Evidence of novelty and specialization behavior in participatory science reporting

Participatory science (or ‘citizen science') records are becoming increasingly useful for wildlife monitoring due to their volume and spatiotemporal coverage. Statistical analysis of these data can be challenging due to the many sources of sampling heterogeneity that need to be accounted for. Many previous studies characterize sampling variability across entire participatory science datasets, such as spatial heterogeneity in sampling effort or species preferences. User‐level heterogeneity in sampling behavior is less well studied, but it may be just as important as dataset‐level variability in contributing to downstream analyses. Here, we investigate user‐level novelty and specialization behavior. Novelty‐seeking behavior occurs when an individual observer preferentially reports species that they have not seen before, while specialization, or novelty‐avoidant behavior, occurs when an observer preferentially reports species they have previously observed (i.e. they specialize in particular species or taxa). We provide the first test of user‐level sampling variability in participatory science data by analyzing the sampling histories of more than 2000 observers on the popular participatory science platform iNaturalist in Pennsylvania, USA. We find evidence of specialization or novelty behavior in the overall sampling behavior of 70% of the observers considered. We identified specialization behavior in more than six times as many observers than novelty behavior, indicating that a large proportion of observers reported species they had reported previously at a higher rate than expected. Within taxonomic groups, 61% of observers deviated from neutral sampling behavior. Novelty and specialization behavior were both common within taxa as well. These findings suggest that iNaturalist observers often specialize in favorite taxa or species, while within taxa some users simultaneously seek out previously unobserved species.