A Meta-Analysis of Priming Effects on Impression Formation Supporting a General Model of Informational Biases

Abstract

Priming researchers have long investigated how providing information about traits in one context can influence the impressions people form of social targets in another. The literature has demonstrated that this can have 3 different effects: Sometimes primes become incorporated in the impression of the target (assimilation), sometimes they are used as standards of comparison (anchoring), and sometimes they cause people to consciously alter their judgments (correction). In this article, we present meta-analyses of these 3 effects. The mean effect size was significant in each case, such that assimilation resulted in impressions biased toward the primes, whereas anchoring and correction resulted in impressions biased away from the primes. Additionally, moderator analyses uncovered a number of variables that influence the strength of these effects, such as applicability, processing capacity, and the type of response measure. Based on these results, we propose a general model of how irrelevant information can bias judgments, detailing when and why assimilation and contrast effects result from default and corrective processes.

Enjoyment intention to use and actual use of a conversational robot by elderly people

“…In analyzing the replies to these statements, we used Cronbach's alpha to test the reliability of the constructs. In psychology, an alpha of 0.7 and higher is considered acceptable [30]. As table 2 shows, the constructs were highly reliable.…”

On the adaptive flexibility of evaluative priming

“…Spreading activation in semantic memory (Collins & Loftus, 1975;DeCoster & Claypool, 2004;De Houwer & Randell, 2004;Ross & Bower, 1981) strictly predicts congruity. It is hardly possible to invent an associative network architecture that predicts incongruity.…”

Evidence for an implicit influence of memory on future thinking

“…There was one between-subjects variable, with three levels (priming condition: social group, academic group, control group), in this experiment. Notably, a between-subjects design was instantiated (in both experiments) based on well-documented evidence from the social psychological literature that significant effects of priming are difficult to obtain when multiple concepts are primed in a single experimental session (DeCoster & Claypool, 2004;Förster & Liberman, 2007).…”