Mayor, E., Bietti, L. M., & Bangerter, A. (2025). Can large language models simulate spoken human conversations? Cognitive Science, 49, e70106. https://doi.org/10.1111/cogs.70106

Mayor, E., & Bietti, L. M. (2025). A social media study of portrayals of bipolar disorders on YouTube: Content and thematic analyses. Journal of Medical Internet Research, 27, e67129. https://doi.org/10.2196/67129

Abstract

Background: Individuals with mental disorders frequently use YouTube to express themselves, reach an audience, or as a means of understanding their condition. Testimonies posted on YouTube provide longer and richer perspectives than the short posts found on other social media platforms. Research focusing on the depiction of mental disorders on YouTube is blossoming. Bipolar disorders (BDs) are disabling mood disorders. The diagnosis of any mental disorder, and more so BD, is often a life-changing event. However, no published study has investigated the portrayal of diagnoses of BD on YouTube.

Objective: This study aims to investigate the portrayals of BDs on YouTube, focusing on the diagnosis narratives and their accompanying narrative context, in particular, reports of personal experiences and reactions.

Methods: We performed a manual content analysis of 39 testimonies (women: n=24, 62%) depicting BDs and their diagnosis by individuals with BD. We also performed a thematic analysis of the corpus relying upon a deductive and inductive approach.

Results: Our manual content analysis revealed that portrayals included the disclosure of diagnoses of BD-I (as per both coders’ agreement: 10 testimonies) and BD-II (11 testimonies) to a similar extent. The reactions to the diagnosis were mostly negative (8 testimonies), followed by positive (5 testimonies), while fewer portrayals indicated a denial of the condition (4 testimonies). Several portrayals made mention of issues in the areas of money and accommodation (15 testimonies), profession and education (13 testimonies), and relationships (20 testimonies). Medication (31 testimonies) and psychotherapy (23 testimonies) were often mentioned as part of treatment for BD, most generally in positive terms. The 8 themes emerging from the thematic analysis were: “reactions on diagnosis, treatment, and health care professionals’ expertise,” “trial and error in medication,” “positive effects of BD,” “disability, stigma, and shame,” “loss,” “family planning and genetics,” “identity change (psychological and physical),” and “human social relationships.”

Conclusions: Overall, our results underline the complexity and richness of the depiction of the diagnosis of BD and its narrative context, and highlight the importance of the moment of the diagnosis, medication, and psychotherapy. Our study emphasizes the need for further exploration of the impact of social media on mental health awareness.

Brandel, N., Schwarz, B. B., Cedar, T., Baker, M. J., Bietti, L. M., Pallarès, G., & Détienne, F. (2024). Dialogue on ethics and ethics of dialogue: An exploratory study. European Journal of Psychology of Education. Advance online publication. https://doi.org/10.1007/s10212-024-00856-z

Abstract: We report on a study bearing implications for ethical learning in schoolchildren during social interaction. The study was conducted as part of a project aimed at promoting ethical learning of socially-oriented values within the context of dialogic education. 172 fourth graders from 7 classes participated in an 8-session series designed to foster empathy, inclusion, and tolerance. Two of these sessions (3 and 8) were pre-selected for analysis. We investigated (1) whether students’ discussion of ethical issues and the ethical aspects of their actual in-class interaction with each other can be reliably measured, and (2) what relation holds between students’ ethical thinking during classroom discussions and the ethical aspects of their behavior. We thus developed an analytical framework comprising two tools for appraising ethical thinking and behavior in in-class interaction: dialogue on ethics (DoE) and ethics of dialogue (EoD). This framework was applied to the dialogues taken from the two sessions. The DoE and EoD tools proved reliable, as inter-rater agreement was substantial. Moreover, the relation between children’s DoE and their EoD was positive where the topic posed for discussion presented a dilemma and students’ interaction proceeded under moderate teacher guidance. In contrast, it was negative when the discussion was conceptual, and the teacher was dominant. We conclude that (1) DoE/EoD is a suitable framework for studying children’s ethical learning and development in social interaction, and (2) ethical learning, in its epistemological and behavioral dimensions, can be boosted or inhibited in a context of dialogic education, depending on design principles.

Mayor, E., & Bietti, L.M. (2024). Language use on Twitter reflects social structure and social disparities. Heliyon. https://doi.org/10.1016/j.heliyon.2023.e23528

Abstract. Large-scale mental health assessments increasingly rely upon user-contributed social media data. It is widely known that mental health and well-being are affected by minority group membership and social disparity. But do these factors manifest in the language use of social media users? We elucidate this question using spatial lag regressions. We examined the county-level (N = 1069) associations of lexical indicators linked to well-being and mental health, notably depression (e.g., first-person singular pronouns, negative emotions) with markers of social disparity (e.g., the Area Deprivation Index–3) and ethnicity, using a sample of approximately 30 million content-coded tweets (U.S. county-level aggregation). Results confirmed most expected associations: County-level lexical indicators of depression are positively linked with county-level area disparity (e.g., economic hardship and inequity) and percentage of ethnic minority groups. Predictive validity checks show that lexical indicators are related to future health and mental health outcomes. Lexical indicators of depression and adjustment coded from tweets aggregated at the county level could play a crucial role in prioritizing public health campaigns, particularly in socially deprived counties.

Bietti, L. M. & Mayor, E. (2023). A longitudinal study of conversational remembering in WhatsApp group messages before, during, and after COVID-19 lockdown. Memory, Mind & Media, 2, E5.  https://doi.org/10.1017/mem.2023.5.

Abstract. Conversational remembering entails that people engage in recalling past experiences, which may themselves have been shared. Conversational remembering comes with social benefits for the person telling the narrative and the one receiving it (e.g., developing and strengthening friendships, fostering entertainment, and consolidating group identity). COVID-19 lockdowns have significantly affected social interaction, including face-to-face interactions where conversational remembering occurs. The aim of this study was to explore how WhatsApp group messages supported conversational remembering in a large group of friends living in Buenos Aires where a complete lockdown was established between 19 March 2020 and 6 November 2020. To accomplish such aim, we conducted a mixed-methods longitudinal study. The data consisted of 32,810 WhatsApp group messages collected over a period of 700 consecutive days, from 13 April 2019 to 13 March 2021. Our study shows that WhatsApp group messages enabled group members to keep connected during the COVID-19 lockdown period. This occurred by remembering together situations, events, and actions associated with the group's identity. The use of WhatsApp group messages may have represented an adaptive collective behaviour in response to changes in global social norms.

Skjuve, M. & Bietti, L.M. (2023). Remembering with my chatbot. ACM Interactions, 30 (1). https://interactions.acm.org/blog/view/remembering-with-my-chatbot

The article begins “Did you know that humans can develop friendly or even romantic feelings toward social chatbots that can turn into close human-chatbot relationships? The phenomenon of human-chatbot relationships is starting to gain substantial media attention, and research on this topic is now emerging. Social chatbots. “What are they?” you might ask. Well, you’ve probably seen them in the App Store or on Google Play under names such as Replika or Kuki. To put it simply, a social chatbot is a form of conversational AI developed with the purpose of having normal, day-to-day conversations with its users. New developments in AI and NLP created the conditions for the design and development of sophisticated social chatbots capable of becoming your best friend, or even romantic partner. Social chatbots are generally good at showing empathy and providing emotional support and companionship. They have essentially grown into conversational affective machines, which makes it possible for users to form close relationships with them…”

Baker, M. J., Pallarès, G., Cedar, T., Brandel N., Bietti, L.M., Schwarz, B. & Détienne, F. (2023). Understanding the moral of the story: Collaborative interpretation of visual narratives. Learning, Culture and Social Interaction, 39, 100700, https://doi.org/10.1016/j.lcsi.2023.100700

Abstract. Fostering moral thinking and cultural literacy are major contemporary concerns in Europe and beyond, as means for young people to co-create social futures. We present a theoretical-methodological approach to understanding students’ moral thinking in the context of collaborative interpretation of visual narratives (“wordless texts”) with ethical implications. Six layers of interpretation are defined, from referential reconstruction of characters’ intentions, through semiotic symbolism, to making the moral of the story explicit in terms of conceptualisations of three key European values (empathy, inclusion and tolerance). Within a case-study approach to analysing computer-mediated dialogues, we show the extent to which students are led to discuss and understand ethical implications of a particular narrative, and how this relates to the quality of collaboration.

Mayor, E., Bietti, L.M. & Canales-Rodríguez, E.J. (2022) Text as signal. A tutorial with case studies focusing on social media (Twitter). Behavior Research Methods. https://doi.org/10.3758/s13428-022-01917-1

Abstract. Sentiment analysis is the automated coding of emotions expressed in text. Sentiment analysis and other types of analyses focusing on the automatic coding of textual documents are increasingly popular in psychology and computer science. However, the potential of treating automatically coded text collected with regular sampling intervals as a signal is currently overlooked. We use the phrase "text as signal" to refer to the application of signal processing techniques to coded textual documents sampled with regularity. In order to illustrate the potential of treating text as signal, we introduce the reader to a variety of such techniques in a tutorial with two case studies in the realm of social media analysis. First, we apply finite response impulse filtering to emotion-coded tweets posted during the US Election Week of 2020 and discuss the visualization of the resulting variation in the filtered signal. We use changepoint detection to highlight the important changes in the emotional signals. Then we examine data interpolation, analysis of periodicity via the fast Fourier transform (FFT), and FFT filtering to personal value-coded tweets from November 2019 to October 2020 and link the variation in the filtered signal to some of the epoch-defining events occurring during this period. Finally, we use block bootstrapping to estimate the variability/uncertainty in the resulting filtered signals. After working through the tutorial, the readers will understand the basics of signal processing to analyze regularly sampled coded text.

 

Bietti, L.M. & Bietti, F.U. (2021). The interactive functions of questions in embodied collaborative work. Frontiers in Psychology, 12:704275. doi: 10.3389/fpsyg.2021.704275

Abstract: Researchers have been interested in the investigation of the interactive functions of questions in conversational contexts. However, limited research has been conducted on the interactive functions of questions in embodied collaborative work, that is, work that involves the manipulation of physical objects. This study aimed to identify the interactive functions of questions in embodied collaborative work. To do so, we conducted a systematic qualitative analysis of a dataset of 1,751 question-answer sequences collected from an experimental study where pairs of participants (N = 67) completed a collaborative food preparation task. The qualitative analysis enabled us to identify three functions of questions: anticipation questions, exploration questions, and confirmation questions. We have discussed in this study how the types of questions identified are associated with: (i) the accomplishment of interactional goals and (ii) complementary temporalities in the collaborative activities.

Mayor, E. & Bietti, L. M. (2021). Twitter, time and emotions. Royal Society Open Science. https://doi.org/10.1098/rsos.201900

Abstract: The study of temporal trajectories of emotions shared in tweets has shown that both positive and negative emotions follow nonlinear circadian (24 h) and circaseptan (7-day) patterns. But to this point, such findings could be instrument-dependent as they rely exclusively on coding using the Linguistic Inquiry Word Count. Further, research has shown that self-referential content has higher relevance and meaning for individuals, compared with other types of content. Investigating the specificity of self-referential material in temporal patterns of emotional expression in tweets is of interest, but current research is based upon generic textual productions. The temporal variations of emotions shared in tweets through emojis have not been compared to textual analyses to date. This study hence focuses on several comparisons: (i) between Self-referencing tweets versus Other topic tweets, (ii) between coding of textual productions versus coding of emojis, and finally (iii) between coding of textual productions using different sentiment analysis tools (the Linguistic Inquiry and Word Count—LIWC; the Valence Aware Dictionary and sEntiment Reasoner—VADER and the Hu Liu sentiment lexicon—Hu Liu). In a collection of more than 7 million Self-referencing and close to 18 million Other topic content-coded tweets, we identified that (i) similarities and differences in terms of shape and amplitude can be observed in temporal trajectories of expressed emotions between Self-referring and Other topic tweets, (ii) that all tools feature significant circadian and circaseptan patterns in both datasets but not always, and there is often a correspondence in the shape of circadian and circaseptan patterns, and finally (iii) that circadian and circaseptan patterns obtained from the coding of emotional expression in emojis sometimes depart from those of the textual analysis, indicating some complementarity in the use of both modes of expression. We discuss the implications of our findings from the perspective of the literature on emotions and well-being.

Our study has been featured in several media outlets, including the Daily Mail, NRK, Aftenposten and Gemini.no.