Kanokwanwijit Wongsook¹, Payachana Chareunsouk^1,2, Sedthapong Chunamchai^1,3, Chaipat Chunharas^1,3,4; ¹Cognitive Clinical and Computational Neuroscience Center of Excellence, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand, ²Medical sciences program, Division of Neuroscience, Faculty of Medicine, Chulalongkorn university, Bangkok, Thailand, ³Division of Neurology, Department of Medicine, Faculty of Medicine, King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand, ⁴Chula Neuroscience Center, King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand

Previous studies suggest that functional relatedness, referring to items’ interactions based on their function, and semantic relatedness, referring to shared contextual associations between items, both improve item recall memory, but their impact on spatial precision and spatial biases remains unclear. We hypothesized that relatedness would improve spatial precision through grouping and simultaneously influence spatial biases: items might be stored separately and later pulled together, creating attractive or repulsive bias, or stored as a single unit, reducing bias. Here, we examined the effects and interaction of functional and semantic relatedness on spatial precision and spatial bias. Across 480 trials, real-world related items were presented under four conditions: functionally and semantically related, functionally related only, semantically related only, and unrelated. Participants were asked to drag items to their original locations after a 1400 ms delay. We quantified spatial accuracy using three Euclidean distance–based metrics: precision, absolute bias, and relative bias. Results from fifteen participants showed higher precision for functional than non-functional related items (F(1,14) = 71.84, p < .001). Semantic related items were remembered more accurately than non-semantic related items (F(1,14) = 7.89, p = .013). Semantic benefit was observed only for functional related items (F(1,14) = 33.93, p < .001) and was not observed for non-functional related items. Non-functional related items showed higher attractive bias than functional related items on absolute (F(1,14) = 61.24, p < .001) and relative positions (F(1,14) = 30.13, p < .001), indicating stronger attractive biases. Overall, functional and semantic relatedness enhanced spatial precision while non-functional related items show stronger attractive bias, suggesting initial separate encoding and later grouping. Together, these results highlight how high-level relationships structure spatial memory.

Acknowledgements: The authors wish to acknowledge the valuable contributions of the CCCN members for their support, data provision, and insightful discussions.