Explore the vast range of titles available.

The eruption of Mount St. Helens in 1980 resulted in a cataclysmic restructuring of its surrounding landscapes. The Pumice Plain is one of these landscapes, where tree species such as Sitka willow (Salix sitchensis) and their dependent communities have been established along newly-formed streams. Thus, the study of these dependent communities provides a unique and rare opportunity to investigate factors influencing metacommunity assembly during true primary succession. We analyzed the influence of landscape connectivity on metacommunity assembly through a novel application of circuit theory, alongside the effects of other factors such as stream locations, willow leaf chemistry, and leaf area. We found that landscape connectivity structures community composition on willows across the Pumice Plain, where the least connected willows favored active flyers such as the western tent caterpillar (Malacosoma fragilis) or the Pacific willow leaf beetle (Pyrrhalta decora carbo). We also found that multiple levels of spatial habitat structure linked via landscape connectivity can predict the presence of organisms lacking high rates of dispersal, such as the invasive stem-boring poplar weevil (Cryptorhynchus lapathi). This is critical for management as we show that the maintenance of a heterogeneous mixture of landscape connectivity and resource locations can facilitate metacommunity dynamics to promote ecosystem function and mitigate the influences of invasive species.

Learning to classify information into concepts and categories is an essential component to educational success. However, learning to correctly classify examples into a given category can be challenging, particularly when learning STEAM (Science, Technology, Engineering, Art, & Math) categories. Two factors that may impact STEAM category learning are category organization and study order. For example, when learning geological categories, the information may be organized into basic categories such as igneous and metamorphic rocks; or the information may be organized into subcategories such as granite and peridotite (which are types of igneous rocks). When studying either subcategories or basic categories, students may decide to study in blocks, such that a student would study several examples from the same category in a row; or they could study the categories in an interleaved order, such that the categories are mixed together. The discriminative contrast hypothesis suggests that the study order most beneficial to learning may be contingent to category organization. Thus, the goal of the present dissertation was to systematically evaluate the impact of study order and category organization from the lens of the discriminative contrast hypothesis. Across a Pilot experiment and two high-powered experiments, study order did not significantly influence classification performance when learning most geological categories. However, interleaving categories during learning was beneficial to novel classification performance for students who learned basic categories with many exemplars. Whereas study order did not impact classification performance for most groups, category structure did impact classification performance. Specifically, students who learned to classify geological subcategories performed significantly better on the classification tests than did students who learned basic geological categories. The present results are inconsistent with the discriminative contrast hypothesis and indicate a nuanced context in which study order influences STEAM category learning.

The present research investigated what type of feedback was most beneficial to category learning. The discriminative contrast hypothesis suggests that comparisons between categories enhances learning (e.g., Kornell & Bjork, 2008). As such, feedback that gave learners additional information and encouraged between-category comparisons with additional information (i.e., contrast feedback) was predicted to improve category learning more than feedback that only gave additional information (i.e., feature feedback), feedback that gave the correct answer (i.e., corrective feedback) or no feedback. To explore this, participants learned categories of organic chemistry compounds and received different types of feedback. The results revealed that providing feedback during category learning was beneficial for learning, and that providing additional information during feedback was beneficial to later performance above and beyond corrective feedback. The contrast feedback group did not perform better than did the feature feedback group; more research should investigate the degree to which contrast feedback improves later performance.

Researchers can investigate many aspects of animal ecology through noninvasive photo–identification. Photo–identification is becoming more efficient as matching individuals between photos is increasingly automated. However, the convolutional neural network models that have facilitated this change need many training images to generalize well. As a result, they have often been developed for individual species that meet this threshold. These single‐species methods might underperform, as they ignore potential similarities in identifying characteristics and the photo–identification process among species. In this paper, we introduce a multi‐species photo–identification model based on a state‐of‐the‐art method in human facial recognition, the ArcFace classification head. Our model uses two such heads to jointly classify species and identities, allowing species to share information and parameters within the network. As a demonstration, we trained this model with 50,796 images from 39 catalogues of 24 cetacean species, evaluating its predictive performance on 21,192 test images from the same catalogues. We further evaluated its predictive performance with two external catalogues entirely composed of identities that the model did not see during training. The model achieved a mean average precision (MAP) of 0.869 on the test set. Of these, 10 catalogues representing seven species achieved a MAP score over 0.95. For some species, there was notable variation in performance among catalogues, largely explained by variation in photo quality. Finally, the model appeared to generalize well, with the two external catalogues scoring similarly to their species' counterparts in the larger test set. From our cetacean application, we provide a list of recommendations for potential users of this model, focusing on those with cetacean photo–identification catalogues. For example, users with high quality images of animals identified by dorsal nicks and notches should expect near optimal performance. Users can expect decreasing performance for catalogues with higher proportions of indistinct individuals or poor quality photos. Finally, we note that this model is currently freely available as code in a GitHub repository and as a graphical user interface, with additional functionality for collaborative data management, via Happywhale.com.