Hacking into the Memory Palace. . A study on the effects of spatial clustering and active learning on item memorization

Abstract

Plenty of research has been conducted on the role of space in organizing memories. Recent studies indicate a connection between spatial clustering and free-recall memory performance. However, the long-term stability of this effect remains poorly understood. While voluntary learning is known to benefit memory, little research has explored whether actively creating spatial structure and information clusters offers additional advantages compared to passive scenarios of previous studies. To address these questions, we created an original, fully immersive virtual reality setup. In the first experiment, we investigated the role of semantic organization in memory and its relationship with spatial clustering. Participants were presented with items categorized under clear semantic groups and were tasked to organize and memorize them spatially. The semantic organization of memory outperformed other methods (e.g., temporal or spatial organization) and remained evident even after two weeks. In the second experiment, we eliminated semantic clustering to examine the effects of volition and active creation of spatial structure and its relation to spatial and temporal clustering. Actively creating and organizing spatial structure significantly reduced forgetting rates for the memorized items. While spatial and temporal clustering were observed immediately after the tests, they were not present after the two-week period. These findings highlight the crucial role of volition in encoding and retaining information, suggesting that actively creating memory structure (spatial, semantic, or temporal) is fundamental for long-term declarative memory. Additionally, the study revealed the persistence of semantic organization over time, whereas temporal and spatial organization vanished after two weeks, indicating different roles in short-term and long-term memory organization. This study offers valuable insights for the development of new learning techniques, memory-friendly environments, and spatial computing interface