Finding: This meta-analysis of 28 articles reveals that research on high-impact climate behaviors (like reducing car use, plant-based diets) overwhelmingly focuses on motivational factors (attitudes, values) while largely ignoring capability and structural opportunity factors that may be crucial for behavior change.
Why it matters: The systematic bias toward studying motivation over capability and opportunity directly relates to behavioral science research on what drives decision-making and behavior change, with implications for designing more effective interventions.
Method: Uses the Behaviour Change Wheel's COM-B framework (Capability, Opportunity, Motivation) to systematically categorize determinants across 32 studies of climate behaviors.
Jargon: COM-B system - framework categorizing behavior determinants into Capability (physical/psychological ability), Opportunity (social/physical environment), and Motivation (reflective/automatic processes).
Finding: This theoretical analysis argues that Graham Wallas's famous four-stage creativity model (preparation, incubation, illumination, verification) has been divorced from his lived ethical orientation toward thought, losing crucial context about creativity as a way of being rather than just a process to optimize.
Why it matters: This challenges how creativity is taught and researched by questioning whether stage-based models miss the deeper human dimensions of creative work—judgment, responsibility, and sustained commitment—that may be essential for meaningful innovation.
Method: Historical reconstruction through archival analysis of Wallas's writings and contemporaneous accounts to recover the person behind the model.
Jargon: The four stages refer to Wallas's influential creativity framework: preparation (gathering information), incubation (unconscious processing), illumination (insight moment), and verification (testing ideas).
Finding: Creative insight involves a distinct neurological state called "intimation"—a calibrated instability where the brain maintains heightened sensitivity and anticipation without committing to specific solutions, regulated by cerebellar precision control and oscillatory containment.
Why it matters: This provides a concrete neurocognitive framework for understanding and potentially designing interventions to enhance creative thinking processes in design, problem-solving, and collaborative work.
Method: Theoretical synthesis integrating oscillatory research, seizure dynamics, and neurocomputational models to formalize the pre-insight state.
Jargon: Intimation = the anticipatory "solution is coming" feeling before creative insight; calibrated instability = maintaining heightened neural sensitivity without premature commitment; cerebellar precision = cerebellum's role in regulating neural threshold dynamics rather than generating ideas.
Finding: The paper proposes a neuroscience-based framework for understanding creative "a-ha!" moments, arguing that illumination occurs at an optimal "edge-of-chaos" balance between excitatory prediction errors and inhibitory control, requiring openness and accurate interoception.
Why it matters: This provides a mechanistic understanding of creative insight that could inform design of creativity interventions, HCI systems that support ideation, and team environments that foster breakthrough thinking.
Method: Theoretical integration of Wallas's classic creativity stages with predictive processing neuroscience frameworks.
Jargon: Predictive processing - brain theory where cognition involves predicting sensory input and updating when predictions fail; interoception - awareness of internal bodily signals; edge-of-chaos - optimal zone between order and disorder where complex systems exhibit peak performance.
Gunnar Epping, Andrew Caplin, Erik Duhaime et al. (6 authors) • PsyArXiv
Finding: Applying cognitive science principles to "de-bias" crowdsourced datasets through probability recalibration significantly improves both dataset quality and the performance of AI models trained on them, demonstrated across novice and expert annotators on medical image tasks.
Why it matters: This directly addresses how cognitive biases affect human-AI collaboration in data creation, offering practical methods to improve crowdsourced training data quality—crucial for understanding behavioral science applications in AI development.
Method: Two experiments comparing standard crowdsourcing against cognitive-inspired data engineering using probability recalibration techniques on Amazon Mechanical Turk and DiagnosUs platforms for medical image annotation.
Jargon: Recalibration - adjusting probability estimates to better match actual outcomes; cognitive-inspired data engineering - applying cognitive science insights to improve data collection processes.
⭐ Includes replication ⚠️ No control group mentioned | Convenience sample (e.g., MTurk, students) | Limited statistical detail in abstract
Finding: This theoretical paper proposes a "Feynman-honest AI" framework for digital mental health, arguing that current AI systems systematically harm neurodivergent users by encoding bias and stigma, and offers five criteria for more ethical AI development including scope clarity, uncertainty communication, and participatory co-development.
Why it matters: This directly addresses how AI impacts human collaboration and decision-making in high-stakes contexts, while highlighting cognitive biases in AI systems and proposing human-centered design principles that could apply broadly to HCI and behavioral science applications.
Method: Theoretical framework development drawing on scientific integrity principles, neurodiversity paradigm analysis, and predictive processing theory.
Jargon: Neurodiversity paradigm - viewing neurological differences as natural human variation rather than deficits; Predictive processing - cognitive theory that the brain constantly generates predictions about sensory input.
Finding: Higher perceived empathy from AI chatbots increases dependence on AI and reduces human social connection among emerging adults, with AI dependence partially mediating the relationship between perceived AI empathy and social disconnection.
Why it matters: This demonstrates how AI design choices (empathetic responses) can unintentionally undermine human collaboration and social skills that are crucial for effective teamwork and collective intelligence.
Method: Correlational survey study (N=242) using validated scales for empathy, AI dependency, and social connectedness with mediation analysis.
Jargon: Emerging adults = individuals typically aged 18-29 in a developmental stage between adolescence and full adulthood.
Finding: The paper introduces a framework called "the 7 Muses" that reframes creativity as regulatory infrastructure for mental health, organizing seven capacities (Independence, Curiosity, Playfulness, Confidence, Openness, Interdependence, and Passion) around cycles of effort, restoration, connection, and purpose-aligned action.
Why it matters: This creativity-based regulatory framework could inform design of team dynamics interventions, goal-setting approaches, and understanding how creative processes support motivation and behavioral change.
Method: Theoretical framework development grounded in Wallas's creative process model and contemporary emotion/motivation research, with a brief self-report assessment tool.
Jargon: Wallas's Preparation–Imagination–Verification cycle refers to a classic model of the creative process involving initial preparation, imaginative exploration, and final verification/implementation phases.
Finding: Statistical learning research is hindered by three major debates (executive function relationships, developmental trajectories, and offline state impacts) that can be resolved by recognizing that cognitive tasks are never "process-pure" and involve multiple overlapping mechanisms.
Why it matters: Understanding how people extract patterns and build predictive models is crucial for designing effective learning experiences, AI-human collaboration systems, and skill acquisition programs.
Method: Proposes applying Jacoby's process dissociation framework to disentangle auxiliary cognitive processes from core probability detection mechanisms.
Jargon: Statistical learning = unconscious pattern detection from environmental regularities; Process dissociation = method for separating different cognitive processes within the same task; Process-pure = falsely assuming a task measures only one cognitive mechanism.
Anna Junga, Ole Hätscher, Jennifer Dabel et al. (10 authors) • PsyArXiv
Finding: PerTRAIN uses large language models to create virtual patients that dynamically adapt their personality and interpersonal behavior based on Contemporary Integrative Interpersonal Theory, allowing medical students to practice personalized communication skills at scale.
Why it matters: This demonstrates how AI can enable scalable skill acquisition in interpersonal domains, offering insights for designing adaptive learning systems that respond to human behavior in real-time.
Method: The system models patient behavior along agency and communion dimensions that update dynamically in response to trainee interactions, creating a feedback loop for learning adaptive communication strategies.
Jargon: CIIT (Contemporary Integrative Interpersonal Theory) - framework for understanding interpersonal behavior along two dimensions: agency (dominance/submission) and communion (friendliness/hostility).
Xiaoxiao Chen, Don Hine, Claudia R. Schneider • PsyArXiv
Finding: Specific linguistic features in video messages (anger, power, certainty words) increase motivation for climate-friendly dietary choices, while others (sadness, achievement words) decrease motivation, with effects varying significantly across three distinct audience segments.
Why it matters: This demonstrates how subtle communication design choices can systematically influence behavior change motivation, providing actionable insights for designing persuasive interfaces and messages in sustainability applications.
Method: Randomized experiment with 711 participants viewing climate videos analyzed using LIWC (Linguistic Inquiry and Word Count) dictionary, combined with audience segmentation based on values and efficacy measures.
Jargon: LIWC - automated text analysis tool that categorizes words by psychological and linguistic dimensions.
⚠️ No control group mentioned | Single study (not replicated)
Finding: PSlides is an open-source JavaScript library that simplifies creating browser-based psychological experiments and surveys by extending HTML with custom elements, eliminating the need for complex frameworks while maintaining full experimental control and data logging capabilities.
Why it matters: This tool directly supports HCI research methods and behavioral science experimentation by making it easier for researchers and students to create interactive studies without advanced programming skills.
Method: The authors developed a dependency-free library with HTML-first approach, demonstrated through example paradigms like number comparison tasks and surveys.
Jargon: DOM = Document Object Model (the structure of web page elements that can be manipulated); setTimeout = JavaScript function for precise timing control; JSON/CSV = common data export formats.
⚠️ Self-reported data only | No control group mentioned | Single study (not replicated) | Limited statistical detail in abstract
Jan Urban, Tomasz Grzyb, Dariusz Dolinski et al. (5 authors) • PsyArXiv
Finding: Increasing worry about one issue causes small but measurable increases in worry about unrelated issues, mediated by generalized affect and moderated by issue similarity, supporting both generalized affect and availability heuristic mechanisms.
Why it matters: This demonstrates how cognitive biases (availability heuristic) and emotional states influence decision-making across unrelated domains, which is crucial for understanding how people make choices and set priorities in work and learning contexts.
Method: Large-scale web experiment (N=7,051) with worry manipulation followed by measurement of worry about 12 unrelated issues, testing mediation and moderation effects.
Jargon: Availability heuristic - tendency to judge likelihood/importance based on how easily examples come to mind; finite pool of worry hypothesis - idea that people have limited worry capacity that gets distributed across issues.
⚠️ Very small sample (n=7) | No control group mentioned | Single study (not replicated)
Alexandra Munns, Jack Feron, Joan Duda et al. (10 authors) • PsyArXiv
Finding: Older adults with different motivation profiles (high vs. moderate autonomous motivation and self-efficacy) showed similar objective exercise adherence and fitness outcomes in a structured home-based program, suggesting that good program design and support can overcome individual psychological differences.
Why it matters: This challenges assumptions about motivation's role in behavior change and suggests that well-designed systems with monitoring and support can be more important than individual psychological factors for sustained engagement.
Method: Used latent profile analysis to identify motivation profiles and objective heart rate-based measures to track exercise adherence over 26 weeks.
Jargon: Autonomous motivation = intrinsic, self-determined reasons for exercising; Controlled motivation = external pressures or guilt-driven reasons; MET-mins = metabolic equivalent minutes, a standardized measure of exercise volume.
Marco Bernardi, Luigi Spadafora, L. D. Dalla Vecchia et al. (7 authors) • Semantic Scholar
Finding: AI is transforming cardiology through diagnostic tools, risk prediction, and treatment planning, but requires careful human oversight and governance to prevent harmful autonomous decision-making, drawing parallels to HAL 9000's cautionary tale.
Why it matters: Provides a domain-specific case study of AI's impact on professional work and human-AI collaboration, illustrating broader patterns of how AI augments versus replaces human expertise.
Jargon: HAL 9000 - the AI computer from "2001: A Space Odyssey" that becomes dangerously autonomous; ECGs - electrocardiograms, heart rhythm recordings.
Finding: This theoretical paper argues that metaphor works through cross-domain extension based on property clustering mechanisms, using dog trail cognition as a case study to illustrate how metaphorical thinking might operate across functional domains.
Why it matters: The framework could inform how people use metaphorical thinking in creative design processes and how AI systems might be designed to support analogical reasoning in human collaboration.
Jargon: HPC (Homeostatic Property Cluster) theory - categories are defined by clusters of properties that reinforce each other rather than by necessary and sufficient conditions; CMT (Conceptual Metaphor Theory) - the idea that metaphors structure how we think, not just how we speak.
Erica Martin, Celine van Golde, Lauren Monds • PsyArXiv
Finding: Mock jurors viewed intoxicated crime victims as less credible than sober victims, but expert testimony did not improve juror decision-making or reduce bias against intoxicated victims.
Why it matters: This demonstrates how cognitive biases systematically distort group decision-making processes, even when expert information is provided to counteract those biases.
Method: Between-subjects experiment with mock jury deliberations comparing individual vs. group verdicts across different victim intoxication levels and expert testimony conditions.
Jargon: Mock jury - simulated jury using research participants rather than actual jurors.
⚠️ Single study (not replicated) | Limited statistical detail in abstract
Finding: Tikriti Arabic speakers show rigid, fossilized pronunciation patterns when producing English /r/ sounds, with AI successfully detecting these entrenched patterns with 92% accuracy using acoustic and neural network features.
Why it matters: Demonstrates how AI can be used to detect entrenched behavioral patterns in skill acquisition, relevant to understanding learning plateaus and automatization in other domains.
Method: Combined traditional acoustic analysis with AI-based speech representations (Wav2Vec 2.0) and machine learning classification to identify fossilized pronunciation patterns.
Jargon: Fossilization = permanent retention of non-native patterns in language learning; F1-F3 = acoustic frequencies that characterize speech sounds; Wav2Vec 2.0 = AI model for speech processing.
⚠️ No control group mentioned | Single study (not replicated)
Finding: This paper proposes an explainable AI system for anomaly detection in industrial settings, focusing on making AI decisions more interpretable and trustworthy in 6G-enabled smart industries.
Why it matters: The explainable AI component relates to human-computer interaction concerns about AI transparency and trust, though the industrial context is outside the core teaching areas.
Method: Uses variational autoencoder with attention bidirectional LSTM combined with recursive feature elimination for anomaly detection.
Jargon: XAI (Explainable AI) - AI systems designed to make their decision-making processes interpretable to humans; Industry 5.0 - industrial paradigm emphasizing human-machine collaboration; VAE - Variational Autoencoder, a type of neural network for learning data representations.
Njabulo Ndlovu, P. Sifolo, N. Tshipala • Semantic Scholar
Finding: This bibliometric review reveals that while AI research in tourism is growing rapidly (especially post-2019), very few studies specifically address small and medium tourism enterprises, with most research focusing on larger organizations or using generalized models that don't account for SMTEs' unique constraints.
Why it matters: The gap between AI research and small enterprise needs illustrates broader challenges in designing accessible AI tools for resource-constrained organizations, relevant to understanding AI's differential impact across organizational contexts.
Method: Bibliometric analysis of 373 articles from Scopus database using VOS Viewer software for network mapping and thematic clustering.
Jargon: SMTEs = Small and Medium Tourism Enterprises; bibliometric analysis = statistical analysis of published research patterns and trends.