Why now?

The format of VWM representations remains an open theoretical question. Sensory recruitment models propose that VWM relies on the same sensory circuits as perception while other accounts suggest its content is reformatted into abstract semantic codes. It remains unclear whether unattended items persist through activity-silent mechanisms or are offloaded into LTM-like reconstructive traces. Recent findings show that semantic prioritisation increases with memory load and retention interval, pointing toward a shift to LTM-like storage, but the underlying attentional and encoding mechanisms have not yet been tested directly. 

What for?

Results will clarify how attention transforms the representational format of working memory, reconciling working memory and long-term memory theories under a shared framework. Beyond cognitive theory, findings may inform strategies to improve memory retention under high cognitive load, with potential translational relevance for education, attentional difficulties and eyewitness memory research.

What's novel?

This project is the first to test whether retrocueing can flexibly reshape the balance between semantic and perceptual information in VWM and whether the temporal structure of encoding (simultaneous versus sequential) modulates this trade-off. It integrates behavioral measures, EEG decoding and drift diffusion modeling to trace not only whether attention reshapes memory format but when and how it does so at the neural and computational level.

Where?

The project is led by Dr. Juan Linde-Domingo and Dr. Fabiano Botta at CIMCYC María de Maeztu Unit of Excellence, University of Granada (Memory and Language Research Group, HUM740, and Cognitive Neuroscience Group, HUM379), in collaboration with Casper Kerrén (Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig) and Carlos González-García from CIMCYC María de Maeztu Unit of Excellence, University of Granada (Human Brain & Cognition, HUM957).