Center for Research in Language CRL Newsletter
July, 2000
M a i n
N e w s
I n f o r m a t i o n

July 2000: Vol. 12, No. 2

CRL Newsletter PDFCRL
Newsletter DOC

Objective Visual Complexity as a Variable in Studies of Picture Naming*

Anna Székely
Eotvos Lorand University, Budapest

Elizabeth Bates
University of California, San Diego

*Supported by NIDCD R01-DC00216

A b s t r a c t

Visual complexity is an important variable for studies working with picture stimuli, including picture naming. Traditionally, subjective ratings by 20-30 subjects have been used for this purpose, an approach that may be influenced by perceptual and cognitive variables (e.g., familiarity with the object) that are not directly related to visual complexity. The present study offers an objective and easy way of measuring visual complexity by taking the file size of picture stimuli material (black-and-white, simple line drawings) as the basis. Over 30 different file types and degrees of compression were compared for 520 object pictures, and analyzed to determine whether these measures differ in their influence on picture-naming behavior. Results suggest that PDF, TIFF and JPG formats may provide valid indices of objective visual complexity. The effect of these objective measures on picture naming were compared with published subjective visual complexity data from an English and a Hungarian study on overlapping items. Comparative analysis with other picture-naming variables shows that these objective measures - unlike subjective ratings - have no effect on RT, are unrelated to word frequency or age of acquisition, and show a more modest word length effect on the dominant response. However, they do affect picture-naming accuracy (production of the target name), an effect not reported in previous studies using subjective ratings of visual complexity. Subjective and objective complexity measures are both useful, and they are correlated, but they also differ in potentially important ways.

Note:  In order to view the CRL Newsletter in PDF format, you will need Acrobat Reader 3.0 or greater. Download the most recent version here. When you click on the link to the featured article, Acrobat Reader may open in your browser window. If it doesn't, save the file to your hard drive and open with Acrobat Reader.

[download Acrobat Reader]