RESULTS & IMPACT

The project is expected to advance our understanding of working memory dynamics through complementary behavioral, neural and computational evidence.

Expected Findings

Behaviorally, the project expects to replicate the semantic prioritization effect, with retrocues attenuating or reversing this bias by reactivating the cued item in a detail-rich state. Sequential encoding is expected to enhance semantic dominance for early items relative to simultaneous encoding. At the neural level, uncued trials should show early reactivation of semantic-category signals followed by perceptual-feature patterns, while retrocued trials should show earlier and stronger reinstatement of perceptual activity alongside greater alpha suppression and contralateral delay activity. Drift diffusion modeling should reveal higher drift rates and shorter non-decision times for semantic judgments in uncued conditions, with these differences reduced under retrocue.

Theoretical Impact

Confirmed hypotheses would provide strong evidence that the format of information in working memory is dynamic rather than static and can be purposefully shifted by internal attention. This would support a model with at least two representational states (an active attention-focused state retaining rich perceptual detail and an activity-silent or latent state holding items in a more conceptual form) reconciling working memory and long-term memory theories under a shared attentional framework.

Translational Relevance

Understanding how attention can rescue or improve memory detail has potential applications in educational and work contexts involving multitasking and heavy memory load, as well as relevance for individuals with attentional or memory difficulties and for eyewitness memory research.

Dissemination and Knowledge Transfer

Results will be published in high-impact peer-reviewed journals (e.g. Nature Communications, Journal of Cognitive Neuroscience), with preprints shared on open-access repositories such as bioRxiv, and presented at international conferences including the Cognitive Neuroscience Society meeting and the Society for Neuroscience. Findings will also be shared through an internal seminar at CIMCYC and María de Maeztu Unit outreach activities. All data and analysis code will be made openly available following anonymization, in line with open science practices.