Dimension integration: Testing models without trade-offs

This paper tests a behavioral property called dimension integration. The test evaluates models, such as lexicographic semi-orders and the priority heuristic, which assume that a person uses only one dimension at a time. It provides a way to compare such models against those that assume a person combines information from different dimensions. The test allows one to test the hypothesis that different people use different lexicographic semi-orders with different threshold parameters. In addition, by use of a “true and error” model, it is possible to “correct” for unreliability of choice in order to estimate the proportions of participants who show different response patterns that can be classified as integrative or not integrative. An experiment with 260 participants was conducted in which people made choices between two-branch gambles. The aggregate results violate the priority heuristic and six lexicographic semi-orders. The data also refute the theory that people use a mixture of these lexicographic semi-orders. In addition, few individuals appear to show response patterns consistent with non-integrative models. Instead, they show that most individuals show patterns consistent with the hypothesis that they combine information between dimensions.     

Sub-semiorder: A model of multidimensional choice with preference intransitivity

The paper develops a model of choice called a sub-semiorder which is a generalization of Luce's semiorder to multidimensional choice. The same reason (imperfect discrimination) that gives rise to the intransitivity of indifference in a semiorder gives rise to the intransitivity of preference in a sub-semiorder. This provides a rational explanation of intransitivity of preference without resorting to the lexicographic semiorder of Tversky. It is shown that the “apparent underlying preference” of a sub-semiorder is transitive but unfortunately it is not complete. However, with a mild condition, there exists a maximal element.     

Multi-attribute,multi-alternative models of choice: Choice,reaction time,and process tracing

The first aim of this research is to compare computational models of multi-alternative, multi-attribute choice when attribute values are explicit. The choice predictions of utility (standard random utility & weighted valuation), heuristic (elimination-by-aspects, lexicographic, & maximum attribute value), and dynamic (multi-alternative decision field theory, MDFT, & a version of the multi-attribute linear ballistic accumulator, MLBA) models are contrasted on both preferential and risky choice data. Using both maximum likelihood and cross-validation fit measures on choice data, the utility and dynamic models are preferred over the heuristic models for risky choice, with a slight overall advantage for the MLBA for preferential choice. The response time predictions of these models (except the MDFT) are then tested. Although the MLBA accurately predicts response time distributions, it only weakly accounts for stimulus-level differences. The other models completely fail to account for stimulus-level differences. Process tracing measures, i.e., eye and mouse tracking, were also collected. None of the qualitative predictions of the models are completely supported by that data. These results suggest that the models may not appropriately represent the interaction of attention and preference formation. To overcome this potential shortcoming, the second aim of this research is to test preference-formation assumptions, independently of attention, by developing the models of attentional sampling (MAS) model family which incorporates the empirical gaze patterns into a sequential sampling framework. An MAS variant that includes attribute values, but only updates the currently viewed alternative and does not contrast values across alternatives, performs well in both experiments. Overall, the results support the dynamic models, but point to the need to incorporate a framework that more accurately reflects the relationship between attention and the preference-formation process.     

Preserved organization of semantic attributes in Alzheimer's disease

This study investigated whether Alzheimer's disease (AD) disrupts the basic organization of the semantic attributes of concepts. Young and normal older subjects and AD patients were presented with a target concept followed by a stimulus word and were to decide whether the stimulus was related to the target. On those trials where it was, the stimulus was either a high-, medium-, or low-dominance attribute of the target. The higher the normative dominance, the more important the attribute to concept meaning. In all 3 subject groups, decision time varied as a function of dominance. The higher the dominance, the faster the decision. Attribute dominance affected the performance of AD patients more than that of normal subjects. These results suggest that AD patients retain their knowledge of the relative importance that the different attributes of a concept have for concept meaning.     

Inferring latent class lexicographic rules from choice data

A lexicographic rule orders multi-attribute alternatives in the same way as a dictionary orders words. Although no utility function can represent lexicographic preference over continuous, real-valued attributes, a constrained linear model suffices for representing such preferences over discrete attributes. We present an algorithm for inferring lexicographic structures from choice data. The primary difficulty in using such data is that it is seldom possible to obtain sufficient information to estimate individual-level preference functions. Instead, one needs to pool the data across latent clusters of individuals. We propose a method that identifies latent clusters of subjects, and estimates a lexicographic rule for each cluster. We describe an application of the method using data collected by a manufacturer of television sets. We compare the predictions of the model with those obtained from a finite-mixture, multinomial-logit model.     

Lexicographic order preservation and stochastic dominance

Stochastic dominance concerns conditions on outcome probabilities that are necessary and sufficient for one act to be (strictly) preferred to another according to all preference relations that share certain properties, one of which customarily is an Archimedean property sufficient to entail existence of real-valued representations. We relax this assumption to permit linear lexicographic utility of finite and known dimensionality. In some situations, levels of the lexicographic hierarchy could correspond to explicit criteria or attributes. In our model, subjective probabilities emerge as matrix premultipliers of the outcome utility vectors. We thus obtain matrix probability generalizations of the familiar cumulative probability conditions for stochastic dominance.     

认知诊断中多分属性与二分属性的对比研究

摘 要：Karelitz（2004）和詹沛达等（2016）认为1个多分属性内部(Lk+1)个水平的关系相当于Lk个部分满足线型层级关系的二分属性。本研究的目的是通过比较多分属性模型和二分属性模型的判准率,从而验证多分属性和二分属性间是否存在以上关系。结果表明：当属性个数较少时,两个模型的模式判准率相当,随着属性个数增加,多分属性模型的模式判准率高于二分属性模型的模式判准率。结论：在一定程度上,多分属性和二分属性之间确实存在以上关系,但两者并非完全等价,二者间的差异随着属性个数增加更加明显。     

Preferences for multi-attributed alternatives: Traces, dominance, and numerical representations

This paper analyzes conjoint measurement models allowing for intransitive and/or incomplete preferences. This analysis is based on the study of marginal traces induced on coordinates by the preference relation and uses conditions guaranteeing that these marginal traces are complete.Within the framework of these models, we propose a simple axiomatic characterization of preference relations compatible with the notion of dominance. We show that all such relations have a nontrivial numerical representation.Our results allow us to establish useful connections between two lines of thought in the area of decision analysis with multiple attributes that have largely remained unrelated: the one based on conjoint measurement and the one emphasizing the idea of dominance.     

Risky choice with heuristics: reply to Birnbaum (2008), Johnson, Schulte-Mecklenbeck, and Willemsen (2008), and Rieger and Wang (2008)

E. Brandst?tter, G. Gigerenzer, and R. Hertwig (2006) showed that the priority heuristic matches or outperforms modifications of expected utility theory in predicting choice in 4 diverse problem sets. M. H. Birnbaum (2008) argued that sets exist in which the opposite is true. The authors agree--but stress that all choice strategies have regions of good and bad performance. The accuracy of various strategies systematically depends on choice difficulty, which the authors consider a triggering variable underlying strategy selection. Agreeing with E. J. Johnson, M. Schulte-Mecklenbeck, and M. C. Willemsen (2008) that process (not "as-if") models need to be formulated, the authors show how quantitative predictions can be derived and test them. Finally, they demonstrate that many of Birnbaum's and M. O. Rieger and M. Wang's (2008) case studies championing their preferred models involved biased tests in which the priority heuristic predicted data, whereas the parameterized models were fitted to the same data. The authors propose an adaptive toolbox approach of risky choice, according to which people first seek a no-conflict solution before resorting to conflict-resolving strategies such as the priority heuristic.     

题目属性向量平衡策略的认知诊断测验设计

为组卷制定的题目属性向量平衡（IAVB）策略强调试卷必须体现认知模型,并将题目属性向量而不是以单个属性作为考察单位。该策略克服严格属性平衡(AB)策略仅适用于独立结构的不足,且在每个题目考察属性个数（大致）相同的条件下,以模式判准率（PMR）为衡量标准,该策略优于非IAVB策略。特别地,若属性层级结构为独立结构时,IAVB策略最优,严格属性平衡策略次之,两种策略均未采用的,则最差。在题目属性数一定条件下,IAVB矩阵提高PMR更显著。     

Process models deserve process data: comment on Brandstätter, Gigerenzer, and Hertwig (2006)

Comments on the article by E. Brandst?tter, G. Gigerenzer, and R. Hertwig. Resolution of debates in cognition usually comes from the introduction of constraints in the form of new data about either the process or representation. Decision research, in contrast, has relied predominantly on testing models by examining their fit to choices. The authors examine a recently proposed choice strategy, the priority heuristic, which provides a novel account of how people make risky choices. The authors identify a number of properties that the priority heuristic should have as a process model and illustrate how they may be tested. The results, along with prior research, suggest that although the priority heuristic captures some variability in the attention paid to outcomes, it fails to account for major characteristics of the data, particularly the frequent transitions between outcomes and their probabilities. The article concludes with a discussion of the properties that should be captured by process models of risky choice and the role of process data in theory development.     

Choice,Rating, and Ranking: Framing Effects with Different Response Modes

In a typical risky choice framing task, people have to choose among two options, which are either positively or negatively framed. Choices in the two framing conditions are then compared. However, different preferences between the conditions can be due to changes in the evaluation of the single constituent options or due to specific processes triggered by the choice task. In order to clarify the source of the framing effect, we investigate the effect with different response modes: choice, rating, and ranking. The rating and ranking findings indicate that what is commonly called a risky choice framing effect is actually a framing effect that changes the evaluation of only the riskless option, although there is little or no effect on the risky option. According to these findings, risky choice framing might be construed as a process of attribute framing, which is independent of risk preference in choice contexts. Copyright © 2011 John Wiley & Sons, Ltd.     

Rough approximation of pairwise comparisons described by multi‐attribute stochastic dominance

Given a finite set A of actions evaluated by a set of attributes, preferential information is considered in the form of a pairwise comparison table including pairs of actions from subset B⊂A described by stochastic dominance relations on particular attributes and a total order on the decision attribute. Using a rough sets approach for the analysis of the subset of preference relations, a set of decision rules is obtained, and these are applied to a set A\B of potential actions. The rough sets approach of looking for the reduction of the set of attributes gives us the possibility of operating on a multi‐attribute stochastic dominance for a reduced number of attributes. Copyright © 1999 John Wiley & Sons, Ltd.     

How context shapes category inferences and attribute preference for new ambiguous products

Extant research suggests that when marketers introduce products with functions that span multiple categories, consumers tend to generate beliefs in line with only a single category. This has been regarded as a major marketing challenge because it leads consumers to ignore key attributes from the product's supplementary category. Contrary to this prediction, the authors find that because consumers tend to classify new hybrid products by contrasting them against the competitive context, attributes from the supplementary category become more salient and thus contribute greater utility in choice. The authors pit the strength of this effect against several of the most dominant and favored category cues. The results confirm that classification inferences and attribute preference for new hybrid products are highly contextual, and as such, single category inferences need not translate directly into attribute preference.     

Buying/Selling Price Preference Reversals: Preference for Environmental Changes in Buying versus Selling Modes

Many studies have shown that the most people are willing to pay to obtain an object often is significantly less than the least they will accept to relinquish the object (i.e., selling prices tend to be higher than buying prices). Most tests of the buying/selling price discrepancy have elicited values either for everyday market items (e.g., mugs, candy bars) or for environmental changes (e.g., a decrease in air quality, a landfill clean-up). The literature indicates a possible interaction between buying/selling prices and commodity type; buying/selling price differences seem greater for environmental improvements than for market items. In other words, people show more relative preference for environmental improvements in selling modes than they do in buying modes. A significant difference in preference due to elicitation mode is commonly termed a "preference reversal." The four experiments presented here establish this new preference reversal and examine the reasons for it. The results from these studies provide information about the nature of preference reversals, the valuation process as a whole, and the unique problem of valuing complex and risky items such as environmental changes.     

Limitations of exemplar models of multi-attribute probabilistic inference

Observers were presented with pairs of objects varying along binary-valued attributes and learned to predict which member of each pair had a greater value on a continuously varying criterion variable. The predictions from exemplar models of categorization were contrasted with classic alternative models, including generalized versions of a "take-the-best" model and a weighted-additive model, by testing structures in which interactions between attributes predicted the magnitude of the criterion variable. Under typical training conditions, observers showed little sensitivity to the attribute interactions, thereby challenging the predictions from the exemplar models. In a condition involving highly extended training, observers eventually learned the relations between the attribute interactions and the criterion variable. However, an analysis of the observers' response times for making their paired-comparison decisions also challenged the exemplar model predictions. Instead, it appeared that most observers recoded the interacting attributes into emergent configural cues. They then applied a set of hierarchically organized rules based on the priority of the cues to make their decisions.     

The attribute priming effect in patients with Alzheimer's disease

Experiments with semantic priming (SP) paradigms have documented early hypopriming in patients with AD when concepts are used as primes and attribute concept features as targets, suggesting that concept attributes are vulnerable to damage very early in the disease course. The aims of this study were to confirm early priming reduction in the attribute condition in patients with AD and to determine which of several semantic indexes (such as the level of distinctiveness, correlation or feature dominance of concept features) best predicts the priming effect size in AD. We administered an SP attribute condition paradigm to 20 mildly demented patients with AD and to 10 NCs. We used concept–attribute pairs for which normative data of semantic indexes relative to both concept primes (i.e., number, type, mean level of dominance, distinctiveness and correlation of features constituting the concepts) and target features (i.e., level of feature dominance, correlation and distinctiveness) were available. Results showed that compared to NCs, the AD group obtained very reduced priming facilitation. Furthermore, the item regression analyses showed that the priming decrement in the AD group was predicted by the feature dominance of the target in the related pairs; that is, the lower the target feature dominance, the lower the priming effect elicited. These results confirmed hypopriming in the attribute condition from the very early phase of AD and support the view that attributes which are more salient for the identification of a given concept are also those most resistant to semantic memory degradation in AD pathology.     

Stating Preference for the Ethereal but Choosing the Concrete: How the Tangibility of Attributes Affects Attribute Weighting in Value Elicitation and Choice

Marketers routinely make use of stated consumer preferences and the relative attribute‐importance weights implied by these preferences when making decisions on issues such as advertising messages and product design. Using this information as a basis for managerial decision making is risky, though, if stated preferences diverge from actual choices. Practical evidence that such a divergence is of concern is provided by the current trend toward the use of stated choice‐based conjoint analysis. This article examines differences between the attribute‐importance weights consumers use during value elicitation and the attribute weights revealed to influence actual choice. The results of an empirical analysis of automobile stated preference and purchase decisions, and an experiment and subsequent qualitative analysis of wine choice, converge to suggest that consumers’ attribute weightings differ in value elicitation versus choice in a reliable manner. Specifically, we demonstrate a tangibility effect—the tendency for tangible attributes to be weighted relatively more heavily than intangible attributes in choice as compared to in value elicitation. The process underlying the tangibility effect is discussed, as are the implications for researchers and managers.