Positive and Negative Hypothesis Testing by Cooperative Groups

In a rule induction problem positive hypothesis tests select evidence that the tester expects to be an example of the correct rule if the hypothesis is correct, whereas negative hypothesis tests select evidence that the tester expects to be a nonexample if the hypothesis is correct. Previous research indicates the general effectiveness of a positive test strategy for individuals, but there has been very little research with cooperative groups. We extend the analysis of Klayman and Ha (Psychological Review, 1987) of ambiguous verification or conclusive falsification of five possible types of hypotheses by positive and negative tests by emphasizing the importance of further examples following hypothesis tests. In two experiments four-person cooperative groups solved rule induction problems by proposing a hypothesis and selecting evidence to test the hypothesis on each of four arrays on each trial. In different conditions the groups were instructed to use different combinations of positive and negative tests on the four arrays. Positive tests were more likely to lead to further examples than negative tests, and the proportion of correct hypotheses corresponded to the proportion of positive tests, in both experiments. We suggest that positive tests are more effective than negative hypothesis tests in generating further evidence, and thus in inducing the correct rule, in experimental rule induction tasks with a criterion of certainty imposed by the researcher.     

Effectiveness of Positive Hypothesis Testing for Cooperative Groups

In a rule induction problem positive hypothesis tests select evidence that the tester expects to be an example of the correct rule if the hypothesis is correct, whereas negative hypothesis tests select evidence that the tester expects to be a nonexample if the hypothesis is correct. We extend previous analyses of the effectiveness of positive and negative tests for ambiguous verification or conclusive falsification of hypotheses by emphasizing the importance of examples following positive or negative tests. Cooperative four-person groups solved rule induction problems from a single known example of the correct rule by proposing hypotheses and selecting evidence on each of four arrays on a series of trials. There were more examples following positive tests than negative tests. The transition probability from an incorrect hypothesis on trialtto the correct hypothesis on trialt+ 1 was higher for positive tests than for negative tests, higher for positive tests followed by examples than positive tests followed by nonexamples, and higher for negative tests followed by examples than negative tests followed by nonexamples. Once the group proposed the correct hypothesis on trialtthey were highly likely to continue to propose the correct hypothesis on trialt+ 1.     

Effectiveness of Positive Hypothesis Testing in Inductive and Deductive Rule Learning

In a positive hypothesis test a person generates or examines evidence that is expected to have the property of interest if the hypothesis is correct, whereas in a negative hypothesis test a person generates or examines evidence that is not expected to have the property of interest if the hypothesis is correct. Two experiments assessed the effectiveness of positive versus negative hypothesis tests on inductive and deductive rule learning problems. In Experiment 1 problem solvers induced a rule by proposing hypotheses and selecting evidence in the eight conditions of a factorial design defined by instructions to use a positive or negative hypothesis test on each of trials 1-5, 6-10, and 11-15. Instructions to use positive tests resulted in more examples, fewer strategic hypotheses, and a higher weighted score for five types of hypotheses than instructions to use negative tests. In Experiment 2 problem solvers identified 1 of a possible 1296 correct rules in the deductive rule learning game Mastermind. When problems were classified in the 16 possible combinations of positive or negative hypothesis tests on trials 2, 3, 4, and 5 there were fewer trials to solution for positive tests on each of the four trials and fewer trials to solution with increasing positive tests. We conclude that positive hypothesis tests are generally more effective than negative hypothesis tests in both inductive and deductive rule learning. Copyright 1999 Academic Press.     

The Influence of Positive Affect on the Decision Rule in Risk Situations: Focus on Outcome (and Especially Avoidance of Loss) Rather Than Probability

Two experiments examined the influence of positive affect on probability estimation and choice. Participants in whom positive affect had been induced, as well as no-manipulation controls, were asked to make both numerical evaluations of verbal probabilities in three-outcome gambles and actual betting decisions about similar gambles. Results from Experiment 1 showed the phenomenon labeledcautious optimism:Positive affect participants significantly overestimated the probabilities associated with phrases for winning relative to their estimates of probability of losing for the same phrases (optimism), while participants in a control condition did not; yet, in actual gambling situations, affect condition participants were much less likely to gamble than were controls when a real loss was possible (caution). Results of the betting task from Experiment 2 further indicated that affect participants used a betting-decision rule that was different from that of controls: They bet less than controls in gambles where potential losses were large, even though probability of loss was small, and they bet more than controls in gambles where the amount of the potential loss was small, even though the probability of loss was moderate or large. These findings suggest that positive affect can promote an overt shift from a decision rule focusing primarily on probabilities to one focusing on utilities or outcome values, especially for losses. Taken together, the results are compatible with an interpretation of the influence of positive affect in terms of an elaboration of positive cognitive material, and purposive behavior in decisions, rather than in terms of mere response bias.     

Context affects the interpretation of low but not high numerical probabilities: A hypothesis testing account of subjective probability

Low numerical probabilities tend to be directionally ambiguous, meaning they can be interpreted either positively, suggesting the occurrence of the target event, or negatively, suggesting its non-occurrence. High numerical probabilities, however, are typically interpreted positively. We argue that the greater directional ambiguity of low numerical probabilities may make them more susceptible than high probabilities to contextual influences. Results from five experiments supported this premise, with perceived base rate affecting the interpretation of an event’s numerical posterior probability more when it was low than high. The effect is consistent with a confirmatory hypothesis testing process, with the relevant perceived base rate suggesting the directional hypothesis which people then test in a confirmatory manner.     

Social discounting rate is negatively correlated with fluid intelligence

The purpose of the study was to verify a hypothesis, inspired by the handicap principle, of a positive relationship between subjective value of a hypothetical monetary reward shared with others and the level of fluid intelligence. Manipulation involved the amount of reward to be shared (small vs. large amount) and subject’s relationship to recipients (related vs. unrelated). As expected, a positive correlation was found between the subjective value of a reward to be shared with others, measured as the area under the curve for the discounting function and Raven Advanced Progressive Matrices scores, but the relationship was only present for rewards shared with relatives. In addition, participants who made altruistic choices in all items scored higher in RPM than those who were not as consistent. The implications of results for the evolutionary interpretation of the relationship between intelligence and altruism are discussed.     

Comparison of the distortion of probability information in decision under risk and an equivalent visual task

Decision makers typically overweight small probabilities and underweight large probabilities. However, there are recent reports that when probability is presented in the form of relative frequencies, this typical pattern reverses. We tested this hypothesis by comparing decision making in two tasks: In one task, probability was stated numerically, and in the other task, it was conveyed through a visual representation. In the visual task, participants chose whether a "stochastic bullet" should be fired at either a large target for a small reward or a small target for a large reward. Participants' knowledge of probability in the visual task was the result of extensive practice firing bullets at targets. In the classical numerical task, participants chose between pairs of lotteries with probabilities and rewards matched to the probabilities and rewards in the visual task. We found that participants' probability-weighting functions were significantly different in the two tasks, but the pattern for the visual task was the typical, not the reversed, pattern.     

The frequentist implications of optional stopping on Bayesian hypothesis tests

Null hypothesis significance testing (NHST) is the most commonly used statistical methodology in psychology. The probability of achieving a value as extreme or more extreme than the statistic obtained from the data is evaluated, and if it is low enough, the null hypothesis is rejected. However, because common experimental practice often clashes with the assumptions underlying NHST, these calculated probabilities are often incorrect. Most commonly, experimenters use tests that assume that sample sizes are fixed in advance of data collection but then use the data to determine when to stop; in the limit, experimenters can use data monitoring to guarantee that the null hypothesis will be rejected. Bayesian hypothesis testing (BHT) provides a solution to these ills because the stopping rule used is irrelevant to the calculation of a Bayes factor. In addition, there are strong mathematical guarantees on the frequentist properties of BHT that are comforting for researchers concerned that stopping rules could influence the Bayes factors produced. Here, we show that these guaranteed bounds have limited scope and often do not apply in psychological research. Specifically, we quantitatively demonstrate the impact of optional stopping on the resulting Bayes factors in two common situations: (1) when the truth is a combination of the hypotheses, such as in a heterogeneous population, and (2) when a hypothesis is composite—taking multiple parameter values—such as the alternative hypothesis in a t-test. We found that, for these situations, while the Bayesian interpretation remains correct regardless of the stopping rule used, the choice of stopping rule can, in some situations, greatly increase the chance of experimenters finding evidence in the direction they desire. We suggest ways to control these frequentist implications of stopping rules on BHT.     

Social Judgment of Abortion: A Black-Sheep Effect in a Catholic Sheepfold

Abstract

French Catholic participants (N = 340) with high or low religious identification read 1 of 8 scenarios presented as an interview with a female target 2 months after she had had an abortion. The experimental device varied situational pressure (pressure vs. no pressure), the target's religious social identity (Catholic vs. neutral), and the consequences of abortion for the target (positive vs. negative). The participants then rated the acceptability of the target's decision. The participants judged abortion more negatively in the no-pressure condition. Moreover, the participants with high religious identification judged abortion more negatively than did those with low religious identification. In partial support of a black-sheep effect, the participants with high religious identification judged the Catholic target more negatively than they judged the neutral one in some conditions (pressure, negative consequences). In other conditions (no pressure, both positive and negative consequences), the participants with low religious identification judged the Catholic target more positively than they judged the neutral one.     

Riskier for me or for others? The role of domain and probability in self-other differences,in risky decision-making

Abstract

In social and economic interactions, people often decide differently for others, as against for themselves, under situations involving risks. This sometimes leads to conflicts or contradictions. Although previous studies have explored such contradictions, the findings have been inconsistent. To reconcile these inconsistencies, this paper investigates the role played by the different domains and probabilities in the self-other differences under risk. Two groups of participants completed a gambling task combining different domains (gain vs. loss) and probabilities (small vs. large). One group made decisions for others and the other group made decisions for themselves. The results revealed a four pattern of discrepancy: the ones who made decisions for others were less risk-seeking than those who made decisions for themselves over the small probability gains. This was reversed over the large probability gains. Conversely, the participants who made decisions for others were more risk-seeking than those who made decisions for themselves over the small probability losses. The results were reversed over the large probability losses. These results reconcile the contradictory findings of the previous studies and suggest the significant role played by contextual factors in such discrepancies.     

Claiming best or better? The effect of target brand's and competitor's puffery on holistic and analytic thinkers

Previous empirical studies have yielded contradictory results about how consumers react to puffed claims in advertisements. This study addresses this issue by considering how consumers' thinking style and competitors' puffery interact to influence consumers' brand attitude in terms of product puffery. Drawing upon experiments using fictitious and real brand names, three studies provide converging evidence that holistic thinkers will form a more positive brand attitude when exposed to the target brand's low‐puffery (vs. no puffery or high puffery) claims. In contrast, analytic thinkers are less sensitive to puffery, and their brand attitude will not change. Further, holistic thinkers are more sensitive to the presence of competitor's puffery. Holistic thinkers exposed to competitor's high‐puffery (vs. low) claims form a more positive brand attitude toward the target brand. For analytic thinkers, competitor's puffery level will not significantly affect their attitude toward the target brand. Our findings shed fresh light on the inconclusive results of prior studies and offer practical implications for marketing puffery.     

Searching for Cognitive Processes in the Cultivation Effect Adult and Adolescent Samples in the United States and Australia

Research on the cultivation hypothesis has focused on whether relationships between television viewing and social reality beliefs truly exist or are artifacts. There is very little evidence about what cognitive processes allow viewers to construct television-biased beliefs. The present study tests two possible processes: First, that perceptions of the television world serve as an intermediate step between fragmented incidental learning from television and beliefs about the real world and second, that beliefs closely linked to television content are an intermediate step in implying more general values and beliefs. These two hypotheses were tested in one adult and three adolescent samples, two in the United States and two in Australia. Across a range of cultivation questions, the basic cultivation result generally replicated that heavy viewers had beliefs about the world that appeared influenced by television. However, neither process hypothesis was supported. Although the null findings on the first hypothesis do not rule out construction from learned fragments, findings on the second hypothesis contradict cultivation's second-order effect. “Close” beliefs and their implied counterparts were unrelated, and cultivation relationships for these implied variables occurred only for those with real-world biased “dose” beliefs. Exploration of this result demonstrated that cultivation of both kinds of beliefs occurred more of ten for adolescents with high academic skills, suggesting that if cultivation occurs, it is a more active and intellectually demanding process than previously proposed.     

善因营销的捐赠水平与消费者态度:一个有中介的调节效应模型检验

通过两个实验,探讨了善因营销的捐赠水平对消费者态度的影响,尤其是考察了道德提升感的中介作用以及产品-公益事业拟合度的调节作用。研究发现,善因营销的捐赠水平对消费者态度有显著的正向影响,且道德提升感在捐赠水平与消费者态度之间起中介作用。与此同时,产品-公益事业拟合度显著地调节了捐赠水平与道德提升感的关系,且该调节关系通过道德提升感的中介作用来进一步影响消费者态度。     

Verbal probabilities: An alternative approach

Previous studies of verbal probabilities have tried to place expressions like a chance, possible, and certain on 0–1 numerical probability scales. We ask instead, out of a range of outcomes, which outcome a verbal probability suggests. When, for instance, a sample of laptop batteries lasts from 1.5 to 3.5 hours, what is a certain and what is a possible duration? Experiment 1 showed that speakers associate certain with low values and possible with (unlikely) high or maximal values. In Experiment 2, this methodology was applied to several positive and negative verbal probability phrases, showing a preference for high rather than low or middle values in a distribution. Experiment 3 showed that such maxima are not universally described by large numbers. For instance, maximum speed is often described in terms of a small number of time units. What can (possibly) happen is accordingly sometimes described with very low and sometimes with very high values, depending upon focus of interest. Finally, participants in Experiment 4 were given the role of hearers rather than speakers and were asked to infer outcome ranges from verbal probabilities. Hearers appeared to be partly aware of speakers' tendencies to describe outcomes at the top of the range.     

Thinking styles and risky decision‐making: Further exploration of the affect–probability weighting link

People usually overweight small probabilities and underweight large probabilities leading to the familiar inverse S‐shaped weighting function. This research explores the link between affect and the structure of probability weighting from the perspective of thinking dispositions, a concept central to dual system theories of reasoning. The effects of affective priming and cognitive load on both probability weighting and the value function are also examined. The evidence suggests that thinking styles do have predictive implications for risky decision‐making. Participants with a more affective thinking style tend to be more risk‐seeking in small probability gambles. However, increasing access to the affective system by affective priming or cognitive load manipulations tend to reduce risk‐seeking behavior in small probability gambles as well as reduce risk averse behavior in large probability gambles. Previous research, manipulating the affective nature of lottery outcomes, found evidence for an increase in curvature (more overweighting of small probabilities and more underweighting of large probabilities) of the weighting function for affect‐rich outcomes, lending support to a hope‐and‐fear deconstruction of probability weighting. The present research suggests that increased anticipatory emotions characterized by the elevation of the weighting function (more overweighting at all probabilities) is also important and could sometimes be more significant than hope‐and‐fear in decision‐making under risk. An integrated approach incorporating the impact of affect on all three, the elevation and curvature of probability weighting as well as the curvature of the value function explains the empirical findings. Copyright © 2010 John Wiley & Sons, Ltd.     

Exact procedures for the analysis of multidimensional contingency tables

The log-linear model for contingency tables expresses the logarithm of a cell frequency as an additive function of main effects, interactions, etc., in a way formally identical with an analysis of variance model. Exact statistical tests are developed to test hypotheses that specific effects or sets of effects are zero, yielding procedures for exploring relationships among qualitative variables which are suitable for small samples. The tests are analogous to Fisher's exact test for a 2 × 2 contingency table. Given a hypothesis, the exact probability of the obtained table is determined, conditional on fixed marginals or other functions of the cell frequencies. The sum of the probabilities of the obtained table and of all less probable ones is the exact probability to be considered in testing the null hypothesis. Procedures for obtaining exact probabilities are explained in detail, with examples given.     

Assessing changes in performance and monitoring processes in individual and collaborative tests according to students' metacognitive skills

This study assessed students' test performances and metacognitive processes in real classroom settings. Psychology undergraduates were categorised according to their metacognitive skills (high vs. low) and had their test performances and monitoring processes in two different types of tests (multiple-choice and short-answer tests) compared in individual and collaborative test conditions. Students' test preparation practices, attributions, and regulatory processes during test-taking were also compared by using open-ended questions. In the collaborative tests, three types of metacognitive pairings were made (high/high vs. high/low vs. low/low). Results showed that: (1) in individual tests, high-metacognitive students presented better performance and higher confidence levels due to their more effective test preparation practices and regulatory skills; (2) Differences in performance and confidence levels due to metacognitive skills disappear when students take tests collaboratively; (3) Over time, collaborative testing had particular positive effects on the low/low metacognitive pairings. Results are discussed focusing on their educational implications.     

Mood and judgments based on sequential sampling

Different adaptive styles characterize cognition and behavior in different affective states. Whereas negative affect supports accommodation (i.e., stimulus‐driven bottom‐up processing), positive affect supports assimilation (i.e., self‐determined top‐down processing). Applying this well‐established rule to binary choices after self‐truncated information sampling, we predicted that positive mood should render choices less dependent on large samples than negative mood. Consequently, the potential primacy advantage underlying Wald's ( 1947 ) sequential testing (i.e., quick and correct decisions from the first few items in a sample) was exploited more efficiently when participants were in positive rather than negative mood. This efficient utilization of small samples in positive mood was obtained under the very conditions derived on a priori ground from a statistical model, namely, when a response criterion or threshold was high and when the true difference between choice options was relatively small. Copyright © 2009 John Wiley & Sons, Ltd.     

Phase attraction in sensorimotor synchronization with auditory sequences: effects of single and periodic distractors on synchronization accuracy

Four experiments showed that both single and periodic distractor tones affected the timing of finger taps produced in synchrony with an isochronous auditory target sequence. Single distractors had only small effects, but periodic distractors occurring at various fixed or changing phase relationships exerted strong phase attraction. The attraction was asymmetric, being stronger when distractors preceded target tones than when they lagged behind. A large pitch difference between target and distractor tones (20 vs. 3 semitones) did not reduce phase attraction substantially, although in the case of continuously changing phase relationships it did prevent complete capture of the taps by the distractors. The results support the hypothesis that phase attraction is an automatic process that is sensitive primarily to event onsets.