This study investigated the effects of three different navigation system human-machine interfaces (HMIs) on driver eye glance behavior, navigational errors, and subjective assessments. Thirty-six drivers drove the same unfamiliar 3-segment urban route in downtown Detroit during weekday daylight hours. The three HMIs were: i)a 2D or ii) a 3D (with level-of-detail icons in perspective) electronic map display, both with standard voice prompts; or iii) a 3D map display augmented by a separate display of photorealistic intersection images and landmark-enhanced voice prompts. Each participant drove the same three route segments in the same order but was assigned a different HMI condition for each segment as per a 3-period/3-treatment crossover experimental design. Results indicated that drivers in real traffic made prudent use of their visual attention with advanced navigation systems HMIs, always within recent US Department of Transportation recommended visual distraction limits. There were significantly more missed turns in the first route segment, regardless of HMI condition, attributable to greater route complexity and one late-onset voice prompt. Participants asked to rate the HMI were substantially influenced by the context in which that HMI was used. Acceptability and usefulness ratings were generally positive, with negative ratings for all three navigation systems predominantly arising from the first route segment. Stress and workload ratings were affected by route segment but not HMI. However, ratings on the TLX ‘Frustration’ workload subscale were significantly lower for the enhanced 3D+ navigation system in comparison to the 2D system, but not significantly different from the 3D system, suggesting the enhanced voice prompts were beneficial. Participant preferences favored the 2D to the 3D map display by a 2 to 1 margin; the availability of photorealistic landmark imagery over no imagery by a 3 to 1 margin; and enhanced over standard voice prompts by a 9 to 1 margin. When presented in a timely fashion, the availability of landmark information may be desirable to support wayfinding, reduce frustration, or accommodate driver preferences.