Andrea Guazzini

Pittawat Taveekitworachai, Chollakorn Nimpattanavong, Mustafa Can Gursesli et al.

This paper presents Dynamic Context Prompting/Programming (DCP/P), a novel framework for interacting with LLMs to generate graph-based content with a dynamic context window history. While there is an existing study utilizing LLMs to generate a visual novel game, the previous study involved a manual process of output extraction and did not provide flexibility in generating a longer, coherent story. We evaluate DCP/P against our baseline, which does not provide context history to an LLM and only relies on the initial story data. Through objective evaluation, we show that simply providing the LLM with a summary leads to a subpar story compared to additionally providing the LLM with the proper context of the story. We also provide an extensive qualitative analysis and discussion. We qualitatively examine the quality of the objectively best-performing generated game from each approach. In addition, we examine biases in word choices and word sentiment of the generated content. We find a consistent observation with previous studies that LLMs are biased towards certain words, even with a different LLM family. Finally, we provide a comprehensive discussion on opportunities for future studies.

Serena Coppolino Perfumi, Chiara Cardelli, Franco Bagnoli et al.

The main aim of our study was to analyse the effects of a virtual environment on social conformity, with particular attention to the effects of different types of task and psychological variables on social influence, on one side, and to the neural correlates related to conformity, measured by means of an Emotiv EPOC device on the other. For our purpose, we replicated the famous Asch's visual task and created two new tasks of increasing ambiguity, assessed through the calculation of the item's entropy. We also administered five scales in order to assess different psychological traits. From the experiment, conducted on 181 university students, emerged that conformity grows according to the ambiguity of the task, but normative influence is significantly weaker in virtual environments, if compared to face-to-face experiments. The analysed psycho-logical traits, however, result not to be relatable to conformity, and they only affect the subjects' response times. From the ERP (Event-related potentials) analysis, we detected N200 and P300 components comparing the plots of conformist and non-conformist subjects, alongside with the detection of their Late Positive Potential, Readiness Potential, and Error-Related Negativity, which appear consistently different for the two typologies.