Same words, different worlds: Conceptual consistency in systems neuroscience

This workshop brings together philosophers, experimentalists, theorists, and engineers to tackle an important but often overlooked source of variability across studies in systems neuroscience: conceptual and methodological drift. While there are growing efforts to establish standardized preprocessing or statistical modeling approaches across labs, much less attention is paid to the impact of inconsistent use of core concepts when researchers use the same words to mean slightly different things. Concepts like “causality”, “connectivity”, or “representation”, are central to how we formulate questions, choose experimental and analytical tools, and interpret results. Yet these terms often carry different meanings across subfields, especially in an inherently interdisciplinary field, such as systems neuroscience. This heterogeneity carries a high risk of conceptual fragmentation, leading to distinct analytical strategies and ultimately to incompatible results. To promote scientific integrity and reproducibility, it is therefore crucial to identify and understand the key conceptual differences that shape how scientific conclusions are drawn. The workshop directly tackles this challenge in its full complexity. Speakers will examine how abstract concepts such as causality or modularity are conceptually defined, and how these definitions can be operationalized. The workshop will also cover seemingly concrete phenomena, such as oscillations, sharp wave-ripples and intrinsic timescales, to examine how even well-established phenomena are shaped by conceptual assumptions and methodological decisions. Talks will range from theory to experimental practice, including the logic behind analysis standardization tools and the epistemology of mechanistic explanations. The format and topic of the workshop are well-suited to promote active participation and open debate. By design, the workshop aims to surface controversial issues and encourage discussing competing approaches and implicit assumptions.