Decision Quality as a Function of the Amount of Information: Small Sample versus Large Sample Advantages in individuals and groups
Final Report Abstract
Taking advice from others offers individuals an effective and efficient strategy to improve judgments under uncertainty. Specifically, averaging one’s own judgment with that of another person cancels out random error and opposing biases. In the present project, we were concerned with the ecological and social boundaries of the success of advice taking. Specifically, we investigated two major modules of social rationality, the active seeking of advice and the discounting of invalid advice. Many findings in the advice taking literature illustrate an underweighting of advice compared to the optimal strategy, which usually consists in equal weighting of the own estimate and advice. Consequently, individuals generally do not exploit advice optimally. It is often assumed that individuals have access to reasons underlying their own judgment, but not to reasons underlying another person’s judgment. As a result, individuals are more confident in their own judgments, which leads to stronger weighting of the own judgment in comparison to the advice. Thus, it appears that the information asymmetry in advice taking situations precludes optimal accuracy gains. However, conventional advice taking research has neglected an inherent feature of interactive decision making, namely the opportunity to collect additional information. In the conventional paradigm, participants are presented with a single piece of advice only, before given an opportunity to revise their initial judgments. In the real world, however, advice taking will often constitute a sequential process in which the individual consecutively consults different advisors. In a paradigm that allows participants to sample as much advice as they desire, we found that participants generally sample several pieces of advice, that they sample more advice the less confident they are in their own judgment, and that they weigh advice more when revising their judgment the more evidence they gathered in favor of an alternative perspective. Thereby, this line of research suggests that advice taking is characterized by large degrees of rationality. However, advice takers must not only manage to optimally combine judgments provided by unreliable, but generally valid information sources. They must critically assess the validity of advice and dismiss potentially biased, invalid, or deceptive advice. Individuals must be able to integrate only advice that can be trusted. In a series of experiments, we demonstrated myopic behavior towards invalid advice. Whether samples were logically appropriate or seriously biased, subsequent judgments were strongly affected by the advice. The uncritical reliance on any advice persisted when participants were sensitized to the contrast of valid and invalid advice, when participants themselves did not believe an advice to be valid, and even after full debriefing about invalid advice. Invalid advice from lay people exerted a similar influence as expert advice. Consequently, rather than being insensitive to advice due to egocentric adherence to their own opinions, judges are often over-sensitive to whatever information happens to be offered by advisors, whose validity and whose information sources are hardly tested critically. We demonstrate on the one hand, that naïve judges exert high degrees of rationality in sampling and integrating valid advice. On the other hand, naïve judges can be heavily influenced by invalid advice. Our results call for a comprehensive theoretical account of advice taking.
Publications
- (2015). The social psychology of politics: Homo politicus revisited. In J. P. Forgas, K. Fiedler, & W. D. Crano (Eds.), Social psychology and politics (pp. 1-17). New York, NY, US: Psychology Press
Forgas, J. P., Fiedler, K., & Crano, W. D.
(See online at https://doi.org/10.4324/9781315717104-6) - (2015). The tragedy of democratic decision making. In J. P. Forgas, K. Fiedler, & W. D. Crano (Eds.), Social psychology and politics (pp. 193-208). New York, NY, US: Psychology Press
Fiedler, K., Hofferbert, J., Woellert, F., Krüger, T., & Koch, A.
(See online at https://doi.org/10.4324/9781315717104-20) - (2016). Heuristics and biases: Beyond Tversky and Kahneman’s (1974) judgment under uncertainty. In M. W. Eysenck & D. Groome (Eds.), Cognitive psychology: Revisiting the classic studies (pp. 146-161). London, UK: Sage
Fiedler, K., & von Sydow, M.
- (2016). Seeking advice: A sampling approach to advice taking. Judgment and Decision Making, 11(4), 401-415
Hütter, M., & Ache, F.
(See online at https://doi.org/10.15496/publikation-13917)
