New forms of visualizations are transforming how people interact with data. This dissertation explored how undergraduates learn with infographics. The following questions guided this research: (i) What do we know about the factors influencing the processing of data visualizations? (ii) How do task-level and learner-level characteristics impact the visual processing and comprehension of infographics? (iii) Can machine learning be used to reliably predict the visual processing and comprehension of infographics using task-level and learner-level characteristics? Systematic review of the literature has shown that data visualization comprehension includes perceptive and conceptual processes, which are influenced by learning task and its complexity, strategies used to convey data, individual learner differences in previous experiences, cognitive and attentional characteristics. A study was conducted with 51 undergraduates in an eye-tracking laboratory at a major southeastern university. The learning task included using infographics with verbal and visual data representations to find answers to questions of three levels of complexity. Learners' working memory, visual search and inhibitory control abilities were evaluated as measures of individual differences in cognition. The results suggest that learners become more engaged and produce slightly more accurate results when they learn using verbal infographics. Complex tasks that require learners to make inferences by connecting newly acquired knowledge with prior knowledge always produce the least accurate results. Regardless of the data representation format, learners' visuospatial working memory and goal-oriented visual search ability significantly influence comprehension. On the contrary, low verbal working memory and inhibitory control hinder the processing of verbal infographics. Further, a random forest algorithm predicted infographics comprehension with 88.04% accuracy with learner-level and task-level characteristics contributing to the predictive performance of the model. Learners' visual processing was explored using gaze-based saliency maps. Machine learning generated less than 50% accuracy using saliency map predictions. Yet, statistical tests revealed that both task-level and learner-level characteristics are significantly associated with saliency maps. This study implies that machine learning and the proposed saliency maps may contribute to the development of adaptive infographics based on learner-level and task-level characteristics. This study contributes insights to existing knowledge on data visualization comprehension and proposes new approaches to enhance learning with information visualizations. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com.bibliotheek.ehb.be/en-US/products/dissertations/individuals.shtml.]