Unrealistic optimism is a phenomenon in which a person perceives oneself as less threatened than another person in the same situation leading to the feeling of control and fear reduction during and before threatening events. A small amount of research on unrealistic pessimism delivers the opposite picture: people perceive themselves as more exposed to negative consequences compared to others. The consequences of these biased perceptions of reality are seminal: the first leads to an increased risk of disregarded negative consequences, whilst the second leads to proactive actions toward self-protection. The present paper tries to replicate the unrealistic pessimism effect in a new context: floods. Surprisingly, results show that people affected by floods report unrealistic optimism. Non-affected participants shared this bias, described as the ‘spillover effect.’ Finally, this illusion persists over time, even when the risk is no longer present, challenging its fear-reduction explanation.
Post-traumatic stress disorder describes a chronic and impairing response to extreme stress that affects daily functioning. One of the most wellknown and commonly used tools for measuring post-traumatic stress disorder symptoms is the PTSD Checklist for DSM-5. This tool has been adapted to many different cultures, but measurement equivalence across cultures has only been examined in a single study of European countries. The aim of these analyses was to verify full measurement equivalence of the PTSD Checklist for DSM-5. Data from eight countries (N > 200) were included: Poland, USA, Japan, Argentina, South Korea, Ireland, United Kingdom, and Kazakhstan, collected during the second, third, and fourth waves of the COVID-19 pandemic. A symptom structure model based on DSM-5 presented good fit to the data, but the best fit was found for a hybrid seven-factor model; measurement equivalence was verified with respect to the latter model. Confirmation was obtained for equivalence in terms of identical structure and factor loadings (configural and metric invariance), but not in terms of intercepts and residuals (scalar and residual invariance). Post-traumatic stress disorder symptoms were positively correlated with the levels of depressive, anxiety, and stress symptoms (strongly or moderately) and with COVID-19-related stressors as well as emotional reactivity (moderately or weakly) in all samples. The results allow us to conclude that the PTSD Checklist for DSM-5 measures culturally universal indicators of post-traumatic stress disorder, but measurement equivalence should be verified in each cross-cultural study to confirm scalar invariance. In future research on PTSD Checklist for DSM-5 measurement equivalence, efforts should be made to select equally sized and demographically homogeneous samples.
Purpose: This study examines the usage patterns of artificial intelligence (AI) tools among 859 university students in Poland, representing three major academic fields. The research aims to identify discipline-specific differences in AI adoption, usage motivations, perceived benefits, and challenges, contributing to evidence-based strategies for advancing Sustainable Development Goal 4 (Quality Education). Design/methodology/approach: An exploratory quantitative approach was employed using a cross-sectional online survey conducted from February to May 2025. Data from 859 students across humanities and social sciences, natural sciences and technology, and arts and communication were analysed using descriptive statistics and nonparametric methods (KruskalWallis and DSCF post-hoc tests). Findings: Significant inter-group differences were identified in AI adoption patterns. Humanities students primarily use AI for knowledge expansion (54%) and translation (48%). Natural sciences students emphasise problem-solving (51%) and knowledge acquisition (62%). Arts students focus on translation (63%) and creative content generation (46%). Despite widespread adoption, students across all disciplines express concerns about data privacy, ethics, and employment impacts. Research limitations/implications: Convenience sampling and unequal group sizes limit generalizability. The cross-sectional design captures a single point during rapid technological change. Future longitudinal and comparative studies are recommended. Practical implications: Universities should develop discipline-specific AI integration strategies, targeted professional development for faculty, and comprehensive AI literacy curricula. Preparing graduates for the evolving business landscape requires aligning AI literacy, critical thinking, and ethical competencies with industry expectations. Social implications: The findings underscore the need for inclusive and equitable AI integration in higher education to prevent new forms of digital inequality, contributing to sustainable economic growth, digital transformation, and responsible workforce development. Originality/value: This study provides empirical evidence on discipline-specific AI adoption patterns in the underexplored Central and Eastern European context, documenting usage during the critical transition from experimental to systematic AI integration in higher education.
