The ability to estimate age has evolved as it enables an individual to evaluate, for example, the suitability of a potential mate 
. Perceived age is socially relevant to many individuals as evidenced by the large and global cosmetics industry. In addition, perceived age has been shown to be predictive of mortality in elderly individuals and associated with predictors of age-related diseases independently of chronological age 
indicating its utility as a biomarker of aging. Biomarkers of aging are measures of an individual's or tissue's biological age (i.e. how well they are aging considering their chronological age) and are important in gerontology and epidemiology research for identifying factors that influence the aging process. However, there has been little work to systematically determine which physiological features predominately influence how old women look for their age and, hence, which features drive the link between perceived and biological age.
The influence of particular features on the perception of age depends on the context in which a subject is viewed. For example, photographs can be viewed with or without hair and clothing cues (e.g. passport-type 
versus facial images 
), and in-person evaluations of age 
can be influenced by speech and body movement cues. The benefit of using photographic images is that the cues present for age estimation can be controlled and standardized. Additionally, estimating age from images has been shown to be highly reproducible when employing large numbers of age assessors 
. Thus, photography is now the predominant method for generating perceived age.
Previous perceived age work has found that increased sun-damage 
, male pattern baldness 
, gray hair 
, under eye wrinkles and bags 
, pigmented spots 
, skin topography (i.e. skin micro-texture and wrinkles) 
and reduced skin color uniformity 
are associated with looking older for one's age. However, the use of different methods to estimate perceived age and the different features measured in each study make it difficult to ascertain which skin and hair aging features dominate the perception of age. Additionally, it is unclear how the use of different photographic images in each study affects the relationship between perceived age and skin and hair aging features. Modern cosmetic surgical techniques target subcutaneous tissues to create a youthful look 
suggesting that physiological changes underneath the skin are important modulators of how old one looks. Evidence for the link, though, between changes to subcutaneous tissue and age perception is limited, partly due to the difficulty in measuring subcutaneous changes.
The rate and degree to which physiological changes occur are determined by genetic and environmental factors. For example, around 80% of the variation in male pattern baldness in young and old men can be attributed to genetic factors 
. However, the degree of influence genetic factors have on features of female appearance are largely unknown. For example, whether it is sun-exposure or genetic factors that mainly underlie the variation in sun-damaged skin present in Caucasian populations is unknown. Estimating the influence genetic factors have on a particular feature can indicate, for example, the utility of using Genome Wide Association (GWA) approaches to investigate the etiology of the feature.
Here, to investigate the link between perceived age and biological age we have investigated the strength of relationship between how old women look for their age and skin, hair and lip aging features. Additionally, we investigated the impact on such relationships of using different types of photographic image to generate perceived age. To determine whether future investigations into the identity of factors that influence facial aging should focus on genetic or environmental factors, we have utilized monozygotic and dizygotic twins to estimate the influence genetic factors have on perceived age as well as on skin, hair and lip aging features. Furthermore, we have constructed composite images of individuals who look either young or old for their age to demonstrate the striking differences in perceived age that can occur between individuals of the same chronological age.