A computational account of the mechanisms underlying face perception biases in depression.

Abstract

Here, we take a computational approach to understand the mechanisms underlying face perception biases in depression. Thirty participants diagnosed with Major Depressive Disorder and thirty healthy control participants took part in three studies involving recognition of identity and emotion in faces. We used signal detection theory to determine whether any perceptual biases exist in depression aside from decisional biases. We found lower sensitivity to happiness in general, and lower sensitivity to both happiness and sadness with ambiguous stimuli. Our use of highly-controlled face stimuli ensures that such asymmetry is truly perceptual in nature, rather than the result of studying expressions with inherently di↵erent discriminability. We found no systematic e↵ect of depression on the perceptual interactions between face expression and identity, suggesting that depression is not associated with di culty attending to one of these dimensions while filtering out the other. We show through simulation that the overall pattern of results, as well as other biases found in the literature, can be explained by a neurocomputational model in which neural populations encoding positive expressions are selectively suppressed.

Information processing biases: The effects of negative emotional symptoms on sampling pleasant and unpleasant information.

“…Indeed, authors have noted the need for a better understanding of the development and nature of processing biases (Koster & Bernstein, 2015;Wells, 2019). Similar to others (e.g., Soto et al, 2021, also see Chen et al, 2015, for a review), we argue here that a prerequisite to understanding processing biases in emotional disorders is a coherent theoretical framework of information processing that can be computationally modeled and parameterized. Such a theoretical framework of information processing would enable more detailed descriptions of the interactions between negative emotional symptoms and sampling pleasant and unpleasant information, as well as the potential of characterizing emotional disorders as interpretable shifts in the model's parameters.…”

Face shape and motion are perceptually separable: Support for a revised model of face processing

“…The learning period was defined as data points before the point where the slope dropped below 0.001, which were discarded from the data of each participant. This method has been implemented and proved useful in multiple previous studies (e.g., Soto et al, 2021).…”

“…A more common approach in the computational modeling literature is to perform simulations to determine at what sample sizes it is possible to recover the true model parameters that have generated a data set. Such simulations (Soto et al, 2021) suggest that a sample size of 20-30 participants per experiment is adequate for accurate parameter estimation. Therefore, 30 complete datasets were collected for each group.…”

Multidimensional signal detection modeling reveals Gestalt-like perceptual integration of face emotion and identity.

“…After exclusion of uninformative data sets (see below), we obtained between 20 and 30 participants in each experiment. Simulation work has shown that these sample sizes provide high parameter recoverability for the model used in our analyses (Soto et al, 2021). Participation was voluntary and compensated with course credit.…”