Understanding how geologists conduct fieldwork through analysis of problem solving has significant potential impact on field instruction methods. Recent progress has been made in this area but the problem solving behaviors displayed by geologists during fieldwork and the associated underlying cognition remains poorly understood. We present research showing how geology students initiate and develop geologic models as part of the problem solving process. We qualitatively analyzed field notes and interview data from 36 undergraduate geoscientists engaged in field exams while enrolled in a six-week advanced field camp. Eight cognitive frameworks grouped in two broad categories emerged from the data that show how students develop geologic models. Students employ both single and multiple model approaches with varying degrees of success and frequency. The success of any given approach is dependent on the level of students' geologic situational awareness. The development of multiple geologic models leads to a higher rate of success in general, because of the inherent flexibility to accommodate newly collected data. Instructors should continue to teach a multiple model approach until students have the proper geologic skills to ensure a high level of situational awareness and exhibit expert characteristics in the field. In addition, we collected GPS navigation data from students during these field exams in order to understand the relationship between navigation, cognition, and performance. From the analysis of this data we found that over half of all stops are 1-4 minutes long, while very few of students' stops are longer than 9 minutes as the frequency of stops decreases as the duration increases. Regardless of performance or framework, there is an increase in shorter stops and decrease in longer stops from exam one to three, indicating that students changed the way they investigated as the field course progressed. Temporal signatures categorized by performance only show slight differences, but do indicate that there is an increase in very short and longer stops with declining performance. In contrast, higher performance is linked to an increase in short and medium length stops, suggesting that stops 4-14 minutes long are a "sweet spot" for investigation. We speculate that a high percentage of very short stops involve basic field tasks such as locating or data collection and that the decreasing frequency of long stops indicates that there is a relationship between the length of the stop and the complexity of the activities performed during that stop. Students increased experience leads to more efficient stops as they become more competent with field tasks and more versed in the regional geology. The GPS data we collected from these students while they took these notes allowing us to connect the duration of a stop to the types of notes produced during that stop. Note taking species occurred in various frequencies with the most common type being those that focused on lithologic, or lithologic & structural data collection. Stops that produced geologic models, specifically structural models, were much less frequent. The more frequent data collection stops are very short in length (typically 1-4 minutes), while the more complex stops tend to be longer in duration as the note taking gets more complex. Poor performing students had a high proportion of stops where they only collect lithologic data or stops where they don't generate any hypotheses. In contrast, successful students have more structural data and hypothesis generation in their notes. From this analysis we conclude that too much effort spent on stops with only basic data collection leaves less time for the cognitive effort required for model development, eventually leading to poor exam performance. Specifically, a higher frequency of lithologic data stops and lack of structural data leads to the development of incomplete geologic models or lack of comprehensive models altogether. Field instructors often educate their students on good note taking practices and critique the content of their students' field notebooks, and these findings can inform the content of that critique. Students' utilize many different problem solving approaches in the field, but we suggest that instructors continue to advocate a multiple model approach until students are capable of using a single model approach. In this approach students should decrease the number of geologic models they are using as the day progresses to avoid overwhelming themselves with data and hypothesis. Furthermore, field instructors need to work with students on developing their geologic investigation skills further and improving their efficiency. Specifically students need to collect a higher proportion of structural data and develop complete geologic models earlier in the day. (Abstract shortened by UMI.). [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/en-US/products/dissertations/individuals.shtml.]