The patterns of life exhibited by large populations have been described andmodeled both as a basic science exercise and for a range of applied goals suchas reducing automotive congestion, improving disaster response, and evenpredicting the location of individuals. However, these studies previously hadlimited access to conversation content, rendering changes in expression as afunction of movement invisible. In addition, they typically use thecommunication between a mobile phone and its nearest antenna tower to inferposition, limiting the spatial resolution of the data to the geographicalregion serviced by each cellphone tower. We use a collection of 37 milliongeolocated tweets to characterize the movement patterns of 180,000 individuals,taking advantage of several orders of magnitude of increased spatial accuracyrelative to previous work. Employing the recently developed sentiment analysisinstrument known as the 'hedonometer', we characterize changes in word usage asa function of movement, and find that expressed happiness increaseslogarithmically with distance from an individual's average location.
展开▼
