Adaptive social approach and avoidance behavior is of substantial importance for social functioning. An imbalance in social approach and avoidance tendencies may constitute a risk factor for the etiology and maintenance of mental illness such as social anxiety (SA). The present study investigated how trait SA influences gaze behavior, place preference, and autonomic responses (heart rate and skin conductance) using a naturalistic two-phase design. The first phase was a half-hour walk on a freely chosen route; the second phase entailed a staged social interaction with an alleged test subject (confederate) starting with a brief waiting period followed by a short interaction initiated by the confederate. Additionally, we introduced a between-subject gaze-camouflage condition. Participants wore shaded or clear glasses to test the assumption that people direct their overt attention in a less socially normative way when their eye-movements cannot be observed. Participants were pre-screened and selected to show a high variance in SA. Preliminary results of the walking phase (N = 30 of pre-registered 90) indicate that SA as measured by the SIAS was negatively associated with the frequency of person-centered fixations. Preliminary results of the lab phase (N = 70 of 90) show a negative relationship between SA and fixations on the interaction partner during the waiting phase, but no significant effect during the interaction. We did not observe a significant effect of SA on autonomic responses in either the first or second phase of the experiment. Furthermore, gaze camouflage had no effect in any phase on any measure.