Following PRISMA-ScR guidelines, this scoping review systematically maps the landscape of Large Language Models (LLMs) in mechanical engineering. A search of four major databases (Scopus, IEEE Xplore, ACM Digital Library, Web of Science) and a rigorous screening process yielded 66 studies for final analysis. The findings reveal a nascent, rapidly accelerating field, with over 68% of publications from 2024 (representing a year-on-year growth of 150% from 2023 to 2024), and applications concentrated on front-end design processes like conceptual design and Computer-Aided Design (CAD) generation. The technological landscape is dominated by OpenAI’s GPT-4 variants. A persistent challenge identified is weak spatial and geometric reasoning, shifting the primary research bottleneck from traditional data scarcity to inherent model limitations. This, alongside reliability concerns, forms the main barrier to deeper integration into engineering workflows. A consensus on future directions points to the need for specialized datasets, multimodal inputs to ground models in engineering realities, and robust, engineering-specific benchmarks. This review concludes that LLMs are currently best positioned as powerful ‘co-pilots’ for engineers rather than autonomous designers, providing an evidence-based roadmap for researchers, practitioners, and educators.

Following PRISMA-ScR guidelines, this scoping review systematically maps the landscape of Large Language Models (LLMs) in mechanical engineering. A search of four major databases (Scopus, IEEE Xplore, ACM Digital Library, Web of Science) yielded 66 studies for analysis. The findings reveal a nascent, rapidly accelerating field, with over 68% of publications from 2024, and applications concentrated on front-end design processes like conceptual design and Computer-Aided Design (CAD) generation. The technological landscape is dominated by OpenAI's GPT-4 variants. A persistent challenge identified is weak spatial and geometric reasoning, shifting the primary research bottleneck from traditional data scarcity to inherent model limitations. This, alongside reliability concerns, forms the main barrier to deeper integration into engineering workflows. A consensus on future directions points to the need for specialized datasets, multimodal inputs to ground models in engineering realities, and robust, engineering-specific benchmarks. This review concludes that LLMs are currently best positioned as powerful 'co-pilots' for engineers rather than autonomous designers, providing an evidence-based roadmap for researchers, practitioners, and educators.

Humans’ sense of the passage of time is subjective and determined by psychophysiological responses to the environment. The passing of time has been perceived to significantly slow in stressful situations, such as accidents or virtual threats. The current study will explore distortions in the perception of passage of time when threat is simulated using virtual reality (VR). 44 participants negotiated a large (13.6 × 8.4 m) virtual environment designed to maximise the realism of a physical threat by exposing participants to a virtual height of 200m. Subjective perception of passage of time and time estimation were used as independent variables, whereas the movement of participants, and ambulatory psychophysiology, e.g., electrodermal activity (EDA), heart rate, served as dependent variables. The independent variables were examined in relation to the dependent variables through a regression analysis, which allowed for the identification of the specific weight of each variable. Our analyses revealed that passage of time was perceived to decrease (i.e., time slowed down) for those participants who exhibited the higher levels of skin conductance (SCL). It is argued that this finding can be explained by individual differences in self-regulatory strategies during the task and the effectiveness of VR as means to simulate threat.