The Nature of Visual Short-Term Memory Training Transfer: Evidence from A Systematical Transfer Test

Whether visual short-term memory (VSTM) training can transfer to untrained tasks has been under intensive debate, one of the critical reasons is the nature of VSTM training improvement is still unclear. In the current study, we trained participants with a delayed estimation task for locations and systemically examined its transfer effects with three types of untrained tasks: tasks with the same paradigm but changed stimuli (i.e., delayed estimation tasks for colors and letters), tasks with the same stimuli but changed paradigms (i.e., complex span and n-back tasks for locations), and tasks with combined changes in paradigms and stimuli (e.g., the complex span task for colors). First, we observed that the location delayed estimation training transferred to both the color delayed estimation task and the location complex span task but not the color complex span task or others. Furthermore, we adopted model fittings to estimate memory quantity and quality in delayed estimation tasks for locations (trained) and colors (untrained). Our results revealed that, in the trained task, both memory quantity and quality increases contributed to the recall improvement, which was predicted by lower baseline performance. In the untrained task, however, we found that mainly increased quality and an optimized quality-quantity trade-off strategy lead to the transferred recall improvement, which was predicted by higher baseline performance. Together, our findings suggested that training improved the stimulus-specific STM resource and optimized the paradigm-specific trade-off strategy. These results extended our understanding of the nature of VSTM training and shed light on future transfer studies.