Biology subjects

Rajananda, S.

Publications and source records attributed to Rajananda, S..

5 recordsLinked to original sources

bioRxiv2021

Performance-optimized neural networks as an explanatory framework for decision confidence

Previous work has sought to understand decision confidence as a prediction of the probability that a decision will be correct, leading to debate over whether these predictions are optimal, and whether they rely on the same decision variable as decisions themselves. This work has generally relied on idealized, low-dimensional models, necessitating strong assumptions about the representations over which confidence is computed. To address this, we used deep neural networks to develop a model of decision confidence that operates directly over high-dimensional, naturalistic stimuli. The model accounts for a number of puzzling dissociations between decisions and confidence, reveals a rational explanation of these dissociations in terms of optimization for the statistics of sensory inputs, and makes the surprising prediction that, despite these dissociations, decisions and confidence depend on a common decision variable.

neuroscience↗

bioRxiv2020

New measures of agency from an adaptive sensorimotor task

Self-agency, the sense that one is the author or owner of ones behaviors, is impaired in multiple psychological and neurological disorders, including functional movement disorders, Parkinsons Disease, alien hand syndrome, schizophrenia, and dystonia. Existing assessments of self-agency, many of which focus on agency of movement, can be prohibitively time-consuming and often yield ambiguous results. Here, we introduce a short online motion tracking task that quantifies movement agency through both first-order perceptual and second-order metacognitive judgments. The task assesses the degree to which a participant can distinguish between a motion stimulus whose trajectory is influenced by the participants cursor movements and a motion stimulus whose trajectory is random. We demonstrate the tasks reliability in healthy participants and discuss how its efficiency, reliability, and ease of online implementation make it a promising new tool for both diagnosing and understanding disorders of agency.

neuroscience↗

bioRxiv2018

Normal observers show no evidence for blindsight in facial emotion perception

It is commonly assumed that normal human observers can exhibit blindsight-like behavior: the ability to discriminate or identify a stimulus without being aware of it. However, we recently used a bias-free task to show that what looks like blindsight may in fact be an artifact of typical experimental paradigms susceptibility to response bias. While those findings challenge many previous reports of blindsight in normal observers, they do not rule out the possibility that different stimuli or techniques could reveal such perception without awareness. One particularly intriguing candidate for this unconscious perception is emotion processing, as processing of emotional stimuli (e.g. fearful/happy faces) has been reported to potentially bypass conscious visual circuits. Here we used the bias-free blindsight paradigm to investigate whether emotion processing might reveal featural blindsight, i.e. the ability to identify a faces emotion without having introspective access to the task-relevant features of the face that led to the discrimination decision. However, despite the purported ability of emotional stimuli to bypass conscious visual processing, we saw no evidence for such emotion processing featural blindsight: like in our previous study, as soon as participants could identify a faces emotion they reported introspective access to the task-relevant features, matching predictions of a Bayesian ideal observer. The present results challenge dominant theory, adding to the growing body of evidence that perceptual discrimination ability in the complete absence of introspective access may not be possible for neurologically intact observers.

neuroscience↗

bioRxiv2017

Subjective inflation of color saturation in the periphery under temporal overload

A long-standing puzzle in perception concerns the subjective impression of vivid color experience in the periphery. While peripheral color processing is not entirely insensitive, the subjective vividness seems unsupported by the relative paucity of representation in the periphery. Inspired by the flashed face distortion effect, in which peripheral faces are perceived as somewhat exaggerated or distorted when they are preceded by flashes of other faces in the same location, we attempted to create an equivalent illusion in the domain of color. The hypothesis is that under temporal overload, patches of colorful dots may be perceived as more exaggerated in terms of saturation. We confirmed this hypothesis with the observation of a significant effect of modest magnitude, which was replicated in a second experiment. These results suggest that subjective inflation of perceived color saturation does occur in the periphery, when the perceptual system is sufficiently occupied temporally and spatially. We discuss the relationship between the observed effects with previous findings of liberal detection biases in the unattended periphery.

neuroscience↗

bioRxiv2017

A random dot kinematogram for online visual psychophysics

Web-{-}based experiments using visual stimuli have become increasingly common in recent years, but many frequently-used stimuli in vision research have yet to be developed for online platforms. Here, we introduce the first open access random-dot kinematogram (RDK) for use in web browsers. This fully customizable RDK offers options to implement several different types of noise (random position, random walk, random direction) and parameters to control aperture shape, coherence level, the number of dots, and other features. We include links to commented JavaScript code for easy implementation in web-based experiments, as well as an example of how this stimulus can be integrated as a plugin with a JavaScript library for online studies (jsPsych).

neuroscience↗