What does brain response to neutral faces tell us about major depression? Evidence from machine learning and fMRI

Oliveira, Leticia, Ladouceur, Cecile D., Phillips, Mary Louise, Brammer, Michael and Mourao-Miranda, Janaina 2013. What does brain response to neutral faces tell us about major depression? Evidence from machine learning and fMRI. PLoS ONE 8 (4) , e60121. 10.1371/journal.pone.0060121

Preview

PDF - Published Version Available under License Creative Commons Attribution. Download (513kB) | Preview

Abstract

Introduction A considerable number of previous studies have shown abnormalities in the processing of emotional faces in major depression. Fewer studies, however, have focused specifically on abnormal processing of neutral faces despite evidence that depressed patients are slow and less accurate at recognizing neutral expressions in comparison with healthy controls. The current study aimed to investigate whether this misclassification described behaviourally for neutral faces also occurred when classifying patterns of brain activation to neutral faces for these patients. Methods Two independent depressed samples: (1) Nineteen medication-free patients with depression and 19 healthy volunteers and (2) Eighteen depressed individuals and 18 age and gender-ratio-matched healthy volunteers viewed emotional faces (sad/neutral; happy/neutral) during an fMRI experiment. We used a new pattern recognition framework: first, we trained the classifier to discriminate between two brain states (e.g. viewing happy faces vs. viewing neutral faces) using data only from healthy controls (HC). Second, we tested the classifier using patterns of brain activation of a patient and a healthy control for the same stimuli. Finally, we tested if the classifier’s predictions (predictive probabilities) for emotional and neutral face classification were different for healthy controls and depressed patients. Results Predictive probabilities to patterns of brain activation to neutral faces in both groups of patients were significantly lower in comparison to the healthy controls. This difference was specific to neutral faces. There were no significant differences in predictive probabilities to patterns of brain activation to sad faces (sample 1) and happy faces (samples 2) between depressed patients and healthy controls. Conclusions Our results suggest that the pattern of brain activation to neutral faces in depressed patients is not consistent with the pattern observed in healthy controls subject to the same stimuli. This difference in brain activation might underlie the behavioural misinterpretation of the neutral faces content by the depressed patients.

Item Type:	Article
Status:	Published
Schools:	Medicine MRC Centre for Neuropsychiatric Genetics and Genomics (CNGG)
Subjects:	R Medicine > R Medicine (General)
Additional Information:	Copyright: © 2013 Oliveira et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Publisher:	Public Library of Science
ISSN:	1932-6203
Date of First Compliant Deposit:	30 March 2016
Last Modified:	14 May 2023 06:41
URI:	https://orca.cardiff.ac.uk/id/eprint/75757

Citation Data

Cited 22 times in Scopus. View in Scopus. Powered By Scopus® Data

Actions (repository staff only)

Edit Item