Computational models of human confidence
Confidence is the “feeling of knowing” that comes with every decision we make. I’m interested in understanding how the brain computes this feeling, leading to a confidence report. Because these reports are basically an statement about probability, I’m interested in the contribution of different probabilistic quantities to confidence.
What are the computations underlying human confidence? How variable are those computations among different individuals? What are the neural circuits supporting those computations?
Wisdom (and madness) of the crowds
When many people make individual judgements, the average of those opinions tend to be very accurate, sometimes even more accurate than the best individual. This effect is known as the “wisdom of crowds”. I’m interested in understanding the conditions under which human groups can be wise.
What is the deliberation procedure that leads to optimal integration of crowd wisdom? Do humans follow this procedure? Why do sometimes collective decisions catastrophically fail leading to “crowd madness”?
Neural correlates of consciousness
When we see a picture, the light reflecting on that image stimulates our retina and triggers a sequence of neural processes. I’m interested in understanding which of these processes lead to conscious perception.
What are the spatio-temporal dynamics of visual awareness? Should we define a single conscious process in the brain? What is the timing of visual awareness? How can we dissociate consciousness from visual attention?
Cognition across eye movements
Reality is an illusion. A prime example is given by eye movements: although we move our eyes three times per second, we perceive the world as an stable environment. I’m interested in understanding how the brain constructs reality within and across eye fixations.
How well do we integrate information across and within eye movements? Are eye movements informative of our choices and preferences?
Unsupervised methods for neural data
We live in the era of Big Data and neural recordings aren’t the exception as everyday they are becoming larger and more complex. To avoid an overflow of data and to prevent subjectivity in data analysis, we need to come up with unsupervised algorithms to automatically extract information from neural signals.
Some years ago, I worked in the development of an online, unsupervised, on-chip spike sorting system with potential applications to brain-machine interfaces. Currently, I’m collaborating in the development of an unsupervised method to extract information from event-related potentials.