Do Glasses Modulate Age Perception? (2024)

Abstract

No formal studies have reported how glasses influence age perception, except for a LondonVision Clinic survey that found that people over 45 look 5 or more years older whenwearing eyeglasses. To investigate the effect of eyeglasses and sunglasses on ageperception while controlling for age and interpersonal differences, we digitallymanipulated the photographs of faces of 50 young adults, to create two age conditions(young and old) and three eyewear conditions (no glasses, eyeglasses, and sunglasses).Participants then estimated the age of the faces, displayed in random order. Contrary tothe generally accepted beliefs that wearing eyeglasses makes you look older and wearingsunglasses make you look younger, our results suggest that the effect of glasses on ageperception is rather small.

Keywords: faces, glasses, sunglasses, age perception, eyewear, eyeglasses

For a society obsessed with beauty and youthful looks, it is surprising that no formal studyhas ever been conducted to investigate how eyewear affects age perception. After all,according to the Vision Council of America, about half of all women and 42% of men weareyeglasses. How glasses alter perception has been investigated along many dimensions such asphysical attractiveness and intelligence (), and identity (), among others. Surprisingly,age was not one of them. The only study, not published in a peer-reviewed journal, wasreported by the Daily Mail on September 21, 2010, and its content was quicklyshared by other news sources. The study was based upon a survey, commissioned by the LondonVision Clinic, where participants were either given 10 pictures of people with eyeglasses or10 pictures of the same people without glasses, and asked to guess their ages. The studyreported that people wearing eyeglasses were perceived as 3.3 years older; glasses-wearersover 45 were thought to be 5 years older.

Although those numbers seem reasonable, the topic merits a more formal approach. To thatextent, we randomly selected 25 male and 25 female faces with neutral emotional expression(see examples at bottom Figure 1A)from the publicly available Karolinska Directed Emotional faces database (Lundqvist etal., 1998). We then usedFaceApp (Wireless Lab, Skolkovo, Russia) to apply an aging filter to make the faces look older(examples at bottom Figure 1C). Thesame application was used to add generic eyeglasses (examples at top of Figure 1A and C), while a different application,Stylish Sun Glasses, was used to add sunglasses (examples at top of Figure 1B and D). Together, the originaland doctored photographs constituted a stimuli set of 300 images. Twenty-one college-aged menand women rated the images, displayed in random order, presented at the center (17° × 23° ofvisual angle) of a Dell 19 in. monitor. More specifically, the participants reported theestimated age of each face using the keyboard. The time limit for each image was 5 seconds. Toanalyze the data, we simply averaged the responses for each participant across each of sixconditions: Two Age Groups (old and young) × Three Eyewear Conditions (eyeglasses, no-glasses,and sunglasses). We then used a paired t test, to probe significance levelsbetween the no-glasses and either sunglasses or eyeglasses condition for each of the two ageconditions. The results indicate that faces of young adults were perceived as slightly (1.3years) but significantly, older when eyeglasses were added (Figure 1A). Adding sunglasses, on the other hand, did notalter the perceived age (Figure1B).

The results obtained with faces digitally altered to look older were remarkably different:adding glasses, in this case, did not alter the perceived age (Figure 1C), whereas adding sunglasses significantlyreduced the perceived age with 1.8 years (Figure 1D). Statistics (two-way repeated measures analysis of variance) also showthat there was a significant interaction between the independent variables (age of face andspectacle type), affecting the perceived age (p = .014).

The notion that sunglasses make older people look younger is not that surprising as thedarkness of the glasses helps to hide the wrinkles around the eyes that show earlier signs ofa*ging. We do not have a straightforward explanation for why our results contradict with thoseconducted, albeit not peer-reviewed, by the London Vision Eye Clinic, which received a lot ofnews coverage. A possible explanation is that participants either rated 10 images of faceswith eyeglasses or 10 without glasses, which unlike our randomized and controlled experiment,might cause priming effects. It is also noteworthy that the London Vision Eye clinic provideslaser eye surgery, which implies a conflict of interest. When posing with eyeglasses, theconditions (facial expression, lighting, makeup, etc.) might have been different, to look lessflattering, compared with posing without glasses. Our study was controlled to avoid such abias (unchanged facial expression and environmental conditions and addition of genericglasses); at the same time, we acknowledge that there is a possibility that the digitallymodified stimuli do not necessarily generalize to real faces. In addition, publicly availableface databases, such as the one we used, mainly contain Caucasian faces, where a more diverseface database is desired. Other potential shortcomings are the use of only one style ofeyeglasses and sunglasses, and reliance of a participant pool consisting of young adults,which can potentially introduce an own-age bias. A follow-up study might reveal whether thisis the case, and whether the reported findings can be extrapolated to older participants. Inconclusion, our study suggest that eyewear hardly modulates how we perceive age, which isimportant because wearing eyeglasses can significantly impact one’s self-esteem (Harris etal., 1982). Our results arealso a reminder of how non-peer reviewed research, funded by entities with specific interests,might shape public perception.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research,authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research,authorship, and/or publication of this article: Nicolas Brunet received grant support by anInstitutional Development Award from the National Institute of General Medical Sciences ofthe National Institutes of Health (INBRE, P20GM103476). Its contents are solely theresponsibility of the authors and do not necessarily represent the official views ofNational Institutes of Health.

ORCID iD

Nicolas M. Brunet https://orcid.org/0000-0002-1145-269X

References

How to cite this article