Academic journal article Canadian Journal of Experimental Psychology

Conformity in a Generative Linguistic Task: The Role of Category and Strategic Nonword Use

Academic journal article Canadian Journal of Experimental Psychology

Conformity in a Generative Linguistic Task: The Role of Category and Strategic Nonword Use

Article excerpt


Marsh, Ward, and Landau (1999) demonstrated that participants asked to create novel words use elements of sample nonwords they are given, even when instructed to avoid use of the examples. In four studies, we replicated the effect of conformity to sample nonwords and found the effect was not influenced by the semantic category of the words unless those words shared orthographic characteristics. We found that although we could increase conformity to examples when word exemplars were grouped by category, it was likely that much of this increase was strategically driven. We propose that the presence of the sample non-words, presented in groups with the same word rules, created an orthographic category used by participants in the word creation task.

Plagiarism is the intentional use of another's ideas or words without crediting the source. However, conformity to another's ideas or words can occur without the conscious intent to plagiarize. This conformity is sometimes termed unconscious or inadvertent plagiarism and has been studied to examine the constraints inherent in everyday creativity. The theory of structured imagination (Ward, 1994) posits that imaginative creations are constrained by established knowledge structures. While novelty requires being original and producing something completely new, this production is created from elements of knowledge already stored.

In tests of the theory of structured imagination, and of conformity, participants are asked to generate some response under different experimental conditions. For instance, Ward (1994) asked participants to create alien creatures, cueing them with a basic starting point such as, a feathered creature living on a planet of molten lava. Ward found that the generated creatures were constrained by the features of Earth animals cued (e.g., wings, cued by feathered) or Earth-bound creatures more generally (e.g., bilateralism).

Conformity to stored or presented information has also been found to occur in the attribution of generated solutions in problem-solving tasks (Marsh, Bink, & Hicks, 1999; Marsh & Bower, 1993; Marsh, Landau, & Hicks, 1997) and in both visual and linguistic generative tasks (Bredart, Ward, & Marczewski, 1997; Marsh, Ward, & Landau, 1999; Ward, 1994). In a study of visual creativity and conformity, it was found that participants would use elements of given examples when asked to supply novel generations. This occurred even when participants were explicitly told to refrain from using examples supplied and that the use of the examples would adversely affect their own ability to be creative (Experiment 3, Smith, Ward, & Schumacher, 1993).

The conformity phenomenon has also been studied using both real and novel words of category members (Tenpenny, Keriazakos, Lew, & Phelan, 1998). Tenpenny et al. (1998) asked participants to learn lists of words from categories for later recall. Some of the categories were comprised of real words and some of the categories were comprised of novel words (non-words). Following recall, when asked to generate their own category members, participants repeated already studied real words as their own contributions but did not conform in this manner to the nonwords. However, as pointed out by Marsh, Ward, et al. (1999), the failure to find a conformity effect for the nonwords may be due to two factors: The study phase for the nonwords used by Tenpenny et al. may not have produced an appropriately strong enough categorical representation for later use and, instead a pseudocategory based on the orthographic properties of the nonwords may have been formed. Participants may have used these orthographic features to produce their nonword examples. Indeed, there was some conformity to letter combinations observed in Tenpenny et al.

To test the orthographic hypothesis, Marsh, Bink, and Hicks (1999) created nonwords that exhibited orthographie regularity and matched these sample nonwords with category exemplars. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.