This article applies to:
E-Prime Challenge Winner 2018!
Experiment Author: Alexandros Rouchitsas, Luleå University of Technology
Both gaze direction and gaze duration convey valuable information about conspecifics’ communicative intentions. But, how do gaze direction and duration interact to affect an inference about such intentions? Do we infer intention to engage socially if we cross gazes for a mere second? Or is prolonged eye contact required in order for any expectation that a social scenario may soon unfold to be raised?
In a facial expression identification task, an avatar face was presented with a neutral expression and direct or averted gaze for 0.5, 1.2, 3.3 or 6 seconds (closed eyes served as our control condition). Then, it produced either a social expression (polite smile) or an arbitrary one (cheek puff). Participants fixated on the avatar’s eyes and made speeded judgments about said expressions, using a two-choice keypress response. In the 0.5 and 1.2-second mutual gaze conditions, we anticipated both expressions to be identified equally fast, since a brief glance should not have raised any expectation of social interaction. In the 3.3-second mutual gaze condition, we anticipated that participants would be faster and more accurate in identifying the social expression, as some expectation should have already been raised. In the 6-second mutual gaze condition, we anticipated both expressions to be identified equally fast, since an expressionless stare should have had a distracting or discomforting effect, causing performance to decline. In the averted gaze and closed eyes conditions, we anticipated both expressions to be identified equally fast, irrespective of gaze duration, as no expectation of social interaction should have ever been raised. Results are discussed in connection with known socio-temporal neurocognitive mechanisms. Implications for designing efficient human-robot interactions are also discussed.