The effect of question order on evaluations of test performance: Can the bias dissolve?

Question difficulty order has been shown to affect students’ global postdictions of test performance. We attempted to eliminate the bias by letting participants experience the question order manipulation multiple times. In all three experiments, participants answered general knowledge questions and self-evaluated their performance. In Experiment 1, participants studied questions and answers in easy–hard or hard–easy question order prior to taking a test in the same order. In Experiment 2, participants took the same test twice in the opposite question order (easy–hard then hard–easy, or hard–easy then easy–hard). In Experiment 3, participants took two different tests in the opposite question order (easy–hard then hard–easy, or hard–easy then easy–hard). In all three experiments, we were unable to eliminate the bias, which suggests that repeated exposure is insufficient to overcome a strong initial anchor.     

Path dependent preferences: The role of early experience and biased search in preference development

What is the role of early experiences in shaping preferences? What are the mechanisms by which such early encounters influence the way preferences are formed? In this research, we examine the impact of the entry position and favorability of initial (and ongoing) experiences on preference development. We predict that the starting point will heavily influence which particular region people select from initially, and favorableness of early experiences and myopic search will both limit their search to that particular region. Across four studies, we find that when the initial experiences are favorable, subjects engage in lower levels of search, experience only a narrow breadth of possible alternatives, demonstrate less ongoing experimentation, and have a reduction in the amount of preference development.     

Origins of the stuttering stereotype: stereotype formation through anchoring-adjustment

The stereotype of people who stutter is predominantly negative, holding that stutterers are excessively nervous, anxious, and reserved. The anchoring–adjustment hypothesis suggests that the stereotype of stuttering arises from a process of first anchoring the stereotype in personal feelings during times of normal speech disfluency, and then adjusting based on a rapid heuristic judgment. The current research sought to test this hypothesis, elaborating on previous research by [White, P. A., & Collins, S. R. (1984). Stereotype formation by inference: A possible explanation for the “stutterer” stereotype. Journal of Speech and Hearing Research, 27, 567–570]. Participants provided ratings of a hypothetical typical person who stutters, a person suffering from normal speech disfluency and a typical male on a 25-item semantic differential scale. Results showed a stereotype of people who stutter similar to that found in previous research. The pattern of results is consistent with the anchoring–adjustment hypothesis. Ratings of a male stutterer are very similar to a male experiencing temporary disfluency, both of which differ from ratings of a typical male. As expected, ratings of a stutterer show a small but statistically significant adjustment on several traits that makes the stereotype of stutterers less negative and less emotionally extreme than the temporarily disfluent male. Based on the results of this research, it appears that stereotype formation is a result of generalization and adjustment from personal experience during normal speech disfluency.

Educational objectives: The reader will be able to: (1) explain how the negative stereotype of people who stutter arises; (2) discuss the negative implications of stereotypes in the lives of people who stutter; and (3) summarize why the stereotype of people who stutter is so consistent and resistant to change.     

Associations between self-rated health and the assessments of anchoring vignettes in cardiovascular patients

Background/ObjectiveAssessments of health can be biased by response shift effects. One method for detecting such effects is the use of anchoring vignettes. The aim of this study was to analyze the relationship between participants’ self-assessed health state and their assessments of these vignettes.MethodA total of 342 cardiovascular patients assessed their own state of health on a 0-100 visual analogue scale. The patients additionally assessed two vignettes featuring fictional persons suffering from specific complaints. A sample of the general population (N = 1,236) served as controls.ResultsThe participants rated the health state of the vignette character featuring physical problems as being significantly better than the general population did (effect size: d = 0.53). The group difference in the assessment of the vignette featuring primarily mental health problems was lower (d = -0.17). Participants’ assessments of the vignettes were positively correlated with their assessments of their own health state (r = .26 and r = .10) and with several quality of life variables.ConclusionsAnchoring vignettes are a useful tool for detecting response shift effects.     

Uninformative Anchors Have Persistent Effects on Valuation Judgments

Anchoring effects influence a wide range of numeric judgments, including valuation judgments, such as willingness‐to‐pay (WTP). However, prior research has not established whether anchoring only temporarily distorts responses or exerts persistent influence on preferences. This article presents three incentive compatible experiments examining the long‐term effects of random anchoring on WTP. Study 1 evaluated the persistence of anchoring effects over long durations, and showed that the strength of the effect decayed but did not disappear completely even 8 weeks later. In Study 2, a random anchor significantly influenced WTP after one week, regardless of whether WTP was also elicited immediately following the anchoring procedure, showing that consistency motivations do not account for persistence of anchoring effects. Study 3 showed relatively low anchors resulted in more stable valuations, compared with participants who reported WTP with no anchoring procedure. Together with the pattern of decay over time in Study 1, this suggests that anchoring facilitates the “imprinting” of valuation judgments for later retrieval. These studies show that anchoring effects can lead to lasting changes in valuation judgments, providing the first demonstration of long‐term persistence of constructed preferences as a result of an uninformative and arbitrary manipulation.     

The limits of anchoring

Anchoring and adjustment is a pervasive bias in which decision makers are influenced by random or uninformative numbers or starting points. As a means of understanding this effect, we explore two limits on anchoring. In Experiments 1 and 2, implausibly extreme anchors had a proportionally smaller effect than anchors close to the expected value of the lotteries evaluated. In Experiments 2 and 3, anchoring occurred only if the anchor and preference judgment were expressed on the same scale. Incompatible anchors and response modes resulted in no anchoring bias. A confirmatory search mechanism is proposed to account for these results.     

Rating of perceived effort but relative to what? A comparison between imposed and self-selected anchors

PurposeCollecting reliable and valid rating of perceived effort (RPE) data requires properly anchoring the scales’ upper limits (i.e., the meaning of 10 on a 0–10 scale). Yet, despite their importance, anchoring procedures remain understudied and theoretically underdeveloped. Here we propose a new task-based anchoring procedure that distinguishes between imposed and self-selected anchors. In the former, researchers impose on participants a specific task as the anchor; in the latter, participants choose the most effortful task experienced or imaginable as the anchor. We compared the impact of these conceptually different anchoring procedures on RPE.MethodsTwenty-five resistance-trained participants (13 females) attended a familiarization and two randomized experimental sessions. In both experimental sessions, participants performed non-fatiguing and fatiguing isometric maximal voluntary contraction (MVC) protocols with the squat followed by the gripper or vice versa. After each MVC, participants reported their RPE on a 0–10 scale relative to an imposed anchor of the performed task (e.g., gripper MVCs anchored to a gripper MVC) or to a self-selected anchor.ResultsIn the non-fatiguing condition, imposed anchors yielded greater RPEs than self-selected anchors for both the squat [on average, 9.4 vs. 5.5; Δ(CI_95%) = 3.9 (3.2, 4.5)] and gripper [9.4 vs. 3.9; Δ = 5.5 (4.7, 6.3)]. Similar results were observed in the fatiguing condition for both the squat [9.7 vs. 6.9; Δ = 2.8 (2.1, 3.5)] and gripper [9.7 vs. 4.5; Δ = 5.2 (4.3, 5.9)].ConclusionsWe found large differences in RPE between the two anchors, independent of exercises and fatigue state. These findings provide a basis for further development and refinement of anchoring procedures and highlight the importance of selecting, justifying, and consistently applying the chosen anchors.     

Socially Constructed Technology

The concepts of omniscience and omnipotence are defined in 2 × 2 ordinal games, and implications for the optimal play of these games, when one player is omniscient or omnipotent and the other player is aware of his omniscience or omnipotence, are derived. Intuitively, omniscience allows a player to predict the strategy choice of an opponent in advance of play, and omnipotence allows a player, after initial strategy choices are made, to continue to move after the other player is forced to stop. Omniscience and its awareness by an opponent may hurt both players, but this problem can always be rectified if the other player is omniscient. This pathology can also be rectified if at least one of the two players is omnipotent, which can override the effects of omniscience. In some games, one player's omnipotence ‐ versus the other's ‐ helps him, whereas in other games the outcome induced does not depend on which player is omnipotent. Deducing whether a player is superior (omniscient or omnipotent) from the nature of his game playing alone raises several problems, however, suggesting the difficulty of devising tests for detecting superior ability in games.     

Elaboration and consequences of anchored estimates: An attitudinal perspective on numerical anchoring

In the standard numerical anchoring paradigm, the influence of externally provided anchors on judgment is typically explained as a result of elaborate thinking (i.e., confirmatory hypothesis testing that selectively activates anchor-consistent information in memory). In contrast, theories of attitude change suggest that the same judgments can result from relatively thoughtful or non-thoughtful processes, with more thoughtful processes resulting in judgments that last longer over time and better resist future attempts at change. Guided by an attitudinal approach to anchoring, four studies manipulated participants’ level of cognitive load to produce relatively high versus low levels of thinking. These studies show that, although anchoring can occur under both high and low thought conditions, anchoring based on a higher level of thinking involves greater use of judgment-relevant background knowledge, persists longer over time, is more resistant to subsequent attempts at social influence, and is less likely to result from direct numeric priming.