotional processing on PTSD and depression symptoms reduction in women victims of IPV.

Intimate partner violence (IPV) is a serious social, physical and mental health issue. Women victims of IPV can develop short- and long-term consequences such as depression and post-traumatic stress disorder (PTSD). Where trauma has been incurred, standard psychotherapies may usefully be complemented by interventions based on expressive writing (EW). Numerous studies have explored the mechanisms underpinning improvement after writing, focusing on the cognitive and emotional processing of traumatic experiences. The aims of this study were to evaluate changes in PTSD and depression symptoms following EW and to examine the mediating effect of emotional and cognitive processing on symptom reduction in subjects who engaged in EW. Seventy-seven abused women (mean age = 41.43, SD = 10.75) were randomly assigned to a three-session expressive writing condition (n = 43) or a neutral writing condition (n = 34). Psychological distress (PTSD and depression) was assessed both before and after the writing sessions. Linguistic inquiry word count software was used to analyze the women's narratives in relation to emotional processing (positive and negative emotions) and cognitive processing (insight and causal attributions). The mediation model indicated that the reduction in depression was fully mediated by negative emotion processing and partially mediated by cognitive processing, while the reduction in PTSD was partially mediated by negative emotion processing and fully mediated by cognitive processing. No effect of positive emotion processing was found. The clinical implications are discussed.