Last updated: 10 June 2026
How do our senses of vision, smell, and taste interact to shape perception and emotion? This talk explores the dynamic interplay among these sensory modalities, drawing on recent behavioral and neuroimaging studies. I will present evidence that visual and olfactory cues can systematically modulate taste perception, and vice versa, revealing consistent patterns of sensory-emotional integration. Using functional MRI, our research identifies brain regions—such as the insula, orbitofrontal cortex, and primary visual areas—that contribute to these cross-modal effects. Together, the findings shed light on the neural pathways supporting multisensory flavor experience and offer implications for affective neuroscience, sensory design, and health-related applications.
Reading acquisition in deaf individuals differs fundamentally from typical hearing pathways due to limited access to auditory and phonological information. My research investigates how deaf readers develop literacy through alternative visual and cognitive strategies, focusing on the mechanisms that both support and constrain reading performance. Across a series of experimental studies, I have examined visual crowding, serial position effects, contextual processing, and parafoveal attention during letter and word recognition. Findings demonstrated that deaf readers rely heavily on visual and orthographic cues in the absence of phonological decoding. The first study revealed heightened visual crowding effects in deaf participants, particularly in peripheral vision, suggesting that enhanced visual attention does not necessarily translate into efficient letter recognition under crowded conditions. The second study identified a strong first-letter advantage, indicating increased reliance on early orthographic information during word recognition. The third study showed that deaf readers benefit from lexical and contextual support in a manner comparable to hearing readers, demonstrating robust orthographic knowledge and effective top-down processing. Finally, the fourth study revealed altered parafoveal processing patterns alongside persistent inefficiencies in multi-letter processing. Together, these findings highlight both the adaptability and limitations of visually based reading strategies in deaf individuals. The research challenges phonocentric models of reading by demonstrating that literacy can develop through alternative cognitive pathways supported by visual and linguistic experience. My work contributes to a broader understanding of reading development and offers important implications for inclusive literacy education and intervention design for deaf readers.
The goal of my research is to understand the computations underlying natural behavior. While most laboratory studies examine behavior in simplified and constrained settings, natural behavior arises from the interaction of perception, learning, and planning in dynamic environments. I believe that natural behavior is best understood not as fundamentally different from laboratory tasks, but as a composition of simpler computational problems that can be studied tractably in the lab. In this talk, I present Resource-rational Context-dependent POMDPs (RCP) as a unified normative framework for understanding perception, learning, and planning under uncertainty and biological resource constraints. I illustrate this framework using three examples from my prior work on hierarchical motion perception, multi-context motor learning, and active sensing, each highlighting a different component of the framework.
High trait impulsive sensation seeking (ISS) is prevalent and associated with risky decision-making and adverse life outcomes. To better understand the neurophysiological mechanisms underlying ISS, we examined brain activity during uncertain reward expectancy (RE)—a context relevant to ISS—using stereo-electroencephalography (SEEG) in 14 epilepsy patients. Building on prior fMRI findings linking ISS to heightened activity in reward-related regions, we focused on RE-related neural dynamics, particularly in the prefrontal cortex. We observed increased beta power at stimulus onset during win trials compared to neutral trials, followed by a subsequent decrease. This beta activity pattern during uncertain RE was associated with greater risky decision-making. Importantly, this relationship was mediated by elevated levels of impulsivity, fun seeking, and experiential seeking (ES). These findings identify objective neural markers of high ISS and suggest potential targets for interventions aimed at reducing risk-taking behavior in young adults with elevated levels of this maladaptive personality trait.
Cortical blindness can occur after a cerebrovascular accident, and is usually characterised by a loss of vision in circumscribed regions (if not the entirety) of the visual field. Visual experience is mediated by brain dynamics, which are, in turn, available to manipulation by the use of stimulus dynamics. Can stimulus dynamics influence the outcome of stroke? Reporting 3 sets of studies, in this talk I will show that the answer to this question is ‘yes’, and, in fact, using specific stimulus-presentation frequencies, we have been able to restore vision to fairly large regions of the visual-blind field - at least for short periods of time. An interesting ‘inverse corollary’ with restoration is inhibition. There appear to be stimulus frequencies that are suboptimal for mediating visual experience. And these might be useful for approaching other diseases, particularly glaucoma. In concluding, I will explain how the use of stimulus dynamics might help unmask glaucoma at early stages of the disease.
Numerosity perception involves distinct mechanisms for subitizing (rapid, accurate enumeration of small quantities) and estimation (slower, approximate enumeration of larger quantities) (Kaufman et al, 1949). Computational models have shown that networks with varied inhibition strengths can replicate human-like number perception across different ranges of numerosities, aligning with behavioral findings that distinguish between subitizing and estimation ranges (Stoianov and Zorzi, 2012; Sengupta et al, 2014; Verma and Sengupta, 2023). In this context, investigating *inhibitory *neural mechanisms in enumeration is essential for understanding how the brain processes numerosities and performs numerical tasks. In our previous works (Sengupta et al., 2014, Verma and Sengupta, 2023) we focused on a network of neurons characterized by self-excitatory and mutual *inhibitory *properties, underscoring the importance of inhibition in regulating neural activity related to numerosity perception. In the current talk we explore the predictions from such a model at varying levels of inhibition in a series of behavioral experiments involving enumeration, perceptual averaging, and go/no-go tasks. This allows us to develop future studies to find neural correlates of such an inhibitory mechanism through neuroimaging studies.
The human brain is inherently complex with within-subject variability across hours, days, and months. A recent shift in neuroimaging studies is towards collecting a large amount of data per subject across multiple sessions to find out strong, robust patterns of brain activations and connectivity across task and rest runs. The talk would highlight some of my previous and ongoing projects towards this precision neuroimaging approach to study cerebro-cerebellar brain circuits and distributed mechanisms of multisensory attention and working memory, followed by recent methods for predicting brain function and cognitive health using connectivity data in healthy and clinical populations.
Human learning is a complex process and, in this Masterclass, Dr. Anveshna will describe through a few concrete examples how understanding the learning process can enable educators and learners to efficiently and effectively integrate technological resources in pedagogy.[watch live]
We spend a substantial amount of our waking time mind-wandering. Despite popular belief, I do not think that is a bad thing necessarily. It can be very useful for planning and creativity. Mind-wandering only becomes harmful when it is so "sticky" that it is difficult to disengage from, and therefore interferes with other important tasks of life. For this reason, I am very interested in tracking such sticky thinking in the lab. I will demonstrate different methods to assess sticky thinking, for example based on computerized tasks or based on EEG. I will show how rumination may be a particular example of sticky thinking, and finally, I will present my ideas on how different contemplative practices may help to reduce the stickiness of our thinking.
Prof. Sinha discussed his project Prakash which is helping children in India recover the ability to see. His work involved both ways to restoring sightedness and how individuals see the world upon recovery, for instance can people recognize shapes from sight alone immediately after being able to see. His spoke about how this work allowed him to solve the Molyneux Problem.
In this talk, Prof. Gold detailed the work his lab has been doing on perceptual decision making. He discussed with us his efforts over the years in trying to model how humans collect perceptual evidence in favour of a decision. He spoke about how he has been trying to model this form of decision making in a simpler and adaptive way, where models do not over-fit data and allow for variance seen in human performance.
Prof. Melcher discussed in his talk results from behavioural and neuroimaging studies that showed temporal constraints on perception, attention and memory. He presented us with results showing the temporal regularities over which these cognitive processes operate. Ultimately to explain how visual systems may construct percepts over time.
In his talk, Prof. Vasisth discussed the basic composition of the frequentist linear mixed modeling framework (including generalized linear models). He covered the following topics: shrinkage, model comparison, model evaluation using simulated data, and power analyses using the linear mixed model.
An expert on time perception, Prof. Wittmann gave a talk on relating concepts of self, body and perceived time. His work explores states of altered consciousness (meditation, experience under psychotropic substances etc). In his talk, he discussed how sense of self and time are altered during such experiences.