Through the Lunar Advanced Science and Exploration Research (LASER) program, NASA has supported an ongoing research program at the University of Maryland and Arizona State University on collaboration between humans in extravehicular activity (EVA) and robotic systems performing scientific exploration of planetary surfaces. These field tests have been named, with a nod to NASA's longstanding series of Desert Research and Technology Studies (RATS) tests, the Desert Field Lessons in Engineering And Science, or Desert FLEAS. This paper presents the results of the third set of field tests in this program, which consisted of a full week of field trials at SP Crater near Flagstaff, Arizona in June, 2012. Trained field geologists from Arizona State University served as test subjects for series of field exploration sorties. Each subject performed three similar sorties: one in shirtsleeves as a control; one in the MX-B pressure suit simulator to replicate the restrictions of an actual EVA; and one in MX-B directly assisted by RAVEN, including providing a ride for the suited test subjects to and from the science sites. Continual full-body biomechanics data was collected by a conformal body suit worn under the liquid cooling garment, which incorporated 18 inertial measurement units which document the motions of all major body joints throughout the sortie. Post-test subjective evaluations were collected based on the NASA task load index (TLX) protocol and Cooper-Harper ratings, along with an evaluation of the scientific exploration performance (in terms of noted observations, collected sample number and quality, and correct interpretation of data in real time) of the subject in each of the operating modes. These tests were performed both in daylight and in darkness, using lighting provided by the suits and the rover. The 2012 Desert FLEAS tests sought to provide rigorous quantitative data on the benefits and limitations of robotic augmentation of EVA for geological science data collection, including a statistically significant number of trained geologist test subjects. Following the presentation of these results, the paper briefly outlines the plans for the last two years of Desert FLEAS testing, which will focus on more extreme terrains, more advanced robotic systems, and multi-person, multi-robot exploration teams.