This isn’t specifically about vision, but Strick et al  (Cognition) report on a study of the effect of stimuli paired with pictures of attractive or unattractive faces with gaze averted from or directed towards the viewer.
In the first trial the images of faces were paired with unknown peppermint brands and in the second the images were paired with positive and negative adjectives. It was ensured that volunteers attended to the faces by the inclusion of images of a third condition (faces with eyes closed). When one of these was detected, the volunteer had to press a button.
To determine the effect of attractiveness and gaze on desire, the volunteers were asked to rate the peppermint brands for desirability at the end of the test. Mean self-reported desire for brands associated with the attractive/direct gaze condition was 4.46 out of a possible 7 and for the attractive/averted condition was 4.05. Direct gaze resulted in significantly higher desirability where the face was attractive (p= .04). For the unattractive faces, averted gaze led to fractionally but not significantly higher ratings than direct gaze.
The adjective experiment is an affective priming test. In affective priming, respondents must indicate whether an adjective they read is positive (‘exciting’, ‘happy’) or negative (‘boring’, ‘angry’) by pressing one of two keys. Responses to positive adjectives are generally faster than responses to negative adjectives, but priming can exaggerate or eradicate this difference. The experiments showed that attractive faces lead to a greater difference in response time than unattractive faces (a mean difference between positive and negative judgements of 55ms for attractive faces versus 36ms for unattractive faces, p =0.4). Attactive faces with direct gaze also made the response time difference significantly longer than unattractive faces with direct or averted gaze (both p < .02). However for unattractive faces there was no significant effect on the difference in in response times.
This doesn’t seem to have made the same splash as last November’s press coverage of the Conway et a  (Royal Soc) study that showed the same face is more attractive when gazing directly at the viewer than when not. It seems that for some of us, the power of our gaze may not be all we’d hoped it might be. Especially if we’re trying to sell mints.
Strick, M., Holland, R.W., van Knippenberg, A. (2008). Seductive eyes: Attractiveness and direct gaze increase desire for associated objects. Cognition, 106(3), 1487-1496. DOI: doi:10.1016/j.cognition.2007.05.008
C.A. Conway, B.C. Jones, L.M. DeBruine, and A.C.Little. “Evidence for Adaptive Design in Human Gaze Preference.” Proc. R. Soc. B 275, no. 1630 (January 2008): 63-39.
* You Could Have it So Much Better, Franz Ferdinand.
“The last message you sent said I looked really down,
and that I ought to come over and talk about it. Well,
I wasn’t down;
I just wasn’t smiling at you.”