Your gaze betrays your age

International audienceVisual attention networks are so pervasive in thehuman brain that eye movements carry a wealth of informationthat can be exploited for many purposes. In this paper, we presentevidence that information derived from observers’ gaze can beused to infer their age. This is the first study showing thatsimple features extracted from the ordered sequence of fixationsand saccades allow us to predict the age of an observer. Eyemovements of 101 participants split into 4 age groups (adults,6-10 year-old, 4-6 year-old and 2 year-old) were recorded whileexploring static images. The analysis of observers’ gaze providesevidence of age-related differences in viewing patterns. Therefore,we extract from the scanpaths several features, including fixationdurations and saccade amplitudes, and learn a direct mappingfrom those features to age using Gentle AdaBoost classifiers.Experimental results show that the proposed image-blind methodsucceeds in predicting the age of the observer up to 92% of thetime. The use of predicted salience does not further improve theclassification’s accuracy