1,720,991 research outputs found
Unmasking Lies: Advancements In Deception Detection Techniques
Full text linkPursuing truth through the unmasking of deception has consistently engaged the scientific community and the general public. This work delves into the intricate domain of lie detection, integrating various methodologies and theories to provide a comprehensive view of the challenges and potential of studying deceit within contemporary society. This dissertation unfolds across three distinct projects, each focusing on a unique approach to detecting deception in different scenarios: self-report assessment questionnaires, workplace drug testing, and face-to-face investigative interviews.
The first project examines using the Term Frequency-Inverse Document Frequency (TF-IDF) index to detect lies within personality questionnaires. Departing from traditional methods that construct comprehensive simulation profiles through control scales, this approach focuses on the detailed analysis of specific responses to identify instances of deceit at the single-item level. The project tests the efficacy of TF-IDF in discerning authentic from fabricated responses, offering a novel analytical tool for forensic psychology. To validate this methodology, three experiments were conducted in "faking good" scenarios, where participants are incentivized to present themselves in an overly positive manner. The results were notably promising: the TF-IDF model effectively distinguished between genuine and counterfeit responses. These findings open new avenues for research in forensic psychology, suggesting that TF-IDF could be a valuable tool for addressing complex challenges like detecting deception in questionnaires at the item level.
The second research project offers a preliminary analysis of drug use in workplaces and methods for its detection. The study investigates standard biological drug testing procedures and their legal and ethical implications, as well as the effectiveness of questionnaire-based tests for substance abuse detection, particularly highlighting their limitations in accuracy due to the possibility of deceptive responses. The research emphasizes the need for more innovative methodologies to effectively tackle this increasingly significant workplace safety concern. In response, this project focuses on introducing and validating a methodology based on the kinematic analysis of mouse movements. This method hypothesizes that mouse movements during responses to double-choice questions can reveal behavioral patterns associated with lying. Applied here, the research aims to detect deceptive responses in workplace drug tests through two online experiments, comparing behaviors of drug-using and non-drug-using employees and exploring the potential of this analysis to understand the mental processes involved in generating deceitful responses.
The third project delves into identity deception, a specific form of deceit where individuals intentionally conceal their identity, impersonate others, or use counterfeit identity documents. This type of deceit, which poses a serious threat to national security, has been exacerbated in the digital age, facilitating terrorists and criminals in evading security protocols, as evidenced by the use of false identities in terrorist attacks. To address this challenge, this research employed techniques such as the method of unexpected questions, utilizing the difficulty liars face in responding to unforeseen queries, and reaction time analyses in face-to-face investigative interviews. Machine learning analyses in this study offer insights into detecting individuals who lie achieving a remarkable level of accuracy.
In conclusion, this dissertation offers a multifaceted perspective on deception detection, highlighting the incorporation of computational tools to augment the accuracy of lie detection in diverse scenarios. This comprehensive exploration not only advances the scientific understanding of deceit but also addresses the practical implications and challenges in detecting lies
L'utilizzo del suolo nel rapporto di competitività delle aree urbane italiane
No full textCon il supporto dei fondi europei JESSICA ogni anno viene prodotto il "Rapporto di competitività delle aree urbane italiane": all'interno dello studio vengono monitorati dati sui tassi di urbanizzazione a livello provinciale, nonché sulla disponibilità di verde pubblico e altre infrastrutture "sostenibili" all'interno delle aree urbane (dotazione di verde pubblico per capoluogo di provincia). I risultati vengono elaborati in un indice sintetico che permette di valutare le performance provinciali su tutto il territorio nazionale. L'articolo si propone di fornire una lettura comparativa dei processi in atto, nonché una valutazione dei trend che hanno caratterizzato il nostro territorio negli ultimi dieci anni, mettendo in luce alcuni significativi aspetti sulle dinamiche in corso. In particolare, nel primo capitolo verrà introdotto il rapporto di competitività delle aree urbane spiegandone la struttura e la mission. Successivamente verrà presentato l'indicatore aggregato sull'utilizzo del suolo con riportate le più recenti analisi svolte nell'ultima edizione. In seguito l'autore approfondirà il fenomeno dell'impermeabilizzazione del suolo nelle province d'Italia e descriverà la situazione attuale basandosi sul risultato dell'elaborazione dei dati della Corine Land Cover. Infine a conclusione dell'articolo saranno presenti alcune riflessioni sul tem
GeoTIPO: geovisualization tool interactive planning oriented
No full textThe following research study describes the implementation of a tool that allows to collect and compare different types of data coming from several databases, with the aim of providing support to territorial planners. The tool is accessible from mobile devices and non-mobile ones (personal computer, tablet, smartphone), and therefore it can be used in many different contexts. Its most remarkable feature is the possibility of interaction with all kind of geo-related maps, in order to make the territory queryable through the use of filters and selections depending on the type of reference data. These characteristics make it possible to identify correlations between different phenomena in variable time frames, in order to quickly suggest or support potential scenarios or solutions. The following paper summarizes the implementation process of this tool, and shows some examples of application in the municipality of Turin, a city that has faced significant urban transformations over the past two decades. Example of application: mortality in urban areas, Turin (Progetto SOPHIE
Large language models and psychiatry
No full textIntegrating Generative Artificial Intelligence and Large Language Models (LLMs) such as GPT-4 is transforming clinical medicine and cognitive psychology. These models exhibit remarkable capabilities in understanding and generating human-like language, which can enhance various aspects of healthcare, including clinical decision-making and psychological counseling. LLMs, trained on vast datasets, function by predicting the next word in a sequence, endowing them with extensive knowledge and reasoning abilities. Their adaptability allows them to perform a wide range of language-related tasks, significantly contributing to advancements in cognitive psychology and psychiatry. These models demonstrate proficiency in tasks such as analogical reasoning, metaphor comprehension, and problem-solving, often achieving performance comparable to neurotypical humans. Despite their impressive capabilities, LLMs still exhibit limitations in causal reasoning and complex planning. However, their continuous improvement, exemplified by the enhanced performance of GPT-4 over its predecessors, suggests a trajectory towards overcoming these challenges. The ongoing debate about the “intelligence” of LLMs revolves around their ability to mimic human-like reasoning and understanding, a focal point of contemporary research. This paper explores the cognitive abilities of LLMs, comparing them with human cognitive processes and examining their performance on various psychological tests. It highlights the emergent properties of LLMs, their potential to transform cognitive psychology, and the different applications of LLMs in psychiatry, highlighting the limitations, the ethical considerations, and the importance of scaling and fine-tuning these models to enhance their capabilities. We also explore the parallels between LLMs and human error patterns, underscoring the significance of using LLMs as models for human cognition. Overall, this paper provides substantial evidence supporting the role of LLMs in reviving associationism as a viable framework for understanding human cognition while acknowledging the current limitations and the need for further research to fully realize their potential
- …
