The cognitive snare of Mona Lisa’s stare

Have you stood in line at the Louvre in Paris to catch a glimpse of famous Mona Lisa’s smile? People say that besides her ambiguous smile, her gaze tends to follow you, which might add to the mysterious aura that has been created around her for centuries.

Leonardo da Vinci. Cropped and relevelled from File:Mona Lisa.

There is a considerable body of evidence that attests to the power of gaze, which is one of the strongest social cues we have and make use of. For instance, even infants who are only a few days old have learned to detect when someone makes eye contact with them which suggests that eye contact is one of the cornerstones in developing social skills.

While there seems to be little dispute about the importance of eye contact — or “direct gaze” — there have been contradictory findings with regard to what influence direct gaze has on your mental capabilities. Some researchers have found that eye contact enhances memory and the corresponding brain areas; however, the same authors have found that direct gaze can also hinder performance in other cognitive tasks such as the Stroop task.

A recent article in Psychonomic Bulletin & Review examined this issue and elucidates what Mona Lisa’s stare can do to your mind.

Researchers Jessica Wang and Ian Apperly focused on the effect direct gaze has on memory for stimuli other than faces, since little is known about that so far. The researchers were particularly interested in visual working memory, since this cognitive ability allows bridging between perceptual inputs and the formation of conceptual representations.

For this purpose, Wang and Apperly used a one-shot change detection paradigm. As shown in the figure below, the change detection displays contained either 3 or 4 colored agents along with a matching number of differently shaped objects. Each display was initially presented for 100 ms, and was followed by a 900-ms retention interval. A test picture was then displayed until participants made a response. In the look-at-object condition, each agent always looked towards an object. In the look-at-you condition, the agents always looked straight ahead giving the impression that they were looking at the participants.

Figure 1 in the featured article.

Half of the time, the test pictures were identical to the sample pictures, the other half of the time, the test pictures contained a shape change of one of the objects or a color change of one of the agents.

Wang and Apperly hypothesized that if direct gaze facilitates encoding, then change detection accuracy should be higher in the look-at-you than the look-at-object condition. Conversely, if direct gaze disrupts visual working memory, then results should be exactly the other way around.

The fact that participants only had 100 ms to encode the initial displays provided a strong incentive to optimize performance by ignoring the irrelevant elements of the display.

Wang and Apperly found that having the agents look at the observers decreased their change detection performance by about 10% (for both objects and agents), which is in line with the predictions of a disruptive effect of direct gaze.

One might wonder whether the inferior performance in the look-at-you condition was really due to disruption resulting when the agents were staring at you, or whether the gaze of the agents towards the objects might increasing encoding performance. There is evidence from a paper previously published in Psychonomic Bulletin & Review that task-irrelevant gaze cues towards objects lead to more gaze-following towards the gazed-at objects. Similarly, other evidence suggests that objects which are close to an agent’s uninformative gaze also seem to be found more quickly.

To test this alternative account the authors simply compared the look-at-object vs. a look-away condition, in which the agents looked away from anything (see the figure above). No difference in performance was observed between those two conditions. So having an agent look at an object does not really buy you anything for a change detection task, implying that the disadvantage of the look-at-you condition likely reflected a disruption.

Digging deeper, Wang and Apperly wondered how long-lived the disturbing effect of direct gaze is and conducted a final experiment in which the initial  100 of direct gaze was followed by an averted gaze for 300 ms following  (the look-at-you-then-away condition). Wang and Apperly compared performance in this condition to two other conditions: Either the initial 100 ms of direct gaze stimulus was simply extended for a further 300 ms before the blank-screen retention interval (allowing some extra time for encoding after spontaneous recovery from initial disruption), or the 100 ms of direct gaze was followed by 300 ms of a blank screen before the retention interval (giving no further encoding opportunity, but matching the overall length of the trial sequence).

Wang and Apperly found that compared to the baseline condition, in which 100 ms of direct gaze was followed by a blank screen, performance improved when the initial direct gaze was either prolonged by another 300 ms or followed by 300 ms of averted gaze. It therefore seems that it does not matter whether the extended exposure to the stimuli included direct or averted gaze; improvement was always observed.

This result suggests that the critical factor for recovery is the additional encoding opportunity, which allows participants to overcome the initial disruption from direct gaze. These latter findings tell us that while direct gaze seems to cause a rapid disruption (within 100 ms), this disruptive effect is quite short-lived (shorter than 400 ms) — at least when participants are occupied with a change detection task.

Taken together with previous findings these latest results imply that direct gaze facilitates tasks where eye gaze is informative — such as those involving processing of facial contents — but it decreases performance when your task at hand does not require processing someone’s eye gaze.

So for those of you planning a visit to the Louvre, you might want to fight the crowds behind you for a moment longer to give Mona Lisa some time until the effect of her disturbing gaze wears off.

Article focused on in this post:

Wang, J. J., & Apperly, I . A. (2016). Just one look: Direct gaze briefly disrupts visual working memory. Psychonomic Bulletin & Review. DOI: 10.3758/s13423-016-1097-3.

The Psychonomic Society (Society) is providing information in the Featured Content section of its website as a benefit and service in furtherance of the Society’s nonprofit and tax-exempt status. The Society does not exert editorial control over such materials, and any opinions expressed in the Featured Content articles are solely those of the individual authors and do not necessarily reflect the opinions or policies of the Society. The Society does not guarantee the accuracy of the content contained in the Featured Content portion of the website and specifically disclaims any and all liability for any claims or damages that result from reliance on such content by third parties.

You may also like