I am a doctoral researcher at the Institute of Human-Centered Computing (under David Garcia) at the Graz University of Technology. Starting from July 2025, I am part of the IDea Lab, the Interdisciplinary Digital Lab at the University of Graz (group Data Analysis, Jana Lasser). During the academic period 2023/24, I was a visiting fellow at the Digital Emotions Lab at Harvard University led by Amit Goldenberg, which was funded by the German Academic Scholarship Foundation (Studienstiftung des Deutschen Volkes), the Austrian Marshall Plan Foundation, and the Agency for Education and Internationalization Austria (OeAD).
My research interests include everything emotions, political psychology, and methods development (check out my current projects below). In particular, I use methods from computer science, such as machine learning and natural language processing, to gain a more naturalistic and integrated understanding of emotion regulation as it happens in real life. I also study counter speech, a citizen-organized response to online hate; and use those questions to develop computational methods for the social sciences. I believe in transparent and reproducible science, good storytelling, interdisciplinarity and the advancement of science through new methods of measurement.
I started my academic career with a bachelor degree in business psychology at the University of Lüneburg. In an earlier life, I was an advertising strategist in Hamburg for clients such as NIVEA, SEAT and E.ON. While obtaining my master degree in psychology at the University of Graz, I realized that methods from computer science provide great opportunities to study psychological phenomena. That's why I enrolled in computer science at the Graz University of Technology - before I started my PhD in the same field.
We introduce a new method - the Construct Mining Pipeline - to infer the structure of psychological constructs from tailor-made text data and demonstarte its applicability on an example from emotion regulation.
While our understanding of hate speech in online discussions has grown significantly over the past few years, counter speech is still poorly understood. We try to change that.
LLMs are doing well in providing emotional support - at least when it comes to reappraisal. What's behind GPT-4's reappraisal success?
Reappraisal (or positive rethinking) is effective, but why? Understanding how different people approach reappraisal reveals distinct techniques and their relationship to reappraisal quality.
Open Science is one remedy against psychology's replication crisis. Those principles are now adopted by many social scientists. How is Open Science taught to the researchers of tomorrow: bachelor and master students of psychology?
Together with Sophie Labat and Jana Lasser, I conceptualized and conducted a half-day workshop on data annotation at ICWSM 2024. How to fine-tune machine learning models for classification is taught exhaustively, but how to arrive at good human annotated data is rarely discussed. In this workshop, we cover the influence of training data on model performance, creating an annotation scheme with Grounded Theory, data perspectivism, and interrater agreement including practical exercises.
I designed a two day introductory workshop for machine learning from psychologist for psychologists or other social science researchers. It includes basic vocabulary and machine learning algorithms, neural networks and NLP, as well as the data science pipeline and programming exercises. The current version is in German, but English is possible as well. Please feel free to contact me, if you are interested.
19 Mar, 2025. When networks can think: The meaning of self-regulation in the presence of humans (Invited talk). In: Self-regulating and learning systems - from neural to social networks. German Physical Society Spring Meeting, Regensburg, Germany.
11 Feb, 2025. Counter speech: Fighting windmills? (Invited talk). Bavarian Research Institute for Digital Transformation (bidt) at LMU, Munich, Germany.
18 Jul, 2024. A computational method to reveal psychological constructs from text data. International Conference for Computational Social Science, Philadelphia, USA.
2 Feb, 2024. Inferring a data-driven taxonomy of emotion regulation strategies with computational methods. In: Understanding affective experiences with natural language processing (Symposium). Annual Conference of the Society for Affective Science, New Orleans, USA.
21 Jul, 2023. Improving discourse quality in online discussions: A large-scale, longitudinal study of influencing speech characteristics, International Conference for Computational Social Science, Copenhagen, Denmark.
14 Jul, 2023. Improving discourse quality in social media networks: A large-scale, longitudinal study of influencing speech characteristics, International Conference on Network Science, Vienna, Austria.
26 Aug, 2022. Breaking the wall of emotional well-being, European Forum Alpbach, Alpbach, Austria.
20 May, 2022. Breaking the wall of emotional well-being, Falling Walls Lab Austria, Vienna, Austria.
10 June, 2022. Deriving emotion regulation startegy classes using a computational grounded theory approach, 8th Conference of the Consortium of European Research on Emotion, Granada, Spain.
Herderich, A., Freudenthaler, H. H., & Garcia, D. (2024). A Computational Method to Reveal Psychological Constructs from Text Data. Psychological Methods, Advance online publication. https://doi.org/10.1037/met0000700
Lasser, J.*, Herderich, A.*, Garland, J., Aroyehun, S. T., Garcia, D., & Galesic, M. (2023). Collective moderation of hate, toxicity, and extremity in online discussions. ArXiv. https://arxiv.org/abs/2303.00357
Herderich, A.*, Lasser, J.*, Galesic, M., Aroyehun, S. T., Garcia, D., & Garland, J. (2024). Measuring complex psychological and sociological constructs in large-scale text. PsyArxiv. https://doi.org/10.31234/osf.io/tzc9p
Li, J. Z., Herderich, A., Nair, P., & Goldenberg, A. (2025). Skill but not Effort Drive GPT Overperformance over Humans in Cognitive Reframing of Negative Scenarios. PsyArxiv. https://doi.org/10.31234/osf.io/fzvd8
Herderich, A., Li, J. Z., & Goldenberg, A. (2024). Data-driven classification of interpersonal reappraisals reveals eight distinct reappraisal strategies [Scholarship Paper]. Austrian Marshall Plan Foundation. https://www.marshallplan.at/papers-2024
Röseler, L., Kaiser, L., Doetsch, C. A., Klett, N., Seida, C., Schütz, A., …, Herderich, A., …, & Zhang, Y. (2024). The Replication Database: Documenting the Replicability of Psychological Science. Journal of Open Psychology Data, 12(8), 1-23. https://doi.org/10.5334/jopd.101
Frank, M., Grüning, D. J., Pronizius, E., Korbmacher, M., Elsherif, M. M., Bergmann, L., … Herderich, A. (2024). Mapping students' Open Science attitudes and preferences: Country and education level differences - PCI RR Stage 1 Report. PsyArxiv. https://doi.org/10.31219/osf.io/7gbvp
If you want to learn more about all the great people in my research group, visit our group's landing page.
Alina Herderich
University of Graz
IDea Lab - The Interdisciplinary Digital Lab
Leechgasse 34/II
8010 Graz AUSTRIA