People often communicate with pointing gestures in their daily life. It was previously found that a point does not indicate a specific location but rather a (small) area. We still do not know, which spatial information pointer expect they are conveying with their gestures. We assume pointers scan the target’s surrounding that they believe the gesture is covering to find an appropriate target description, so pointers’ gaze pattern should represent the area of subjective uncertainty associated with their pointing gesture. We hypothesized that pointers are overconfident concerning their gesture’s accuracy and therefore, the fixated area should be smaller than the area in which the observers’ interpretation fall.
To examine on this, two participants holding either the pointer or observer role were invited to the lab and were seated in front of a big screen. In each trial, the pointer had to point towards one target shape among 179 similar distractor shapes – or depending on condition, could additionally describe it – while the observer had to identify the target shape. To measure subjective uncertainty, eye movements of both, observers’ answers and identification confidence were recorded. Unsurprisingly, correct target identification was enormously facilitated, and confidence rating was higher when pointing was accompanied by verbal description. Furthermore, the results support our hypothesis as the fixated area by pointers was smaller than the observers’ interpretation area and additionally, both centers significantly differ. Thus, pointers overestimate the accuracy of points and did not consider interpretation biases and situational factors.