People with high levels of social anxiety do not necessarily perceive others' gazes as directed at them more often than others do. This is the conclusion of a group of researchers led by Madeleine E. Moses-Payne of University College London. The work was published in the Journal of Affective Disorders.
In psychology there is a concept of the "gaze cone" - the range of angles at which a person believes an interlocutor is looking directly at them. It was assumed that socially anxious people have a wider cone: they more often interpret a gaze "past" them as directed at themselves. However, the new data cast doubt on the stability of this link.
First, the researchers conducted a meta-analysis - a statistical pooling of the results of several independent studies. The first review included ten studies with a total of 403 participants. Overall, a weak link was found between social anxiety and a wider gaze cone, but the data proved extremely heterogeneous. Small samples and a large spread of results did not allow for a reliable effect to be claimed.
The second meta-analysis (nine studies, 523 participants) tested how emotions affect gaze perception. People slightly more often perceived angry faces as looking directly at them than neutral ones, but here too the spread between studies was very high.
Because past data were not stable, the authors conducted three new online experiments with 514 participants. Participants were shown faces with eyes deviated at various angles for fractions of a second - from a direct gaze to eight degrees to the side. They had to determine whether the person was looking at them. All also completed a standard questionnaire on social anxiety symptoms.
In the first experiment (193 people), the faces were neutral. No link was found between anxiety level and gaze cone width: anxious and non-anxious participants assessed gaze approximately the same way.
In the second experiment (172 people), angry faces were shown. Here a very weak effect appeared: for each one-point increase in anxiety score, the cone widened by only 0.01 degrees. However, this result was not confirmed in the third experiment (149 people), where faces were both neutral and angry. Neither anxiety nor emotion affected gaze perception.
A combined analysis of all three new experiments found no overall link between social anxiety and the gaze cone. The only positive result from the second experiment dissolved in the overall data.
The authors note limitations. Participants were not clinically diagnosed - they only completed a questionnaire. In addition, the experiments were conducted online, without a researcher present. The live presence of an observer could have increased anxiety and influenced perception.
The scientists emphasize: correlation (a statistical link) between anxiety and gaze perception features does not in itself prove causation. The results obtained do not confirm that the gaze cone can be used as a biological marker of social anxiety disorder. Further work could test whether perception changes when social stress is created, for example before a public speech, or using eye-tracking - technology for tracking eye movements.