Bayesian models of cognition revisited: Setting optimality aside and letting data drive psychological theory.

Tauber, Sean; Navarro, Daniel J.; Perfors, Amy; Steyvers, Mark

doi:10.1037/rev0000052

Cited by 100 publications

(63 citation statements)

References 89 publications

(191 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given the ubiquity of individual differences across nearly every domain of human cognition, understanding the effects of heterogeneity is an important part of appropriately applying iterated learning methodologically as well as understanding cultural and linguistic evolution in the world. refers to a much weaker claim: that their behavior can be described using the Bayes' rule, as discussed in Tauber, Navarro, Perfors, and Steyvers (2017). 2.…”

Section: What Are the Biases That Iterated Learning Reveals?mentioning

confidence: 99%

When Extremists Win: Cultural Transmission Via Iterated Learning When Populations Are Heterogeneous

et al. 2018

View full text Add to dashboard Cite

How does the process of information transmission affect the cultural or linguistic products that emerge? This question is often studied experimentally and computationally via iterated learning, a procedure in which participants learn from previous participants in a chain. Iterated learning is a powerful tool because, when all participants share the same priors, the stationary distributions of the iterated learning chains reveal those priors. In many situations, however, it is unreasonable to assume that all participants share the same prior beliefs. We present four simulation studies and one experiment demonstrating that when the population of learners is heterogeneous, the behavior of an iterated learning chain can be unpredictable and is often systematically distorted by the learners with the most extreme biases. This results in group-level outcomes that reflect neither the behavior of any individuals within the population nor the overall population average. We discuss implications for the use of iterated learning as a methodological tool as well as for the processes that might have shaped cultural and linguistic evolution in the real world.

show abstract

Section: What Are the Biases That Iterated Learning Reveals?mentioning

confidence: 99%

When Extremists Win: Cultural Transmission Via Iterated Learning When Populations Are Heterogeneous

et al. 2018

View full text Add to dashboard Cite

show abstract

“…While this model is a Bayesian model, it makes no claims about the optimality of pilot behavior with respect to the world, given it was not populated with real-world data. It, thus, represents a descriptive model rather than a normative or rational one (see Tauber, Navarro, Perfors, & Steyvers, 2017). Gigerenzer and Todd (1999) have reported that people commonly do not consider all available information when making decisions but instead adopt "fast and frugal heuristics" where search is based on only a subset of the available information.…”

Section: Discussionmentioning

confidence: 99%

Inference in the Wild: A Framework for Human Situation Assessment and a Case Study of Air Combat

McAnally

Davey²,

White

et al. 2018

Cognitive Science

View full text Add to dashboard Cite

Situation awareness is a key construct in human factors and arises from a process of situation assessment (SA). SA comprises the perception of information, its integration with existing knowledge, the search for new information, and the prediction of the future state of the world, including the consequences of planned actions. Causal models implemented as Bayesian networks (BNs) are attractive for modeling all of these processes within a single, unified framework. We elicited declarative knowledge from two Royal Australian Air Force (RAAF) fighter pilots about the information sources used in the identification (ID) of airborne entities and the causal relationships between these sources. This knowledge was represented in a BN (the declarative model) that was evaluated against the performance of 19 RAAF fighter pilots in a low-fidelity simulation. Pilot behavior was well predicted by a simple associative model (the behavioral model) with only three attributes of ID. Search for information by pilots was largely compensatory and was near-optimal with respect to the behavioral model. The average revision of beliefs in response to evidence was close to Bayesian, but there was substantial variability. Together, these results demonstrate the value of BNs for modeling human SA.

show abstract

“…Indeed, the implication that the absolute response bias is not monotonic in σ P is not considered in and Davis (2013,2015), Mozer, Pashler, and Homaei (2008), Perfors, Tenenbaum, Griffiths, andXu (2011), Petzschner, Glasauer, andStephan (2015), Sailor and Antoine (2005), Tauber et al (2017), and Tenenbaum, Griffiths, and Kemp (2006). 41 For instance, we note two violations of the chain rule on page 239.…”

Section: Discussionmentioning

confidence: 99%

On the Category Adjustment Model: Another Look at Huttenlocher, Hedges, and Vevea (2000)

Duffy

Smith

2017

SSRN Journal

View full text Add to dashboard Cite

Huttenlocher, Hedges, and Vevea (2000) (Why do categories affect stimulus judgment? Journal of Experimental Psychology: General, 129, introduce the category adjustment model (CAM), which posits that participants imperfectly remember stimuli in serial judgment tasks. In order to maximize accuracy, CAM holds that participants use information about the distribution of the stimuli to improve their judgments. CAM predicts that judgments will be a weighted average of imperfect memories of the stimuli and the mean of the distribution of stimuli.Huttenlocher, Hedges, and Vevea (2000) report on three experiments and the authors conclude that CAM is "verified." We attempt to replicate Experiment 3 from Huttenlocher et al. (2000).We analyze judgment-level data rather than averaged data. We find evidence of a bias toward a set of recent stimuli rather than a bias toward the running mean. We also do not find evidence of the joint hypothesis that the participants learned the distribution of stimuli and employed this information in their judgments. The judgments in our dataset are not consistent with CAM. We discuss how the apparent defects in HHV went unnoticed and how such mistakes can be avoided in future research. Finally, we hope that the techniques that we employ will be used to test other datasets that are currently regarded as consistent with CAM or any Bayesian model of judgment.Keywords: judgment, memory, category adjustment model, central tendency bias, recency effects, Bayesian judgments Running Head: ON THE CATEGORY ADJUSTMENT MODEL 3 It has been known for some time that when participants perform magnitude judgments, there is a bias toward the mean (Hollingworth, 1910;Poulton, 1979). For instance, in the judgment of the length of lines, longer lines tend to be underestimated and shorter lines tend to be overestimated. This effect is sometimes referred to as the central tendency bias. 1 Huttenlocher, Hedges, and Vevea (2000), hereafter referred to as HHV, propose that participants imperfectly remember and perceive the stimuli. Due to these imperfections, HHV posit that participants use information about the distribution of the stimuli to improve their judgments. HHV refer to this as the category adjustment model, hereafter referred to as CAM.CAM predicts that judgments will be a weighted average of imperfect memory of the stimuli and the mean of the distribution (category) of stimuli. In the description of CAM, the authors state, "This process can be likened to a Bayesian statistical procedure designed to maximize the average accuracy of estimation" (p. 220). Since judgments will be an optimal weighted average of the imperfect memory of the stimulus and the mean of the distribution, CAM offers a Bayesian explanation of the central tendency bias.HHV describe one prediction of CAM as the following, "…the concentration of instances in the category should affect the variability of stimulus estimates. In particular, the variability of estimates of all categorized stimuli should be less when the prior distribu...

show abstract

Bayesian models of cognition revisited: Setting optimality aside and letting data drive psychological theory.

Cited by 100 publications

References 89 publications

When Extremists Win: Cultural Transmission Via Iterated Learning When Populations Are Heterogeneous

When Extremists Win: Cultural Transmission Via Iterated Learning When Populations Are Heterogeneous

Inference in the Wild: A Framework for Human Situation Assessment and a Case Study of Air Combat

On the Category Adjustment Model: Another Look at Huttenlocher, Hedges, and Vevea (2000)

Contact Info

Product

Resources

About