The Fundamental Attribution Error in Attributing Fictional Figures' Characteristics to the Actors

Abstract

Two studies attempted to document the occurrence of the psychological phenomenon known as the fundamental attribution error (FAE) in the audiovisual medium. The FAE refers to the human tendency to attribute people's behavior to internal traits more than to external factors. In Study 1, we demonstrated that in the audiovisual medium, viewers tend to attribute an actor's behavior in television dramas to the actor's personality, ignoring the existence of a script dictating the actor's behavior. Study 2 replicated this finding and also demonstrated that the tendency to make the FAE is related to the degree to which the person reports being transported into the narrative of the television drama. Furthermore, we showed that the tendency to attribute character traits to the actor is not diminished following exposure to the same actor playing 2 opposing roles. The last scene viewed was found to determine the evaluation of the actor's characteristics.

Notes

¹From a May 2002 interview on an Israeli talk show, Good Evening with Guy Pines, concerning the Israeli Channel 3 soap opera Lechayey Haahava.

²This study was approved by the Israeli Ministry of Education. Their committee made the decision that no parental approval was needed due to the nature of the experiments.

³In a pilot test (n = 15), the mean evaluation of the actor based on the information given about him, and without watching him act, was 3.96 on a scale ranging from 1 to 7. A t-test confirmed that this mean was not significantly different than 4 (the neutral value), t(13) = −0.418, p > .1

⁴One of the series is called Tironut (Basic Training). This series was produced by Uri and Benni Barbash and was broadcast in 1998 on Israel's Channel 2. The other series is called Cnafaim (Wings). This series was produced by Shmuel Imberman and was broadcast in 2000 on Israel's Channel One.

⁵The differential alphas on the positive and negative character's dependant variables might be accounted for by the differential perception of negative and positive information. As is claimed in the general discussion, people give more weight to negative information in impression formation (Kunda, 1999). As is well documented, extreme traits are known to result in a halo effect (e.g., Asch, 1946). In sum, an actor who is perceived negatively may be viewed as thoroughly evil, whereas an actor who makes a positive impression may be viewed as more multidimensional.

⁶The effect size conventions for eta-squared values are: small (0.01), medium (0.06), and large (0.14; Evans, Wei, & Spyridakis, 2004). Hence, in Study 1 there is a large effect, and in Study 2 there are small and medium effects.

⁷In accordance with these results, we also included in the analyses the five participants who did not mention their gender. However, we also conducted the analyses when those participants were excluded. This did not change the results.