Message processing of evidence and long-term retention and judgment of beliefs.

• Main Author: Baesler, Erland James.
• Other Authors: Burgoon, Judee
• Language: en
• Published: The University of Arizona. 1991
• Subjects: Long-term memory; Evidence; Juvenile delinquency
• Online Access: http://hdl.handle.net/10150/185392

This communication study investigated characteristics of evidence that influenced memory and beliefs about juvenile delinquency across multiple time periods. Four hypotheses were proposed: (H1) vivid evidence is more memorable than nonvivid evidence, (H2) story evidence is more memorable than statistical evidence, (H3) vivid evidence is more persuasive than nonvivid evidence after 48 hours, but not after one week, and (H4) story evidence is more persuasive than statistical evidence after 1 week, but not after 48 hours. A 2 x 2 x 3 factorial design with an offset control was employed, using evidence (story or statistical), vividness (vivid or nonvivid), and time (immediate, or 48 hour delay, or 1 week delay) as independent variables, and recognition memory and judgment of belief as dependent variables. Four written messages, reflecting a complete crossing of evidence and vividness, were used as different types of evidence to attempt to persuade beliefs. A total of 280 undergraduate college students participated in the experiment. Hypotheses 1 and 2 were supported by main effects for vividness and evidence, and by a significant ordinal two-way interaction between vividness and evidence such that vivid story was the most memorable form of evidence. The two-way interactions used to test Hypotheses 3 and 4 were nonsignificant. A main effect for evidence related to Hypothesis 4 indicated that statistical evidence was more persuasive than story evidence at the delayed time periods. Thus, Hypotheses 3 and 4 were not supported. Alternative explanations were discussed to account for the persuasiveness of statistical evidence and the lack of persuasiveness of story evidence at the delayed time periods. Limitations of the study were noted, such as the small amount of experimental variance accounted for in some of the findings, and the limited generalizability of the findings. Finally, several suggestions for future research, including reconceptualizing evidence as a multidimensional construct, were presented.