Overview | Research directions | Members | Publications

Overview

The Lab investigates how the human brain organizes knowledge, memories, and concepts, combining cognitive neuroscience, neuroimaging, eye tracking, and computational modeling to study the neural basis of thought.

Research directions

1. Memory, knowledge organization, and cognitive maps
A central focus of the lab is understanding how memories, concepts, and knowledge are structured in the brain. A guiding idea behind our work is that the neurocognitive machinery mammals evolved to represent physical space-centered on the hippocampal-entorhinal system - has been recycled to organize memory and knowledge more generally. In this view, cognitive maps provide a common representational format that allows the mind to “locate” experiences and ideas in structured spaces and to move through them flexibly.

We study how these map-like representations support mental time travel (reconstructing the past and simulating possible futures), goal-directed reasoning (traversing intermediate steps until a solution is found), and the ability to understand relations - between concepts, between events, and between ourselves and other people. Using behavioral paradigms, neuroimaging, and computational modeling, we examine how structured knowledge emerges, how it is updated with experience, and how it enables generalization and flexible behavior across domains (spatial, conceptual, semantic, and social).

2. Eye movements as a window onto the human mind
The lab also investigates eye movements as an external marker of internal cognitive processes. We study how gaze and attentional dynamics reflect memory retrieval, conceptual navigation, and relational thinking - even in the absence of visual input (“looking at nothing”). This line of research treats eye movements not merely as motor behavior, but as a powerful window into the structure and dynamics of thought.

3. Brain-computer interfaces and reconstruction of mental content
A third research direction explores brain-computer interfaces and the reconstruction of mental content from neural signals. Using deep learning and generative models, we work on reconstructing perceived images as well as internally generated experiences, including mental imagery and dream-like states, from neuroimaging data. This research bridges cognitive neuroscience, artificial intelligence, and computational neuroimaging.

Methods

Across these research lines, the lab employs a wide range of methodologies, including functional MRI (fMRI), magnetoencephalography (MEG), EEG, intracranial EEG, eye tracking, and deep learning–based computational approaches.
 

Members

Publications

For a complete list see Roberto Bottini personal page