The Seductive Cartoons Effect
Sometimes authors insert cartoons as adjuncts to text in the hopes that the reading experience will be more enjoyable and interesting. The literature on the role of illustrations as adjuncts to text suggests five possible functions: decorational, representational, organizational, interpretational, and transformational (Levin, 1981). A meta-analysis revealed that only studies where illustrations were used for decorational functions resulted in no beneficial text-learning effect, whereas illustrations used for the other functions resulted in positive effects (Levin, Anglin, & Carney, 1987). Beyond no effects, can certain types of decorational illustrations actually produce negative effects?
An attempt by authors to make text reading more interesting that may produce negative effects is to include interesting, but not always relevant, details in texts. Certain topics, such as danger, power, and sex, are almost universally interesting (Hidi & Baird, 1988). Such details are sometime referred to as seductive details (Garner, Brown, Sanders, & Menke, 1992; Garner, Gillingham, & White, 1989; Harp & Maslich, 2005; Harp & Mayer, 1997, 1998; Schraw, 1998; Wade, Schraw, Buxton, & Hayes, 1993).
Although adding seductive details may enhance interest and capture students’ attention, these details may not make other material (i.e., target material that students are expected to learn) more interesting (Dewey, 1913; Harp & Maslich, 2005; Harp & Mayer, 1998; Mayer, Heiser, & Lonn, 2001; Mayer & Jackson, 2005). In fact, adding seductive details may actually hinder learning target material. Garner et al. (1989) had students read expository passages with and without seductive details and then identify target material in text. Seductive details decreased student learning of target material (i.e., target content), even with skilled readers. Garner et al. (1992) replicated these results and suggested that including seductive details in text almost always interferes with learning of target material.
In the present study, we sought to determine if the use of illustrated cartoons as adjuncts to text could actually impair students’ ability to comprehend the text. We selected six pages from a leading educational psychology textbook (Woolfolk, 2006) that included illustrated cartoons. To create a version without cartoons, we simply overlaid the cartoons with white paper and photocopied the pages again. Using the publisher’s test bank, we selected 14 multiple-choice questions that referred to information presented in the six pages.
In Experiment 1, ninety-one undergraduates were given 35 min. to read the text that either contained cartoons or not, and then had 10 min. to answer the 14 comprehension questions. There was no effect for cartoons, t(89) = 1.02, p = .16. Students who saw no cartoons (M = 9.98, SD = 2.04) did not score statistically higher than those who did see cartoons (M = 9.34, SD = 2.12).
We found two potential reasons for our null effect. First, several students reported they did not have enough time to read all of the pages. Second, we used items straight from a test bank. Several of the items had distractors that no one selected. There was no evidence of a ceiling effect. Both groups averaged around nine out of the 14 items correct, with standard deviations of around two.
In Experiment 2, we gave students more time to read (45 min. vs. 35 min.) and we also attempted to make the comprehension items more difficult by improving the distractors. Thus, we simply wanted to (1) ensure that all text was read and (2) make the test more difficult to see if this would reveal a cartoons effect, realizing that the length of our materials and learning sessions were brief compared to typical classroom learning conditions.
In Experiment 2, thirty-six undergraduates participated. This time there was a cartoons effect, t(34) = 2.93, p = .006. Students who viewed cartoons (M = 7.61, SD = 1.69) scored worse than those who did not see cartoons (M = 8.94, SD = 0.94).
These findings have both theoretical and practical implications. Theoretically, we found that certain types of decorational illustrations may not only have zero effect on learning, but may actually be detrimental to learning. It is likely that this seductive cartoons effect operates similarly to the seductive details effect. These interesting, yet irrelevant, snippets distract learners from more important content by seducing learners into devoting attention to them.
Practically, we find it somewhat ironic that a leading educational psychology textbook, that is filled with advice on how to improve learning, inserts cartoon illustrations that are detrimental to learning. Of course, we do not blame the author here. We imagine that there are several other textbooks that contain such seductive cartoons.
Back to Expanded View Close this window