Selective associations and causality judgments: Presence of a strong causal factor may reduce judgments of a weaker one.

Baker, A. G.; Mercier, Pierre; Vallée-Tourangeau, Frédéric; Frank, Robert H.; Pan, Maria

doi:10.1037/0278-7393.19.2.414

Cited by 121 publications

(192 citation statements)

References 50 publications

Supporting

Mentioning

169

Contrasting

Unclassified

Order By: Relevance

“…In fact, the data from several articles that claimed to show that people were poor reasoners because they deviated far from D P when reasoning about multiple causes of effects (e.g., Baker, Mercier, Vallée-Tourangeau, Frank, & Pan, 1993;Chapman, 1991;Chapman & Robbins, 1990;Price & Yates, 1993 can be reanalyzed to show that people seem to be using the "smarter" conditional contingency strategy (Cheng, 1993;Melz, Cheng, Holyoak, & Waldmann, 1993;Shanks, 1993Shanks, , 1995Spellman, 1993Spellman, , 1996aSpellman, , 1996b Schaller and colleagues (Schaller, 1992a(Schaller, , 1992bSchaller & O' Brien, 1992) have investigated the use of something akin to conditionalization-what he calls "intuitive analysis of covariance"-in tasks that do not involve causal reasoning. For instance, in a study that is reminiscent of our baseball example, subjects were presented with information about the racquetball prowess of two potential doubles partners.…”

Section: Simpson's Paradox: the Mathematical Problem Of Differing Basmentioning

confidence: 99%

How two causes are different from one: The use of (un)conditional information in Simpson's paradox

2001

View full text Add to dashboard Cite

In a causally complex world, two (or more) factors may simultaneously be potential causes of an effect. To evaluate the causal efficacy of a factor, the alternative factors must be controlled for (or conditionalized on). Subjects judged the causal strength of two potential causes of an effect that covaried with each other, thereby setting up a Simpson's paradox-a situation in which causal judgments should vary widely depending on whether or not they are conditionalized on the alternative potential cause. In Experiments 1 (table format) and 2 (trial-by-trialformat), the subjects did conditionalize their judgments for one causal factor on a known alternative cause. The subjects also demonstrated that they knew what information was needed to properly make causal judgments when two potential causes are available. In Experiment 3 (trial-by-trial),those subjects who were not told about the causal mechanism by which the alternative cause operated were less likely to conditionalize on it. However, the more a subject recognized the covariation between the alternative cause and the effect, the more the subject conditionalized on it. Such behavior may arise from the interaction between bottom-up and top-down processing.

show abstract

Section: Simpson's Paradox: the Mathematical Problem Of Differing Basmentioning

confidence: 99%

How two causes are different from one: The use of (un)conditional information in Simpson's paradox

2001

View full text Add to dashboard Cite

show abstract

“…The impact of the SS prior is that when two (or more) possible generative causes of E cooccur and one cause has a stronger statistical link to E than the other, the presence of the stronger cause will tend to reduce the judged strength of the weaker one (cf. Baker, Mercier, Vallée-Tourangeau, Frank, & Pan, 1993;Busemeyer, Myung, & McDaniel, 1993). Figure 3 shows an example of the posterior distribution of w 1 obtained given contingency data of p(e ϩ |c Ϫ ) ϭ 12/16 and p(e ϩ |c ϩ ) ϭ 16/16, based on either SS or uniform priors (see Appendix A for derivation).…”

Section: Generic Priors For Sparse and Strong Causesmentioning

confidence: 99%

Bayesian generic priors for causal learning.

Lu¹,

Yuille²,

Liljeholm³

et al. 2008

Psychological Review

171

300

View full text Add to dashboard Cite

The article presents a Bayesian model of causal learning that incorporates generic priors-systematic assumptions about abstract properties of a system of cause-effect relations. The proposed generic priors for causal learning favor sparse and strong (SS) causes-causes that are few in number and high in their individual powers to produce or prevent effects. The SS power model couples these generic priors with a causal generating function based on the assumption that unobservable causal influences on an effect operate independently (P. W. Cheng, 1997). The authors tested this and other Bayesian models, as well as leading nonnormative models, by fitting multiple data sets in which several parameters were varied parametrically across multiple types of judgments. The SS power model accounted for data concerning judgments of both causal strength and causal structure (whether a causal link exists). The model explains why human judgments of causal structure (relative to a Bayesian model lacking these generic priors) are influenced more by causal power and the base rate of the effect and less by sample size. Broader implications of the Bayesian framework for human learning are discussed.

show abstract

“…Arcediano, Ortega, and Matute (1996) developed the "Martians" game to explore classical conditioning using Martians and explosions as stimuli (see also Baeyens et al, 2005;Blanco, Matute, & Vadillo, 2010;Franssen, Clarysse, Beckers, van Vooren, & Baeyens, 2010). Gamelike tasks have been used to study instrumental learning with stimuli presented as balloons that must be shot from the sky (Krageloh, Zapanta, Shepherd, & Landon, 2010), minefields to be navigated (Baker, Mercier, Vallee-Tourangeau, Frank, & Pan, 1993), or a host of similar examples (Lie, Harper, & Hunt, 2009;Molet, Jozefowiez, & Miller, 2010;Paredes-Olay, Abad, Gamez, & Rosas, 2002;Stokes & Balsam, 2001;Stokes & Harrison, 2002). Discrimination and generalization learning have been presented as melodies that participants must classify as belonging to different composers (Artigas, Chamizio, & Peris, 2001) or as torpedoes to be launched at certain flying objects but not others (Nelson & Sanjuan, 2008;Nelson, Sanjuan, Vadillo-Ruiz, & Perez, 2011).…”

Section: Gaming-up Experimentsmentioning

confidence: 99%

Gamelike features might not improve data

et al. 2012

View full text Add to dashboard Cite

Many psychological experiments require participants to complete lots of trials in a monotonous task, which often induces boredom. An increasingly popular approach to alleviate such boredom is to incorporate gamelike features into standard experimental tasks. Games are assumed to be interesting and, hence, motivating, and better motivated participants might produce better data (with fewer lapses in attention and greater accuracy). Despite its apparent prevalence, the assumption that gamelike features improve data is almost completely untested. We test this assumption by presenting a choice task and a change detection task in both gamelike and standard forms. Response latency, accuracy, and overall task performance were unchanged by gamelike features in both experiments. We present a novel cognitive model for the choice task, based on particle filtering, to decorrelate the dependent variables and measure performance in a more psychologically meaningful manner. The model-based analyses are consistent with the hypothesis that gamelike features did not alter cognition. A postexperimental questionnaire indicated that the gamelike version provided a more positive and enjoyable experience for participants than the standard task, even though this subjective experience did not translate into data effects. Although our results hold only for the two experiments examined, the gamelike features we incorporated into both tasks were typical of-and at least as salient and interesting as those usually used by-experimental psychologists. Our results suggest that modifying an experiment to include gamelike features, while leaving the basic task unchanged, may not improve the quality of the data collected, but it may provide participants with a better experimental experience.

show abstract

Selective associations and causality judgments: Presence of a strong causal factor may reduce judgments of a weaker one.

Abstract: This document is copyrighted by the American Psychological Association or one of its allied publishers.This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Cited by 121 publications

References 50 publications

How two causes are different from one: The use of (un)conditional information in Simpson's paradox

How two causes are different from one: The use of (un)conditional information in Simpson's paradox

Bayesian generic priors for causal learning.

Gamelike features might not improve data

Contact Info

Product

Resources

About