‘Selfies’ Reveal Systematic Deviations from Known Principles of Photographic Composition

in Art & Perception
Restricted Access
Get Access to Full Text
Rent on DeepDyve

Have an Access Token?

Enter your access token to activate and access content online.

Please login and go to your personal user account to enter your access token.


Have Institutional Access?

Access content through your institution. Any other coaching guidance?


We used ‘selfies’, self-portraits taken with a hand-held smartphone camera, to test three known principles of photographic composition: The rule of thirds, the golden ratio rule, and the eye centering principle. Although they are often taught and discussed, the origin of these principles remains unclear. It is possible that they stem from constraints on human perceptual processes. Alternatively, these principles might serve more practical purposes, such as forcing photographers to explore all quadrants of the image. Selfies provide an ideal test bed for these questions due to the control they give self-photographers when they compose the photograph. We used a database of images created by non-professional photographers (N=388). After analysis, we conclude that there little support for any of the three principles, suggesting that none is strongly rooted in spontaneous perceptual preferences.

‘Selfies’ Reveal Systematic Deviations from Known Principles of Photographic Composition

in Art & Perception



AmirshahiS. A.RediesC.DenzlerJ. (2012). Assessment of the Rule of Thirds in Photographs Taken by Amateurs. Paper presented at VSAC Conference Alghero Italy.

BBC News magazine (2013). Self-portraits and social media: the rise of the ‘selfie’. 7 June 2013. http://www.bbc.co.uk/news/magazine-22511650.

BertaminiM.BennettK. M.BodeC. (2011). The anterior bias in visual art: the case of images of animalsLaterality 16673689.

BrunoN.BertaminiM. (2013). Self-portraits: smartphones reveal a side bias in non-artistsPLoS One 8e55141.

DattaR.JoshiD.LiJ.WangJ. Z. (2006). Studying aesthetics in photographic images using a computational approach in: ECCV 2006 Lecture Notes in Computer ScienceLeonardisA.BischofH.PinzA. (Eds) pp.  288301. Springer VerlagBerlin, Heidelberg, Germany.

FausingB. (2013). Selfies and the Search for Recognition. See for Your Selfie. Web article retrieved on November 1 2013 http://www.academia.edu/4418191/Selfies_and_the_Search_for_Recognition._See_for_your_Selfie.

FechnerG. Th. (1876). Vorschule der Aesthetik2nd edn. Breitkopf and HärtelLeipzig, Germany.

GionniR. (2013). Autoritratti con lo smartphone: una Ricerca Empirica. Unpublished undergraduate thesis Università di Parma Italy.

GreenC. D. (1995). All that glitters: a review of psychological research on the aesthetics of the golden sectionPerception 24937968.

HögeH. (1997). The golden section hypothesis: its last funeralEmpir. Stud. Arts 15233255.

HoughtonD.JoinsonA.CaldwellN.MarderB. (2013). Tagger’s Delight? Disclosure and Liking in Facebook: the Effects of Sharing Photographs amongst Multiple Known Social Circles Discussion Paper University of Birmingham Discussion Paper Series Birmingham.

KragesB. (2005). Photography: The Art of Composition. Allworth PressNew York, NY, USA.

LenmanR. (Ed.) (2005). The Oxford Companion to the Photograph. Oxford University PressOxford, UK.

LiuL.ChenR.WolfL.Cohen-OrD. (2010). Optimizing photo compositionComput. Graph. Forum 29469478.

LuoY.TangX. (2008). Photo and video quality evaluation: focusing on the subject in: ECCV 2008 Part III LNCS 5304 D. Forsyth P. Torr and A. Zimmerman (Eds) pp. 386–399. Springer Verlag Berlin Heidelberg Germany.

MachajdikJ.HanburyA. (2010). Affective image classification using features inspired by psychology and art theor Paper presented at the International Conference on Multimedia New York NY USA.

McManusI. C.HumphreyN. K. (1973). Turning the left cheekNature 243271272.

McManusI.StöverK.KimD. (2011). Arnheim’s Gestalt theory of visual balance: examining the compositional structure of art photographs and abstract imagesi-Perception 2615647.

McManusI. C.ThomasP. (2007). Eye centring in portraits: a theoretical and empirical evaluationPerception 36167182; discussion 183–188.

McManusI. C.WeatherbyP. (1997). The golden section and the aesthetics of form and composition: a cognitive modelEmpir. Stud. Arts 15209232.

OhtaH. (1999). Preferences in quadrangles reconsideredPerception 28505517.

PacioliL. (1509). De divina proportione (English: On the Divine Proportion) Luca Paganinem de Paganinus de Brescia (Antonio Capella): Venice.

PetersonB. (2003). Learning to See Creatively: Design Color and Composition in Photography. Amphoto PressNew York, NY, USA.

PhillipsF.NormanJ. F.BeersA. M. (2010). Fechner’s aesthetics revisitedSeeing Perceiving 23263271.

PlugC. (1980). The golden section hypothesisAm. J. Psychol. 93467487.

RussellP. A. (2000). The aesthetics of rectangle proportion: effects of judgment scale and contextAm. J. Psychol. 1132742.

SantellaA.AgrawalaM.deCarloD.SalesinD.CohenM. (2006). Gaze-based interaction for semi-automatic photo cropping Paper presented at the SIGCHI Conference on Human Factors in Computing Systems New York NY USA.

SmithJ. T. (1797). Remarks on Rural Scenery with Twenty Etchings of Cottages from Nature and Some Observations and Precepts Relative to the Picturesque printed for and sold by Nathaniel Smith Ancient Print Seller at Rembrandt’s Head May’s Buildings St. Martin’s Lane and I. T. Smith at No 40 Trith Street Soho London UK.

TylerC. W. (1998). Painters centre one eye in portraitsNature 392877878.

TylerC. W. (2007a). Some principles of spatial organization in artSpat. Vis. 20509530.

TylerC. W. (2007b). Eye-centering in portraits: reply to McManus and ThomasPerception 36183188.


  • View in gallery

    Portrait condition results. Relative positions of five candidate face features (right eye pupil, left eye pupil, nose center, right mouth corner, left mouth corner) are plotted against the predictions of the ROT (dashed grey lines), GRR (solid grey), and ECP (dotted grey). Right and left refer to the self-portrait facing towards the viewer, not to the position relative to the page. Thus the right eye is actually plotted on the left in the diagram. This figure is published in color in the online version.

  • View in gallery

    Landscape condition results. All plotting conventions as in Fig. 1. This figure is published in color in the online version.

  • View in gallery

    Portrait forcing 3/4 condition results. All plotting conventions as in Figs 1 and 2. This figure is published in color in the online version.

  • View in gallery

    Relative frequencies of selfies with at least one feature falling within a given tolerance area around the lines predicted by the ROT (left) and the GRR (right). With a 2% tolerance, about 30% of images were consistent with the ROT predictions; about 40% were consistent with the GRR predictions. This figure is published in color in the online version.


Content Metrics

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 107 107 50
Full Text Views 67 67 42
PDF Downloads 9 9 5
EPUB Downloads 7 7 0