Feeling torn and fearing rue: Attitude ambivalence and anticipated regret as antecedents of biased information seeking

The Listening Circle is a method for improving listening in organizations. It involves people sitting in a circle where only one talks at a time. Talking turns are signaled by a talking object. Although there are several reports regarding the effectiveness of the Listening Circle, most are based on case studies, or confounded with another intervention, and do not use theory to predict the listening-induced outcomes. We predicted that perceiving good listening decreases employees’ social anxiety, which allows them to engage in deeper introspection, as reflected by increased self-awareness. This increased self-awareness enables an acknowledgment of the pros and cons of various work-related attitudes and can lead to attitudes that are objectively more ambivalent and less extreme. Further, we hypothesized that experiencing good listening will enable speakers to accept their contradictions without the evaluative conflict usually associated with it (subjective-attitude ambivalence). In three quasi-experiments (Ns = 31, 66 and 83), we compared the effects of a Listening Circle workshop to a self-enhancement workshop (Studies 1 and 2), to a conflict management workshop (Study 2) and to employees who did not receive any training (Study 3), and found consistent support for the hypotheses. Our results suggest that the Listening Circle is an effective intervention that can benefit organizations.

The quality of listening in interpersonal contexts was hypothesized to improve a variety of work outcomes. However, research of this general hypothesis is dispersed across multiple disciplines and mostly atheoretical. We propose that perceived listening improves job performance through its efects on afect, cognition, and relationship quality. To test our theory, we conducted a registered systematic review and multiple meta-analyses, using three-level meta-analysis models, based on 664 efect sizes and 400,020 observations. Our results suggest a strong positive correlation between perceived listening and work outcomes, r = .39, 95%CI=[.36, .43], 휌 = .44, with the efect on relationship quality, r =.51, being stronger than the efect on performance, r =.36. These fndings partially support our theory, indicating that perceived listening may enhance job performance by improving relationship quality. However, 75% of the literature relied on self-reports raising concerns about discriminant validity. Despite this limitation, removing data solely based on self-reports still produced substantial estimates of the association between listening and work outcomes (e.g., listening and job performance, r = .21, 95%CI=[.13, .29], 휌 = .23). Our meta-analyses suggest further research into (a) the relationship between listening and job knowledge, (b) measures assessing poor listening behaviors, (c) the incremental validity of listening in predicting listeners’ and speakers’ job performance, and (d) listening as a means to improve relationships at work.