Reduction of pairwise comparisons in decision making via a duality approach

Although pairwise comparisons have been seen by many as an effective and intuitive way for eliciting qualitative data for multi‐criteria decision making problems, a major drawback is that the number of the required comparisons increases quadratically with the number of the entities to be compared. Thus, often even data for medium size decision problems may be impractical to be elicited via pairwise comparisons. The more the comparisons are, the higher is the likelihood that the decision maker will introduce erroneous data. This paper introduces a dual formulation to a given multi‐criteria decision making problem, which can significantly alleviate the previous problems. Some theoretical results establish that this is possible when the number of alternatives is greater than the number of decision criteria plus one. Copyright © 1999 John Wiley & Sons, Ltd.     

Rethinking value elicitation for personal consequential decisions

The application of multiattribute decision analysis to personal consequential decision problems can be unhelpful particularly where the analysis fails to help the decision maker to discover new values, or where the decision maker is unable to visualize the experience that may result from choosing a particular course of action. We concur with James March that values are best discovered by the act of choosing and experimenting but acknowledge that the act of choice can lead to irreversible consequences. Both of these problems suggest that techniques which assist the decision maker to envisage the consequences of choice need to be developed and incorporated into the modelling process. We suggest some forms that these techniques might take. Copyright © 1999 John Wiley & Sons, Ltd.     

Outcome effects: The impact of decision process and outcome controllability

An ‘outcome effect’ refers to the phenomenon whereby performance evaluations of decision makers are affected by the outcomes of those decisions. Although some consider such an effect to be a judgmental error, judgment by outcomes may not be dysfunctional when the evaluator does not know how the decision maker chose his or her action. In such situations, outcomes may provide some noisy information about decision quality. We test whether an outcome effect will still occur when the decision methodology and quality are more explicitly identified. Further, we test whether outcome controllability, a previously unexplored moderator variable, will have an impact on the outcome effect. Our first experiment, using undergraduates as subjects, shows that decision quality and controllability have an impact on performance evaluations but that the ubiquitous outcome effect still obtains. These results were replicated with experienced business people, except that controllability only affected their judgments in the case of negative outcomes. Implications of these results are discussed. © 1997 John Wiley & Sons, Ltd.     

Simulation of Preference Information in an Interactive Reference Point‐Based Method for the Bi‐Objective Inventory Routing Problem

In the research area of multiple criteria decision making, very few publications exist that explicitly design the simulation of a decision maker (DM) in an interactive approach. For this reason, we outline some methods widely used in the literature to identify common assumptions of simulating the DM's responses and the required input preference information. Our paper aims at covering the identified gap by introducing experimental concepts. Such concepts are used for theoretical analyses of a combined search‐and‐decision‐making procedure. Simulating the DM is a fruitful idea because the algorithm can be tested without integrating a human decision maker. Finally, we conduct experiments based on the proposed settings for a multiobjective inventory routing problem, which is a relevant and challenging logistic problem. Copyright © 2014 John Wiley & Sons, Ltd.     

Type of learning task impacts performance and strategy selection in decision making

In order to be adaptive, cognition requires knowledge about the statistical structure of the environment. We show that decision performance and the selection between cue-based and exemplar-based inference mechanisms can depend critically on how this knowledge is acquired. Two types of learning tasks are distinguished: learning by comparison, by which the decision maker learns which of two objects has a higher criterion value, and direct criterion learning, by which the decision maker learns an object's criterion value directly. In three experiments, participants were trained either with learning by comparison or with direct criterion learning and subsequently tested with paired-comparison, classification, and estimation tasks. Experiments 1 and 2 showed that although providing less information, learning by comparison led to better generalization (at test), both when generalizing to new objects and when the task format at test differed from the task format during training. Moreover, learning by comparison enabled participants to provide rather accurate continuous estimates. Computational modeling suggests that the advantage of learning by comparison is due to differences in strategy selection: whereas direct criterion learning fosters the reliance on exemplar processing, learning by comparison fosters cue-based mechanisms. The pattern in decision performance reversed when the task environment was changed from a linear (Experiments 1 and 2) to a nonlinear structure (Experiment 3), where direct criterion learning led to better decisions. Our results demonstrate the critical impact of learning conditions for the subsequent selection of decision strategies and highlight the key role of comparison processes in cognition.     

Robustness of Additive Value Function Methods in MCDM

A simulation model is constructed of choice between a discrete number of non-dominated alternatives. The long-run goals of the decision maker are assumed to be consistent with a hypothetical preference structure which satisfies assumptions of completeness, transitivity and additive independence for an ideal set of criteria. The use of additive value functions to aid the decision maker in this choice is simulated for a variety of contexts and under a number of non-idealities such as the omission of criteria, confounding of criteria and inconsistent responses. It is found that ideal preference orderings are well identified by additive value functions provided that the non-idealities are moderate and that sufficient effort is put into modelling changing marginal values for different levels of performance. One of the potentially most sensitive areas is that of shifts in the decision maker's reference points as a result of the types of preference information asked.     

R&D Project Selection: A Decision Process Approach

The fundamental decision problem of an R&D firm is to select projects in which to invest. Most reported models dealing with this subject use sophisticated mathematical models maximizing specific attributes of the projects without dealing with the underlying motivation of the decision maker. This paper summarizes a successful application of a decision-making process involving multicriteria in the selection of R&D projects in the Ecogen Israel Partnership. We emphasize the structure of the decision problem, with the main goal of helping the decision maker to better understand the nature of his problem. We systematically generate the objective hierarchy of the decision maker, including his main criteria, subcriteria attributes and alternatives. The ‘best’ alternative can then be found using most interactive procedures found in the literature. As the decision maker is totally involved in the entire decision-making process he is better able to understand his problem and preferences.     

Anchor effects in decision making can be reduced by the interaction between goal monitoring and the level of the decision maker’s executive functions

Models of decision making postulate that interactions between contextual conditions and characteristics of the decision maker determine decision-making performance. We tested this assumption by using a possible positive contextual influence (goals) and a possible negative contextual influence (anchor) in a risky decision-making task (Game of Dice Task, GDT). In this task, making advantageous choices is well known to be closely related to a specific decision maker variable: the individual level of executive functions. One hundred subjects played the GDT in one of four conditions: with self-set goal for final balance (n?=?25), with presentation of an anchor (a fictitious Top 10 list, showing high gains of other participants; n?=?25), with anchor and goal definition (n?=?25), and with neither anchor nor goal setting (n?=?25). Subjects in the conditions with anchor made more risky decisions irrespective of the negative feedback, but this anchor effect was influenced by goal monitoring and moderated by the level of the subjects' executive functions. The findings imply that impacts of situational influences on decision making as they frequently occur in real life depend upon the individual's cognitive abilities. Anchor effects can be overcome by subjects with good cognitive abilities.     

Towards Automating Decision Aiding Through Argumentation

Decision aiding can be abstractly described as the process of assisting a user/client/decision maker by recommending possible courses of his action. This process has to be able to cope with incomplete and/or inconsistent information and must adapt to the dynamics of the environment in which it is carried out. Indeed, on the one hand, complete information about the environment is almost impossible, and on the other hand, the information provided by the user is often affected by uncertainty; it may contains inconsistencies and may dynamically be revised because of various reasons. The aim of this paper is to present a model of the decision aiding process that is amenable to automation. The main features of the approach is that it models decision aiding as an iterative defeasible reasoning process, and it uses argumentation for capturing important aspects of the process. More specifically, argumentation is used for representing the relations between the cognitive artefacts that are involved in decision aiding, as well as for modelling the artefacts themselves. In modelling the cognitive artefacts, we make use of the notion of argument schemes and specify the related critical questions. More specifically, the work reported here aims at initiating a systematic study of the use of argumentation in future decision aiding tools. Our ambition is twofold: (i) enhance decision support capabilities of an analyst representing explicitly and accountably the reasons for which he recommend a solution for a decision maker and (ii) enhance decision support capabilities of an (semi) automatic device to handle (at least partially) the dialogue with the user. Copyright © 2011 John Wiley & Sons, Ltd.     

Justification pressure in risky decision making: search for risk defusing operators

Under justification pressure, the decision maker knows in advance that the decision has to be justified to somebody afterwards. The effect of justification pressure on the search for risk defusing operators (RDOs) and the role of RDOs in the justification texts were investigated. An RDO is an action intended by the decision maker to be performed in addition to an otherwise attractive alternative to decrease the risk. As predicted, in Experiment 1 participants (60 non-students) under justification pressure searched more RDOs. Additionally, in Experiment 2 (80 non-students) RDO search success was varied. Under justification pressure, persistence of RDO search was higher when no RDO could be detected. In the justification texts, the existence or non-existence of RDOs played a prominent role. Searching for RDOs supports people in their goal to make a good decision and in their attempt to convince the addressee of their justification that the decision was good.     

Adaptive learning and risk taking

Humans and animals learn from experience by reducing the probability of sampling alternatives with poor past outcomes. Using simulations, J. G. March (1996) illustrated how such adaptive sampling could lead to risk-averse as well as risk-seeking behavior. In this article, the author develops a formal theory of how adaptive sampling influences risk taking. He shows that a risk-neutral decision maker may learn to prefer a sure thing to an uncertain alternative with identical expected value and a symmetric distribution, even if the decision maker follows an optimal policy of learning. If the distribution of the uncertain alternative is negatively skewed, risk-seeking behavior can emerge. Consistent with recent experiments, the model implies that information about foregone payoffs increases risk taking.     

THE DISJUNCTION EFFECT IN CHOICE UNDER UNCERTAINTY

Abstract —One of the basic axioms of the rational theory of decision under uncertainty IS Savage's (1954) sure-thing principle (STP) It states that if prospect x is preferred to y knowing that Event A occurred, and if x IS preferred to y knowing that A did not occur, then x should be preferred to y even when it is not known whether A occurred We present examples in which the decision maker has good reasons for accepting x if A occurs, and different reasons for accepting X if A does not occur Not knowing whether or not A occurs, however, the decision maker may lack a clear reason for accepting X and may opt for another option We suggest that, in the presence of uncertainty, people are often reluctant to think through the implications of each outcome and, as a result, may violate STP This interpretation is supported by the observation that STP is satisfied when people are made aware of their preferences given each outcome     

Sequential decision making: An aggregate scheduling methodology

Recent studies of sequential decision making performance have utilized optimal dynamic programming solutions as the criteria for evaluating the quality of human decisions. The difficulties incumbent in increasing the dimensionality of dynamic programming formulations effectively prohibit their extension beyond the case of a single decision variable. In the following article an alternative optimization formulation for a two-decision-variable case is presented, followed by the development of a corresponding one-decision-variable problem. In addition, related methodological problems limiting the comparability of results among previous studies are presented. It is suggested that the proposed problem formulation and its accompanying interpretation for the decision maker largely eliminate these previous methodological problems. Furthermore, the problem formulation permits quantification of the relative weights that should optimally be placed on future decision periods in the multiple-stage decision problem.     

Evaluating Decision Makers: The Role of Choice Congruence and Outcome

Two studies were conducted in which decision makers were evaluated by subjects who had agreed or disagreed with the decision maker's choice. Subjects read one of two vignettes describing the alternatives available to the decision maker, indicated which alternative they personally favored, and then learned about the decision maker's choice and the outcome that occurred. Study 1 varied whether the outcomes of the decision maker's choice were positive or negative, and whether the subject's preferred option matched (congruent choice) or did not match (incongruent choice) that of the decision maker. Subjects rated the extent to which they thought the decision maker was worthy of praise (in the case of positive outcomes) or blame (in the case of negative outcomes), and the decision maker's likableness and competence. Results revealed a strong effect of congruence on attributions of praise and blame: More praise was ascribed to an agreeing decision maker and more blame to a disagreeing decision maker. The degree to which the decision maker was seen as likable was affected by congruence only, whereas perceived competence was influenced by both outcome and congruence. Study 2 addresses some methodological issues that were unresolved in Study 1 and replicated the results of the first study, using new stimulus materials and an expanded set of dependent measures.     

Copy if dissatisfied,innovate if not: contrasting egg-laying decision making in an insect

The use of conspecific cues as social information in decision making is widespread among animals; but, because this social information is indirect, it is error-prone. During resource acquisition, conspecific cues also indicate the presence of competitors; therefore, decision makers are expected to utilize direct information from resources and modify their responses to social information accordingly. Here, we show that, in a non-social insect, unattractive egg-laying resources alter the behavioural response to conspecific cues from avoidance to preference, leading to resource sharing. Females of the adzuki bean beetle Callosobruchus chinensis avoid laying eggs onto beans that already have conspecific eggs. However, when we provided females with bean-sized clean glass beads with and without conspecific eggs, the females preferred to add their eggs onto the beads with eggs. The glass beads, once coated with water extracts of adzuki beans, enabled the females to behave as if they were provided with the beans: the females preferred bean-odoured glass beads to clean glass beads and they avoided the substrates with eggs. When females are provided with unattractive egg-laying substrates only, joining behaviour (i.e. copying) might be advantageous, as it takes advantage of information about positive attributes of the substrate that the focal animal might have missed. Our results suggest that given only unsatisfactory options, the benefits of copying outweigh the costs of resource competition. Our study highlights the importance of integrating multiple information sources in animal decision making.     

Information about expert decision and post‐decision distortion of facts of own decision

Cognitive representations of decision problems are dynamic. During and after a decision, evaluations and representations of facts change to support the decision made by a decision maker her‐ or himself (Svenson, 2003). We investigated post‐decision distortion of facts (consolidation). Participants were given vignettes with facts about two terminally ill patients, only one of whom could be given lifesaving surgery. In Study 1, contrary to the prediction, the results showed that facts were distorted after a decision both by participants who were responsible for the decisions themselves and when doctors had made the decision. In Study 2 we investigated the influence of knowledge about expert decisions on a participant's own decision and post‐decisional distortion of facts. Facts were significantly more distorted when the participant's decision agreed with an expert's decision than when the participant and expert decisions disagreed. The findings imply that knowledge about experts' decisions can distort memories of facts and therefore may obstruct rational analyses of earlier decisions. This is particularly important when a decision made by a person, who is assumed to be an expert, makes a decision that is biased or wrong.     

On the shape of the probability weighting function

Empirical studies have shown that decision makers do not usually treat probabilities linearly. Instead, people tend to overweight small probabilities and underweight large probabilities. One way to model such distortions in decision making under risk is through a probability weighting function. We present a nonparametric estimation procedure for assessing the probability weighting function and value function at the level of the individual subject. The evidence in the domain of gains supports a two-parameter weighting function, where each parameter is given a psychological interpretation: one parameter measures how the decision maker discriminates probabilities, and the other parameter measures how attractive the decision maker views gambling. These findings are consistent with a growing body of empirical and theoretical work attempting to establish a psychological rationale for the probability weighting function.     

The Effects of Decision Outcome Dispersion Upon Organizational Decision Making

The effects of varying decision outcome dispersion on organizational decision making were investigated under individual and group decision making conditions. Thirty-six female and pg]36 male subjects made decisions for organizational decision scenarios in which outcomes affected primarily the decision maker, people other than the decision maker, or a group of which the decision maker was a member. Subjects rated their levels of perceived risk and confidence in their decisions and made decisions within a simulated context of either a small or a large organization. Results indicated that subjects perceived significantly less risk and more confidence in their decisions when outcomes affected primarily themselves rather than others regardless of whether the decisions were made individually or by a group. Males perceived their decisions as significantly more risky than females. Induced organizational size did not significantly influence decision making.

     

从风险决策中的多次博弈到单次博弈：量变还是质变?

诺贝尔经济学奖获得者Samuelson于1963年发现人们在单次和多次博弈条件下决策行为不一致。文章综述了两种博弈条件下人们决策行为的差异并质疑了这种差异的传统理论解释机制。描述或预测决策行为的风险决策理论其实只采用了一种评价法则—— 期望法则, 始终没有跳出“最大化”的窠臼。基于实验证据, 我们推测, 多次博弈时人们遵守了期望法则, 而单次博弈时人们所遵循的是非补偿性法则。从多次博弈到单次博弈, 不单单是一种博弈次数上的变化(量变), 而是代表了从期望法则(补偿性法则)到非补偿性法则两种策略之间的转变(质变)。最后, 文章介绍了单次、多次博弈问题在医疗、应急管理以及投资领域的体现, 并呼吁更多的研究者关注单次、多次博弈问题。