mixtur: An R package for designing, analysing, and modelling continuous report visual short-term memory studies

Abstract

Visual short-term memory (vSTM) is often measured via continuous-report tasks whereby participants are presented with stimuli that vary along a continuous dimension (e.g., colour) with the goal of memorising the stimuli features. At test, participants are probed to recall the feature value of one of the memoranda in a continuous manner (e.g., by clicking on a colour wheel). The angular deviation between the participant response and the true feature value provides an estimate of recall---and hence, vSTM---precision. Two prominent models of performance on such tasks are the two- and three-component mixture models (Bays et al., 2009; Zhang & Luck, 2008). Both models decompose participant responses into probabilistic mixtures of: (1) responses to the true target value based on a noisy memory representation; (2) random guessing when memory fails. In addition, the three-component model proposes (3) responses to a non-target feature value (i.e., binding errors). Here we report the development of mixtur, an open-source package written for the statistical programming language R that facilitates the fitting of the 2- and 3-component mixture models to continuous report data. We also report the results of several simulations conducted to develop recommendations for researchers on trial numbers, set-sizes and memoranda similarity, as well as conducting parameter recovery and model recovery simulations. It is our hope that mixtur will lower the barrier of entry for utilising mixture modelling

Shielding working-memory representations from temporally predictable external interference

“…The relative proportion of the non-target (i.e., interfering item) distribution was estimated separately for each participant and condition (i.e., interference type and temporal predictability). The mixture modelling analysis was performed using the ‘mixtur’ package (version 1.0.0; Grange & Moore, 2020 ).…”

A tutorial for estimating mixture models for visual working memory tasks in brms: Introducing the Bayesian Measurement Modeling (bmm) package for R

“…Until recently, most researchers have used custom-built code to implement the existing mixture models. Although there is software that implements some of these measurement models (e.g., Grange et al, 2021), this software mostly uses a two-step procedure. First, the parameters of the model are estimated separately for each subject in each condition using maximum likelihood methods.…”

“…These methods constrain parameter estimation in several ways and can either lead to the over-or underestimation of standard errors in statistical tests (Boehm et al, 2018;Skrondal & Laake, 2001). Furthermore, to obtain robust parameter estimates maximum likelihood estimation requires at least 200 trials per subject per condition (Grange et al, 2021).…”

Sensory Recruitment in Visual Short-Term Memory: A Systematic Review and Meta-Analysis of Sensory Visual Cortex Interference Using Transcranial Magnetic Stimulation

“…Subsequent studies reported experiments that provide further evidence for the involvement of the sensory visual cortex in VSTM encoding using more sensitive statistical methods, such as mixture models (see Grange et al, 2021). Koivisto et al (2017) conducted two experiments to investigate whether TMS affects precision or guessing rates in a VSTM task, and whether these are affected dichotomously ('all or nothing') or gradually.…”