Present is greater than absent: influence of positive testing and feature positive effect on social judgment

Social judgments are subject to systematic tendencies as the confirmation bias. The present research (N = 362) aims at exploring the combined influence of positive testing (i.e. the tendency to explore features that are expected to be present) and of the Feature Positive Effect (FPE; whereby the presence of features outweighs the absence of features) when people are gathering new data or evaluating diagnostic cues. In Study 1 and Study 2 participants were presented with a variant of Wason’s selection task (1968) concerning a criminal scenario (e.g "if the proof x is present, then Mr X is guilty") and asked to search information. Evidence on the presence or absence of clues was provided and data impact on the final judgment was assessed. The findings suggest that individuals are prone to FPE, irrespective of the hypothesis (focal vs. alternative) they are testing. In Study 3 and Study 4, instead of choosing the hypothesis testing strategy, participants were asked to evaluate the utility of all the possible queries and the impact of all diagnostic and non-diagnostic pieces of information when present or absent. Data show that people are reasonably more affected by diagnostic data than by non-diagnostic ones; however, diagnosticity is processed more accurately when cues are on the presence rather than on the absence. This tendency to neglect diagnostic absence, in conjunction with the positive testing, may lead to overestimation of the a-priori hypothesis and may prevent a reasonable falsification of our beliefs. Theoretical and practical implications of these results within and beyond the juridical context are discussed. Studies and participants: Four studies, total of 362 participants. Short Abstract: Four experiments showed that, during the social hypothesis-testing process, people are prone to positive testing and Feature Positive Effect; moreover individuals are affected by FPE when processing diagosticity. This tendency may lead to an unjustified overestimation o