Detecting emotional reactions to videos of depression

Loading...
Thumbnail Image

Date

Authors

Zhu, Xuanying
Gedeon, Tom
Caldwell, Sabrina
Jones, Richard

Journal Title

Journal ISSN

Volume Title

Publisher

IEEE

Abstract

We can physically sense depression in others, and this recognition can be detected by using neural networks to analyse our physiological responses to observing individuals with depression. The behaviour of 16 individuals suffering from various levels depression were shown in short videos to 12 experiment participants (observers) whose physiological signals we recorded. Consciously, depression is not normally interesting, so does not provoke strong conscious recognition, and hence barely over chance, at 27%. However, at emotional levels, depression is interesting, so provokes physiological reactions we can measure, leading to neural network classification of 92%.

Description

Citation

Source

INES 2019 - IEEE 23rd International Conference on Intelligent Engineering Systems, Proceedings

Book Title

Entity type

Access Statement

License Rights

Restricted until

2099-12-31