Safran Safar Almakaty, Mohammad Ali Alquaary • Semantic Scholar
Finding: A systematic review of 112 studies proposes the Need–Gratification–Dependency (NGD) cycle, arguing that users engage media to satisfy needs (Uses & Gratifications), but algorithmic personalization and persuasive design features create dependency that reshapes those needs into a self-reinforcing loop.
Why it matters: Directly addresses persuasive technology, habit formation, addiction by design, dark patterns (infinite scroll, variable-reward notifications), and the behavioral mechanisms (dopaminergic reward, variable reinforcement) central to behavior change design and user engagement ethics.
Method: PRISMA-guided systematic review of 112 peer-reviewed studies (2015–2024).
Jargon: *Media Dependency Theory (MDT)*: users become structurally reliant on media systems to meet needs; *Uses and Gratifications (U&G)*: users actively choose media to satisfy specific goals; *guided activeness*: user agency that is real but constrained and shaped by algorithmic architecture; *variable reinforcement*: unpredictable reward schedules that strengthen compulsive behavior (slot-machine logic).
Finding: A six-step problem-solving framework grounded in Kahneman's dual-process theory (System 1/fast vs. System 2/slow thinking), implemented via structured journals, produced 24% higher exam scores than online homework alone in introductory physics.
Why it matters: Directly applies behavioral science (dual-process theory, cognitive biases) to learning and skill acquisition, with implications for designing interventions that scaffold deliberate thinking.
Method: Controlled classroom study comparing students using structured written journals vs. online-only homework.
Mohammad Atari, Ivan Kroupin, Liora Morhayim et al. (6 authors) • PsyArXiv
Finding: Psychological science suffers not just from WEIRD-participant bias but also WEIRD-question bias, where researchers sample topics and constructs from a narrow cultural slice, leaving many cognitive, emotional, and behavioral phenomena understudied.
Why it matters: Directly relevant to behavioral science and research methodology — raises important questions about the generalizability of findings in decision-making, cognitive biases, and behavior change research that professors draw on.
Jargon: *WEIRD* = Western, Educated, Industrialized, Rich, Democratic — shorthand for the narrow population slice that dominates psychological research; *conceptual sampling* = the selection of which topics/constructs to study, as distinct from sampling of participants.
⚠️ No control group mentioned | Single study (not replicated) | Limited statistical detail in abstract
Finding: A WeChat-based microlearning intervention improved undergraduate English speaking skills and self-regulation over five weeks, with gains explained through five mechanisms including scaffolding, peer support, and feedback loops.
Why it matters: Marginally relevant to learning sciences and habit/skill acquisition, but the focus on oral communication and graduate employability is peripheral to the professor's core interests.
Laura Vásquez-Rodríguez, Bertrand Audrin, S. Michel et al. (7 authors) • Semantic Scholar
Finding: AI systems (especially LLMs) extract skills from resumes differently than humans—LLMs produce longer, more detailed skill descriptions but are semantically closer to human annotations than traditional systems.
Why it matters: Tangentially relevant to AI's impact on work, but focused narrowly on HR tech pipeline mechanics rather than human collaboration, behavior, or learning.