Rakshitaasai Karthinarayanan, SIAS Cohort of 2024-28, has co-authored a new preprint titled Arousal as competition, valence as drive: a two-control account of affective modulation in the magnitude system with Dr Rakesh Sengupta, Assistant Professor, Psychology, SIAS. The research, from the Computational Cognition Lab, bridges cognitive psychology and computational modeling. The paper explores how our emotional states, specifically our levels of arousal and valence, affect our brain’s ability to estimate magnitudes, such as the quantity of items in a set.
The work was posted to Research Square on 9 September 2026, and is currently under review in the journal Biological Cybernetics.
Research Highlights
The Missing Link: A Theory of Magnitude (ATOM) suggests that time, space, and number are estimated over a shared substrate in the brain, but it currently offers no account of how human affect shapes those estimates.
The Innovation: The authors tackle this by mapping emotions onto an on-centre off-surround (OCOS) recurrent neural network. Specifically, they map “arousal” onto the network’s lateral inhibition (which acts as competition) and “valence” onto the network’s gain (which acts as drive).
The Key Prediction: Unlike previous models that reduce affect to a single dimension, this two-control account predicts that heightened arousal actually contracts the range over which a number can be estimated, rather than simply producing a uniform bias.
