From 53b6c4aa83fabd7e889ca3035c287a5d43ceeee3 Mon Sep 17 00:00:00 2001 From: Marta Sokol <116507237+msokol99@users.noreply.github.com> Date: Thu, 23 May 2024 23:59:27 +0200 Subject: [PATCH] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 613f3df..9906d63 100644 --- a/README.md +++ b/README.md @@ -7,7 +7,7 @@ a student of Cognitive Science at Osnabrueck University in the summer semester 2 ## Abstract

-Emerging evidence from phenomenological research suggests that borderline personality disorder (BPD) is associated with altered time perception, yet empirical support for this claim is sparse. This study compared differences in subjective time perception between individuals with BPD and an age- and sex-matched control group, as expressed in Polish autobiographical texts (N=10). A computational model based on a fine-tuned transformer neural network was trained for sentiment analysis and supplemented with classical natural language processing techniques. This model extracted objective language-based proxy variables, such as word length and frequency, as well as it predicted subjective sentiments to reflect various aspects of time perception. Outputs were analyzed as a time series, where each sentence in a single autobiographical corpus was assigned a time perception value, representing the perception of time by an individual participant. The resulting time series were compared across the groups utilizing sample entropy and autocorrelation as indicators of the short- and long-term predictability and orderliness of time perception. The results reveal a significantly lower sample entropy (p < .001) and a non-significant, yet marked decrease in autocorrelation (p = .120) in the BPD group compared to the control group, indicating altered time perception in BPD individuals. This suggests that the temporal experience of BPD patients is characterized by the presence of repetitive, predictable motifs, accompanied by relatively high short-term temporal state variability. This study not only provides more accurate understanding of how BPD affects time perception but also highlights the value of computational methods in exploring the subjective experiences associated with psychiatric conditions. +Emerging evidence from phenomenological research suggests that borderline personality disorder (BPD) is associated with altered time perception; however, empirical support for this claim is sparse. This study used Polish autobiographical texts to compare differences in subjective time perception between individuals with BPD (n = 5) and a sex-matched control group (n = 5). A computational model based on a fine-tuned transformer neural network was trained for sentiment analysis and supplemented with classical natural language processing techniques. This model extracted objective language-based proxy variables such as word length and frequency and predicted subjective sentiments to estimate time perception in real-time. Outputs were analyzed as a time series where each sentence in a single autobiographical corpus was assigned a time perception value representing the perception of time by an individual participant. The resulting time series were compared across the groups using sample entropy and autocorrelation as indicators of time perception's short- and long-term predictability and orderliness. The results reveal significantly lower sample entropy (p < .001) and significantly decreased autocorrelation (p = .032) in the BPD group compared to the control group, indicating altered time perception in BPD individuals. These findings suggest that the temporal experience of BPD patients is characterized by repetitive, predictable motifs accompanied by relatively high short-term temporal state variability. This study not only provides a more accurate understanding of how BPD affects time perception but also highlights the value of computational methods in exploring the subjective experiences associated with psychiatric conditions.

## Directory structure