Hidden Markov Models of Evidence Accumulation in Speeded Decision Tasks

Abstract

Speeded decision tasks are usually modeled within the evidence accumulation framework, enabling inferences on latent cognitive parameters, and capturing dependencies between the observed response times and accuracy. An example is the speed-accuracy trade-off, where people sacrifice speed for accuracy (or vice versa). Different views on this phenomenon lead to the idea that participants may not be able to control this trade-off on a continuum, but rather switch between distinct states (Dutilh et al., Cognitive Science 35(2):211–250, 2010). Hidden Markov models are used to account for switching between distinct states. However, combining evidence accumulation models with a hidden Markov structure is a challenging problem, as evidence accumulation models typically come with identification and computational issues that make them challenging on their own. Thus, an integration of hidden Markov models with evidence accumulation models has still remained elusive, even though such models would allow researchers to capture potential dependencies between response times and accuracy within the states, while concomitantly capturing different behavioral modes during cognitive processing. This article presents a model that uses an evidence accumulation model as part of a hidden Markov structure. This model is considered as a proof of principle that evidence accumulation models can be combined with Markov switching models. As such, the article considers a very simple case of a simplified Linear Ballistic Accumulation. An extensive simulation study was conducted to validate the model’s implementation according to principles of robust Bayesian workflow. Example reanalysis of data from Dutilh et al. (Cognitive Science 35(2):211–250, 2010) demonstrates the application of the new model. The article concludes with limitations and future extensions or alternatives to the model and its application.

Subjective Evidence Evaluation Survey For Many-Analysts Studies

“…In addition to examining prior and posterior predictive distributions, we ensured that the proposed computational model as well as the implemented Markov chain Monte Carlo (MCMC) algorithm is able to recover the prior distribution when no data are observed, that the implemented MCMC algorithm returns unbiased estimates, and that the data effectively update the prior beliefs. These additional checks were proposed by Kucharský et al (2021), based on the recommendations by Talts et al (2018) and Schad et al (2021).…”

Validation and Comparison of Non-Stationary Cognitive Models: A Diffusion Model Application

“…These fluctuations can have a significant impact on our cognitive functioning, but they are often overlooked or simplified in traditional models of cognition. And while these often assume cognitive processes to be stable and time-invariant, there has been a growing recognition that traditional models do not fully capture the complexity and variability of real-world cognition (Beer, 2023;Cochrane et al, 2023;Evans & Brown, 2017;Gunawan et al, 2022;Kucharský et al, 2021;Li et al, 2023;. Common approaches to address variability in the components of cognitive models can be broadly classified into four categories: stationary variability, trial binning, regression approach, and frontend-backend models.…”

Neural superstatistics for Bayesian estimation of dynamic cognitive models

“…A viable approach for modeling parameter transitions is offered by hidden Markov models (HMMs). For instance 27 , accounted for different response states during a decision-making task by combining a HMM with an evidence accumulation model of decision-making. This model combination allows for discontinuous changes on longer time scales and continuous changes on shorter time scales.…”

“…Nevertheless, superstatistical models can be rigorously validated in the same way as their static counterparts, using standard model criticism methods, such as simulation-based calibration (SBC) to assess computational faithfulness, parameter recovery for inferential calibration, posterior re-simulation checks for assessing model adequacy, as well as cross-validation for assessing predictive performance 43 , 44 . Superstatistical models allow us to address questions about how cognitive systems undergo distinct transitions in various settings 27 . Further, one can examine which model parameters explain behavioral fluctuations without predefined equations that fix the hypothesized temporal evolution of specific parameters.…”