The generative capacity of language entails an ability to flexibly combine concepts with each other. Conceptual combination can occur either by using an attribute of one concept to describe another (attributive combination) or by forming some relation between two concepts to create a new one (relational combination). Prior research has addressed whether common or distinct processes support these two putatively different types of combinations. We turn the question around and ask whether the consequences of these combination types on our conceptual system might differ, by comparing semantic memory networks before and after participants perform either attributive or relational conceptual combinations. We find a general effect on the semantic networks: the structure of network decreases after participants conceptually combine some of the concepts in the network. However, the relational combination manipulation has a greater effect. Furthermore, only the relational combination manipulation leads to an increase in the network’s connectivity.

The associative theory of creativity proposes that creative ability relies on the organization of semantic memory, yet the relationships between semantic memory structure and brain connectivity, in relation to creativity, remain poorly understood. Here, we explored the relationships between the network properties of individual semantic memory, patterns of functional brain connectivity, and real-life creativity. To this end, we acquired functional magnetic resonance imaging data while participants underwent a semantic relatedness judgment task. Participants’ relatedness ratings between word-pairs were used to estimate their individual semantic networks, whose network properties were significantly related to their real-life creativity. Using a connectome-based predictive modeling approach, we identified patterns of on-task functional connectivity that predicted creativity-related semantic memory network properties in novel individuals. Furthermore, the predicted semantic network properties partially mediated the relationship between functional connectivity and real-life creativity. These results provide new insights on how brain connectivity supports the associative mechanisms of creativity.

Creative thinking has long been associated with spreading ofactivation through concepts within semantic networks. Herewe examine one potential influence on spreading activationknown as the fan effect: increasing concept knowledge leads toincreasing interference from related concepts. We testedwhether cue association size—an index of semantic richnessreflecting the average number of elements associated with aconcept—impacts the quantity and quality of responsesgenerated during the alternate uses task (AUT). Wehypothesized that low-association AUT cues should benefitquality at the cost of quantity because such cues are embeddedwithin a semantic network with fewer conceptual elements,thus yielding lesser interference from closely-related concepts.This hypothesis was confirmed in Study 1. Study 2 replicatedthe effect and found an interaction with fluid intelligence,indicating that cognitive control can overcome constraints ofsemantic knowledge. The findings thus highlight costs andbenefits of semantic knowledge for creative cognition.

Cognition is complex. This complexity is related tomultiple, distributed neurocognitive processes dynamicallyoperating across parallel scales, resulting in cognitiveprocessing. A major challenge in studying this complexity,relates to the abstractness of theoretical cognitive constructs,such as language, memory, or thinking in general. Suchabstractness is operationalized, indirectly, via behavioral,measures or in neural activity. In the past two decades, anincreasing number of studies have been applying networkscience methodologies across diverse scientific fields tostudy complex systems.Network science is based on mathematical graph theory,providing quantitative methods to investigate complexsystems as networks (Baronchelli, Ferrer-i-Cancho, Pastor-Satorras, Chater, & Christiansen, 2013; Siew, Wulff,Beckage, & Kenett, 2018). A network is comprised fromnodes, that represent the basic unit of the system (e.g.,concepts in semantic memory) and links, or edges, thatsignify the relations between them (e.g. semantic similarity).While the application of network science methodologies hasbecome an extremely popular approach to study brainstructure and function, it has been used to study cognitivephenomena to a much lesser extent. This, despite classiccognitive theory in language and memory being highlyrelated to a network perspective (Collins & Loftus, 1975;Siew et al., 2018). Already, network science in cognitivescience has enabled the direct examination of the theory thathigh creative individuals have a more flexible semanticmemory structure, identified mechanisms of languagedevelopment through preferential attachment, shed novellight on statistical learning, shown how specific semanticmemory network parameters influence memory retrieval,and provided new insight on the structure of semanticnetwork of second language in bilinguals (Siew et al.,2018).The aim of the current symposia is to demonstrate thepotential and strength of applying network sciencemethodologies to study cognition. This will be achieved bybringing together leading researchers that apply suchmethods to study various aspects of cognition, includinglanguage, learning, aging, and creativity. The presentationswill describe state-of-the-art progress and perspectives thatare achieved in applying these methods to study cognition.Importantly, these talks aim at stimulating discussion of thefruitfulness of such an approach and how such an approachcan powerfully and quantitatively study the complexity ofcognitive phenomena. Finally, this symposium aims todemonstrate how network science in cognitive science canbe used to quantitatively bridge across different levels ofanalysis, spanning the computational, behavioral, neural,and social.

The associative theory of creativity suggests that creativeabilities rely on the organization of semantic associations inmemory. Recent research has demonstrated that semanticnetwork methods allow testing this hypothesis. The aim of thecurrent study was to investigate the properties of semanticnetworks at the individual level, in relation to creative abilities.Semantic judgement ratings were used to estimate individualsemantic networks, whose topological properties measured byseveral graph metrics were correlated with individual creativityscores. We found a correlation between the theoretical semanticdistance of our stimuli and the relatedness ratings given by theparticipants, demonstrating the validity of our approach.Importantly, we found a close relationship between creativeabilities assessed by an achievement questionnaire and divergentthinking tasks and individual semantic network metrics,replicating and extending previous similar findings.

Semantic networks have been used extensively in psychologyto describe how humans organize facts and knowledge inmemory. Numerous methods have been proposed to constructsemantic networks using data from memory retrieval tasks,such as the semantic fluency task (listing items in a category).However these methods typically generate group-levelnetworks, and sometimes require a very large amount ofparticipant data. We present a novel computational methodfor estimating an individual’s semantic network usingsemantic fluency data that requires very little data. Weestablish its efficacy by examining the semantic relatedness ofassociations estimated by the model.