The Automatic Evaluation Effect: Unconditional Automatic Attitude Activation with a Pronunciation Task

Abstract

Previous demonstrations of automatic attitude activation have used the same critical test of the presence of automaticity (Bargh, Chaiken, Govender, & Pratto, 1992; Chaiken & Bargh, 1993; Fazio, Sanbonmatsu, Powell, & Kardes, 1986). In this task, subjects first classify as quickly as they can each of a series of adjectives as to whether each adjective is good or bad in meaning. Subsequently, on each priming trial, a word corresponding to an attitude object briefly precedes the adjective target. Whether this attitude objectprimeautomatically activates its corresponding evaluation in memory is indicated by whether the presence of that prime word on a trial facilitates or interferes with the conscious and intended evaluation of the adjective target. Although there is some disagreement as to the generality of the automaticity effect (Chaiken & Bargh, 1993; Fazio, 1993), all relevant experiments have obtained it for at least some of the subject's attitudes. The present experiments investigated the role that the adjective evaluation task itself may play in producing the automaticity effect. If the effect is truly unconditional, requiring only the mere presence of the attitude object in the environment, it should not depend on the subject consciously thinking in terms of evaluation. Three experiments designed to remove all such evaluative features from the basic paradigm revealed that the automaticity effect does not depend on such strategic evaluative processing. In fact, consistent with theaffect primacyhypothesis (Murphy & Zajonc, 1993), removing conscious processing aspects from the paradigm serves to increase, rather than decrease, the generality of the automatic evaluation effect.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 505 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.