Most data analyses rely on models. To complement statistical models, psychologists have developed cognitive models, which translate observed variables into psychologically interesting constructs. Response time models, in particular, assume that response time and accuracy are the observed expression of latent variables including 1) ease of processing, 2) response caution, 3) response bias, and 4) non-decision time. Inferences about these psychological factors hinge upon the validity of the models' parameters. Here, we use a blinded, collaborative approach to assess the validity of such model-based inferences. Seventeen teams of researchers analyzed the same 14 data sets. In each of these two-condition data sets, we manipulated properties of participants' behavior in a two-alternative forced choice task. The contributing teams were blind to the manipulations, and had to infer what aspect of behavior was changed using their method of choice. The contributors chose to employ a variety of models, estimation methods, and inference procedures. Our results show that, although conclusions were similar across different methods, these "modeler's degrees of freedom" did affect their inferences. Interestingly, many of the simpler approaches yielded as robust and accurate inferences as the more complex methods. We recommend that, in general, cognitive models become a typical analysis tool for response time data. In particular, we argue that the simpler models and procedures are sufficient for standard experimental designs. We finish by outlining situations in which more complicated models and methods may be necessary, and discuss potential pitfalls when interpreting the output from response time models.

Most data analyses rely on models. To complement statistical models, psychologists have developed cognitive models, which translate observed variables into psychologically interesting constructs. Response time models, in particular, assume that response time and accuracy are the observed expression of latent variables including 1) ease of processing, 2) response caution, 3) response bias, and 4) nondecision time. Inferences about these psychological factors, hinge upon the validity of the models' parameters. Here, we use a blinded, collaborative approach to assess the validity of such model-based inferences. Seventeen teams of researchers analyzed the same 14 data sets. In each of these two-condition data sets, we manipulated properties of participants' behavior in a two-alternative forced choice task. The contributing teams were blind to the manipulations, and had to infer what aspect of behavior was changed using their method of choice. The contributors chose to employ a variety of models, estimation methods, and inference procedures. Our results show that, although conclusions were similar across different methods, these "modeler's degrees of freedom" did affect their inferences. Interestingly, many of the simpler approaches yielded as robust and accurate inferences as the more complex methods. We recommend that, in general, cognitive models become a typical analysis tool for response time data. In particular, we argue that the simpler models and procedures are sufficient for standard experimental designs. We finish by outlining situations in which more complicated models and methods may be necessary, and discuss potential pitfalls when interpreting the output from response time models.

Speed-accuracy tradeoff (SAT) is an adaptive process balancing urgency and caution when making decisions. Computational cognitive theories, known as “evidence accumulation models”, have explained SATs via a manipulation of the amount of evidence necessary to trigger response selection. New light has been shed on these processes by single-cell recordings from monkeys who were adjusting their SAT settings. Those data have been interpreted as inconsistent with existing evidence accumulation theories, prompting the addition of new mechanisms to the models. We show that this interpretation was wrong, by demonstrating that the neural spiking data, and the behavioural data are consistent with existing evidence accumulation theories, without positing additional mechanisms. Our approach succeeds by using the neural data to provide constraints on the cognitive model. Open questions remain about the locus of the link between certain elements of the cognitive models and the neurophysiology, and about the relationship between activity in cortical neurons identified with decision-making vs. activity in downstream areas more closely linked with motor effectors.

Theory development in both psychology and neuroscience can benefit by consideration of both behavioral and neural data sets. However, the development of appropriate methods for linking these data sets is a difficult statistical and conceptual problem. Over the past decades, different linking approaches have been employed in the study of perceptual decision-making, beginning with rudimentary linking of the data sets at a qualitative, structural level, culminating in sophisticated statistical approaches with quantitative links. We outline a new approach, in which a single model is developed that jointly addresses neural and behavioral data. This approach allows for specification and testing of quantitative links between neural and behavioral aspects of the model. Estimating the model in a Bayesian framework allows both data sets to equally inform the estimation of all model parameters. The use of a hierarchical model architecture allows for a model, which accounts for and measures the variability between neurons. We demonstrate the approach by re-analysis of a classic data set containing behavioral recordings of decision-making with accompanying single-cell neural recordings. The joint model is able to School of Psychology, University of Newcastle, Callaghan, New South Wales, 2308, Australia capture most aspects of both data sets, and also supports the analysis of interesting questions about prediction, including predicting the times at which responses are made, and the corresponding neural firing rates.

Reasoning and inference are well-studied aspects of basic cognition that have been explained as statistically optimal Bayesian inference. Using a simplified experimental design, we conducted quantitative comparisons between Bayesian inference and human inference at the level of individuals. In 3 experiments, with more than 13,000 participants, we asked people for prior and posterior inferences about the probability that 1 of 2 coins would generate certain outcomes. Most participants' inferences were inconsistent with Bayes' rule. Only in the simplest version of the task did the majority of participants adhere to Bayes' rule, but even in that case, there was a significant proportion that failed to do so. The current results highlight the importance of close quantitative comparisons between Bayesian inference and human data at the individual-subject level when evaluating models of cognition.