Ontology-driven conceptual modeling in Cognitive Neuroscience: A way towards reusable data?

Managing the vast amounts of data produced daily might very well be one of the most critical challenges Cognitive Neuroscientists need to face within the next decade. However, it is no longer the sheer volume that poses a challenge, but mainly the variety of data. To answer complex questions, we need to integrate different sources of data as well as the interpretations derived from there. Thus, it is inevitable to address the issue of semantic interoperability. Semantic interoperability focuses on disambiguating the meaning of data to enable exchange between multiple information systems. This, in turn, requires harmonizing the meaning of domain-relevant concepts, their properties, and interrelations. So far, knowledge on the relations between our objects of research, theoretical accounts, experimental settings, methodological choices, and analysis decisions mainly guides our research process implicitly. Yet, to guarantee semantic interoperability and hence knowledge discovery, we are obliged to make our implicit knowledge explicit. Ontology-driven conceptual modeling is a technique in Computer Science that enables us to explicate our knowledge unambiguously so that it becomes machine-actionable. Here, we exemplify the expressivity of the theoretical framework and methodological environment provided by the Unified Foundational Ontology with a model fragment. Ultimately, a comprehensive ontology-based annotation system for data indexing and retrieval can act as an essential backbone for effective reusability, rendering data semantically situated and clearly described within a controlled meta-data schema.