Talk:The Big 5 as a Model of Personality Perception: “That line’s got personality!”
From PsychWiki - A Collaborative Psychology Wiki
This is not the first experiment in which participants were instructed to use words to describe lines without inherent meanings. Gestalt psychologist Wolfgang Köhler (1929) presented research participants with two line scribbles and asked them to match the line scribbles to two nonsense words, baluma and takete (he later changed the words to maluma and takete - see http://www.swarthmore.edu/SocSci/Linguistics/Papers/2007/hunter-smith_sarah.pdf). What is intriguing about the Köhler study is that neither the lines nor the nonsense words had formal meanings, yet participants perceived something about the qualities of both the lines and the words that enabled them to match them with complete consensus. (Later attempts to replicate and extend the study cross-culturally have been only partially successful).
But there is a big difference between the Köhler study and the present one. The Köhler study is actually more similar to an individual differences study in that two line scribbles that differed in form were judged to be baluma or takete. The present study calls for Big 5 ratings of a single, simple line. If several different line forms had been used, it would be interesting to see if they would be differentially described and what the degree of consensus across judges would be for the different line forms. As the current study was conceived and executed, there is nothing at all in the line to call forth particular judgments along the Big 5 dimensions. That the Big 5 were represented in a confirmatory factor analysis of judgments of the line could only be due to the Big 5 serving as categories of perception that are projected onto the line. There are at least two different ways that one could interpret this.
One interpretation would draw a parallel to the expression and perception of emotion. Numerous studies have shown that facial expressions of the basic emotions are universal across cultures as is the perception of emotion in facial expressions. The categories of perception for facial expression therefore allow for the veridical perception of emotions in facial expressions.
A second interpretation is more discomforting. It is reminiscent of anthropologists D'Andrade's (1974) and Shweder's (1975) (see also Shweder & D'Andrade, 1979, 1980) semantic similarity criticism of personality ratings. Their argument is that when judges rate personality with semantically similar trait words, a factor analysis of these ratings reveals mere semantic similarities across words rather than covariation in actual behavior. Although Block, Weiss, and Thorne (1979) and Romer & Revelle (1984) provided effective refutations of D'Andrade and Shweder's thesis when considering personality judgments of real people, the results of the present study employing a line as the target of personality judgment might indeed be explained in terms of the D'Andrade and Schweder (1979) hypothesis.