Measuring insight is a central challenge in the problem-solving field. Recent studies show that classic insight tasks can be solved through either insightful thinking or analytical thinking. The present work examined relationships between cognitive reasoning tasks and insight tasks. Additionally, we explored differences in insight phenomenology when solving insight puzzles and when completing the Cognitive Reflection Test. We conducted two independent studies. Study 1 was a conceptual replication of previously observed relationships between insight puzzles, the Cognitive Reflection Test (CRT), and the Verbal Cognitive Reflection Test (VCRT). Study 2 examined differences between analytical and insightful tasks using self-report insight scales. Five dimensions of insight were measured: suddenness, confidence, pleasure, the Aha! experience, and arousal. Across both studies, insight-puzzle performance correlated positively with reflective tests (CRT and VCRT), consistent with prior findings. In Study 2, Aha! ratings did not differ between insight puzzles and CRT. Unexpectedly, Aha! experience was stronger for VCRT than for insight puzzles. Trial-level analyses that incorporated response accuracy further indicated that Aha! ratings in VCRT and CRT did not reliably differentiate correct from incorrect responses, whereas accuracy effects emerged in other task comparisons. Overall, the dissociation between performance overlap and phenomenology suggests that subjective insight is not uniquely tied to tasks traditionally classified as insight problems and that Aha! experience alone should not be treated as a direct marker of restructuring.
Introduction: The aim of this study was to adapt the full 45-item, seven-subscale Intrinsic Motivation Inventory (IMI) into Polish and evaluate its psychometric properties, including cross-language measurement invariance in Polish–English bilinguals.
Methods: Two independent online samples were pooled (N = 275), with participants completing the IMI in both Polish and English. Reliability (α, ω), cross-language correlations, and confirmatory factor analyses (CFA; WLSMV/THETA) were conducted, while language invariance was tested using a CT-C(M–1) framework comparing marker-configural and strict metric models. Convergent and divergent validity were assessed through associations with external constructs related to motivation, control, curiosity, and stress appraisal.
Results: CFA supported the expected seven-factor structure in both languages with good model fit, and language invariance analysis indicated minimal configural–metric differences, suggesting approximate measurement invariance. The Polish IMI demonstrated strong internal consistency across all subscales, closely matching the English version, while cross-language correlations confirmed strong subscale-level equivalence, though a small number of items showed weaker psychometric properties. Subscale intercorrelations followed theoretical expectations, supporting construct validity, with subscales reflecting engagement, autonomy, and competence aligning most strongly with positive motivational constructs. Overall, the Polish IMI is suitable for research in Polish, with all materials, data, and analysis code openly available.
Tischner and Stasiuk (IIC 54:26–60 (2023), https://doi.org/10.1007/s40319-022-01274-8) concluded that the Audi trademark does not influence the evaluation of independently manufactured automotive spare parts, arguing that both consumers and experts interpret such trademarks primarily as descriptive cues of intended use rather than as indicators of origin. The present paper revisits their empirical evidence and challenges this conclusion by reanalyzing the original raw dataset made publicly available in the OSF repository. We argue that the original authors’ conclusions stem from a methodological error – specifically, the aggregation of evaluations across participants who differed fundamentally in their perceptions of the part’s manufacturer. Using the original data, we reclassified both Audi owners and automotive experts according to the manufacturer they recalled after product exposure: Audi, the independent manufacturers named in the description, or other/unknown manufacturers. Separate analyses of variance were then conducted within these groups for four evaluation dimensions: perceived quality, material durability, appearance, and purchase intention. The reanalysis focused on the radiator grille stimulus, across four trademark presentation conditions. Contrary to Tischner and Stasiuk’s conclusions, the results consistently show that brand recall significantly shaped product evaluations. Both consumers and experts who mistakenly identified Audi as the manufacturer evaluated the spare parts more favorably than those who correctly identified independent manufacturers or expressed uncertainty. These effects were particularly pronounced when the Audi trademark was embedded in the product or visually integrated into its design, and in several conditions reached statistical significance. The high rate of manufacturer misidentification – despite explicit textual information – underscores the influential role of trademarks as associative signals affecting perceived quality and value. Overall, this reanalysis demonstrates that the Audi trademark did influence the evaluation of automotive spare parts. The original claim that trademarks have lost their origin-identifying and evaluative function is therefore unsupported and inconsistent with both the empirical evidence and the broader literature on branding and consumer perception.
Pozostałe osiągnięcia naukoweArtykuły (zamknięty dostęp)Journal article
The aim of this article is to develop and validate the AI-MEVAL Model, which integrates multi-stakeholder evaluation with the AI Feedback Loop, in order to enhance the reliability of feedback. Qualitative research confirms that the model shortens the PDCA cycle, provided that transparency and the Human-in-the-Loop principle are maintained. It constitutes a practical tool supporting organisational learning.
