Predictions of Age and Mood Based on Changes in Saccades Parameters

Abstract

We have measured saccades of 12 subjects in age below 30 (younger adults) and over 60 (older adults) during musical sessions with energetic and relaxing music. We wanted to check if saccade parameters are sensitive to emotions caused by music and if so, check if there ability to distinguish and classify those parameters to one of defined group of Age (young/old) or group of Mood (energetic/relaxed) induced by the music. We used combination of different types of filters and classifiers from WEKA [5] in search of possible correlations between saccade parameters and characteristics of respondents. Results showed statistical changes between age groups in the latency (23.6% of difference) and in the pupils size (16,6% of difference), both found significant (P < 0.0001). In case of Mood, results showed changes in the group of younger adults in the latency (P = 0.4532), the amplitude (P = 0.0001) and for the average velocity (P = 0.0048). Prediction of age group showed the accuracy of 91.4%, in case of Mood groups 97%. For both types of groups, best predictions were obtained by the Random Forest and the Multilayer Perceptron.