Towards a Naturalistic Study of Behavior

We developed the Sensory Island Task (SIT) to allow for the study of neural processes and behaviors underlying perception in ethologically relevant scenes. 

In SIT, animals explore an open-field arena to find a sensory target relying solely on changes in the presented stimulus, which is controlled by closed-loop position tracking in real-time.

It is a flexible and easily implementable behavioral paradigm that uniquely incorporates self-motion and natural exploratory behavior, which are essential for naturalistic sensory processing. SIT is readily applicable across species and sensory modalities and extendable to use for neurophysiological investigations.

Social Cognition

The study of behavior (including learning) has traditionally focused on an isolated individual. However, many species (including ourselves) carry out their lives in a social context.

I am using the Sensory Island Task to investigate the complex social interactions of gerbils during an auditory task performance.

Neural Correlates of Active Sensing in Humans

The 'Sensory Island Task for humans' (SITh) allows for the study of auditory perception and decision making in unrestricted subjects.

SITh is inspired by the children's game 'Hot and Cold', in which one finds the location of an object by external auditory cues, and it has succesfully reinforced the intuition that incorporating more naturalistic environments into the lab is a worthwhile effort. 

By pairing it with freely moving, wireless EEG recordings, we study the neural processing of active sensing while subjects perform auditory tasks and manipulating the visual input available to them.

Crowd Cognition and Forecasting

Predicting the future can bring enormous advantages, if the predictions are accurate!

Here, we go beyond the 'Wisdom of Crowds' (i.e. the aggregation of many independent estimates which can outperform the most accurate individual judgments). We apply 'Wisdom of Structured Crowds', a causal intervention seeking consensus within each sub-group and aggregating consensus opinions across groups to remove biases (e.g. herding and polarization).

Our results show that this method improves accuracy of geopolitical forecasting both in the laboratory setting as well as in the 'real world'.

Evolution of Cortical Maps

Since the Nobel prize to Hubel and Wiesel for the discovery of orientation preference maps (OPMs) in visual cortex in carnivores and primates, no reports about rodents' showed similar structure.

Theoreticians hypothesized that a main driver for OPM evolution may have been the size of the brain, due to optimization of wiring between neurons with similar orientation preference.

During my PhD, I studied the orientation preference in the visual cortex of the Agouti (a large Amazonian rodent, similar in size to a cat). We showed that, contrary to theoretical models, the largest ever recorded rodent does not show OPMs, but conserves a similar organization to smaller rodents.

Protein Dynamics for Intelligent Drug Design

Discovering new treatments for disease is a never-ending struggle against new mutations (e.g. cancer) and the development of antibiotic resistance by pathogens (e.g. tuberculosis). Therefore, optimizing the efficiency of new drug development is advantegeous.

Using bioinformatics, computer modeling and in-silico simulations of protein dynamics, we described the active site structure and dynamics details of key proteins that would allow the intelligent design of drugs against cancer and tuberculosis.