Stefan Herzog
Fluency heuristic: A model of how the mind exploits a by-product of information retrieval
Authors
Stefan Herzog (University of Basel)
Ralph Hertwig (University of Basel)
Lael Schooler (Max Planck Institute for Human Development, Berlin)
Which of the following two companies had higher sales in 2003: Siemens or Porsche? The fluency heuristic (FH; Schooler & Hertwig, 2005) predicts that of two recognized objects, the one recognized faster scores higher on a quantitative criterion. The FH is ecologically valid in environments in which recognition speed is a valid cue for this criterion. It exploits a characteristic of human memory, namely that objects can be retrieved faster from memory the more recently and frequently they have been encountered. Recognition speed is therefore only a valid cue for a criterion if environmental mediators (e.g., media mentions) reflect the criterion, inducing an ecological correlation between the criterion and the environmental frequencies, and, in turn, between the environmental frequencies and recognition speed. The results of several empirical studies and ecological analyses suggest that fluency can be an ecologically valid predictor and, in addition, the FH provides a psychologically plausible and valid description of people’s inferences. The current research thus establishes the viability of the FH as a model of human inferences.
|
|
|
Ulrich Hoffrage
The Use of Simple Heuristics Under Time Pressure
Authors
Ulrich Hoffrage (University of Lausanne)
Jörg Rieskamp (Max Planck Institute for Human Development, Berlin)
To what extent are simple heuristics adequate models for predicting inferences under different forms of time pressure? This question is addressed in three experimental studies. In Study 1, participants made inferences after searching for information on a computerized information board. Opportunity costs in terms of time induced the requisite time pressure. A simple lexicographic heuristic achieved the best fit in predicting participants’ inferences. Studies 2 and 3 show that high time pressure can be induced either by limiting the time for each choice or by having opportunity costs in terms of time, a common form of time pressure that has rarely been examined in the literature. The decision process under high time pressure, regardless of how it was induced, differed from the decision process under low time pressure. When assuming that people are equipped with a repertoire of decision strategies, then people seem to select strategies adaptively according to the inference situation.
|
|
|
Natalia Karelaia
Heuristic and linear models of judgment: Matching rules and environments.
Authors
Natalia Karelaia (University of Lausanne)
Robin M. Hogarth (Universitat Pompeu Fabra, Barcelona (E) )
Much research has highlighted incoherent implications of judgmental heuristics, and yet, other findings demonstrate high correspondence between predictions and outcomes. At the same time, judgment has been well-modeled in the form of “as if” linear models. Accepting the probabilistic nature of the environment, we use statistical tools to model how the performance of heuristic rules varies as a function of environmental characteristics. We further characterize the human use of linear models by exploring effects of different levels of cognitive ability. We illustrate with both theoretical analyses and simulations. Results are linked to the empirical literature by a meta-analysis of lens model studies. Using the same tasks, we estimate the performance of both heuristics and humans where the latter are assumed to use linear models. Our results emphasize that judgmental accuracy depends on matching characteristics of rules and environments and highlight the trade-off between using linear models and heuristics. Whereas the former can be cognitively demanding, the latter are simple to implement. However, they require knowledge to indicate when they should be used.
|
|
|
Robin Hau
Simple Strategies in Decisions from Experience
Authors
Robin Hau (University of Basel)
Timothy J. Pleskac (University of Indiana)
Ralph Hertwig (University of Basel)
When people make decisions, more often than not, they base their choices on experience with similar situations. Correspondingly, in studies of decisions from experience, participants gather experience with their options prior to making their choice. This experience replaces the objective information provided in decisions from description. Studies have found a gap between experience-based choices and decisions made from summary descriptions of the lotteries. Specifically, rare events seem to have less impact in decisions from experience than in decisions from description. This may be explained by sampling error due to small samples. To reduce sampling error, we conducted two studies that had respondents draw large samples. This indeed reduced the description-experience gap, but did not close it. To investigate the remaining differences, we employed several models to predict individual choices based on sample experience. We found that strategies that are good predictors of decisions from description (prospect theory, priority heuristic) fail to predict decisions from experience well. Conversely, decisions from experience can be predicted by remarkably simple strategies that do not predict an appreciable proportion of choices in decisions from description. These strategies embody risk-neutral or even risk-seeking behaviour. We conclude that the description-experience gap is the combined product of sampling error and the use of different strategies in the two paradigms.
|
|
|
Lukas Baenninger
Motivational Rationality - A Heuristic Theory of Decisions Under Risk
Authors
Lukas Baenninger (University of Zurich)
Damian Laege (University of Zurich )
The Prospect Theory (PT) explains experimentally exposed systematic deviations from the classic code of conduct of a profit-maximizing individual: excessive striving for security and risk seeking are ascribed to a systematic distortion of human perception of probability scale, described by the weighting function.
The empirical results that lead to this function are decision tasks with two equal monetary risk prospects, e.g. to win CHF 100000 with p = 0.01 or CHF 50000 with p = 0.02. Here, the majority of subjects opt for the riskier alternative. Is this a distortion of probability cognition?
The theory presented here postulates that the alternatives of choice, which are purely monetary in the experiments of PT, are linked to motivationally defined classes of different situations; e.g. the situation of winning with small probabilities is grouped as „status gain“ (status goods are by definition reserved for a minority of the population, their probability of occurrence is thus typically small). In such situations people usually choose – consistent with results of the experiments of PT – the riskier alternative that promises higher status.
To test the theory, subjects were confronted with some of the original experiments of PT, but these experiments were embedded in the context of everyday life. The conclusion is that the reduction of the process of decision under risk to a systematic distortion of human perception of probabilities, as proposed by PT, leads to wrong predictions.
|