SMI paper draft

schizophrenie & mental images - draft hypothesis

LX
psycholinguistics
CLIP
mental images
SMI
Author

st. schwarz

Published

February 7, 2026

1 research question and hypothesis

1.1 preliminary

1.2 Q:

  • how do mental images relate in wordfields i.e. what are typical resp. atypical semantic fields in which imageability of patients language increases or decreases?
  • how do word categories from concrete to abstract influence imageablity in relation to controls language?
  • are there genres/fields which are avoided/promoted?

1.3 hypothesis

we assume that:

  1. violence wordfield increase imageability
    1. body associated wordfields in contrast decrease imageability
  2. food wordfield decrease imageability
  3. weather/climate/atmosphere wordfield decrease imageability
  4. abstract categories increase in imageability
  5. concrete categories decrease in imageability

References

2 references

Bates, Douglas, Martin Mächler, Ben Bolker, and Steve Walker. 2015. “Fitting Linear Mixed-Effects Models Using Lme4.” Journal of Statistical Software 67 (1): 1–48. https://doi.org/10.18637/jss.v067.i01.
huggingface. n.d. “Google Colab.” Accessed December 15, 2025. https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/stable_diffusion.ipynb.
Nenchev, Ivan. 2026. “Clip Score Computation.” https://github.com/esteeschwarz/SPUND-LX/blob/main/mental-img/clip_scores.ipynb.
Nenchev, Ivan, Berlin Institute of Health at Charité – Universitätsmedizin Berlin, Germany, Christiane Montag, Department of Psychiatry and Psychotherapy, Charité Campus Mitte, Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Germany, Sandra Anna Just, Department of Psychiatry and Psychotherapy, Charité Campus Mitte, Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Germany, and Department of Clinical Medicine, UiT – The Arctic University of Norway, Tromsø, Norway. 2025. “Reverse Prompting: A Novel Computational Paradigm in Schizophrenia Based on Large Language Models.” In, 797–806. https://doi.org/10.26615/978-954-452-098-4-092.
Patil, Suraj, Pedro Cuenca, Nathan Lambert, and Patrick von Platen. 2022. “Stable Diffusion with 🧨 Diffusers.” Hugging Face Blog. https://huggingface.co/blog/stable_diffusion.
Pillny, Matthias, David J. Hallford, and Kerem Böge. 2024. “The Nature of Mental Imagery and Its Relationship With Amotivational Psychopathology in People With Schizophrenia Spectrum Disorders.” Behavior Therapy 55 (4): 885–97. https://doi.org/10.1016/j.beth.2024.01.009.
Princeton U. 2005. “Current Version.” Wordnet. https://wordnet.princeton.edu/download/current-version.
Richter, Raffael. 2026. “Evaluation Script (Clip Evaluation, Visualisations) for p… · Esteeschwarz/SPUND-LX@01d293b.” https://github.com/esteeschwarz/SPUND-LX/commit/01d293bfa731f80944ec1298699c15543d6dbcd7.
Schwarz, St. 2026. “This Paper Scripts.” https://github.com/esteeschwarz/SPUND-LX/tree/main/mental-img.
Tucker, Benjamin V., Daniel Brenner, D. Kyle Danielson, Matthew C. Kelley, Filip Nenadić, and Michelle Sims. 2019. “The Massive Auditory Lexical Decision (MALD) Database.” Behavior Research Methods 51 (3): 1187–1204. https://doi.org/10.3758/s13428-018-1056-1.
Wu, Si. 2025. “Swsiwu/Composition_and_deformance.” https://github.com/swsiwu/composition_and_deformance.
Yates, Andrew, Bart Desmet, Emily Prud’hommeaux, Ayah Zirikly, Steven Bedrick, Sean MacAvaney, Kfir Bar, Molly Ireland, and Yaakov Ophir, eds. 2024. Proceedings of the 9th Workshop on Computational Linguistics and Clinical Psychology (CLPsych 2024). St. Julians, Malta: Association for Computational Linguistics. https://doi.org/10.18653/v1/2024.clpsych-1.0.