Changelog

hBayesDM 1.3.1

• Add plot functions for Hierarchical Gaussian Filter models: plot_hgf_ibrb, plot_hgf_ibrb_single.

hBayesDM 1.3.0

• Added a Hierarchical Gaussian Filter model for binary inputs and binary responses: hgf_ibrb for hierarchical Bayesian analysis and hgf_ibrb_single for individual Bayesian analysis.
• Added new article: Hierarchical Bayesian Analysis on Hierarchical Gaussian Filter

hBayesDM 1.2.12022-09-23

• Fixed a pkgdown error.

hBayesDM 1.2.02022-09-10

• Added a drift diffusion model and two reinforcement learning-drift diffision models for the probabilistic selection task: pstRT_ddm, pstRT_rlddm1, and pstRT_rlddm6.
• Added multiple models for the banditNarm task: banditNarm_2par_lapse, banditNarm_4par, banditNarm_delta, banditNarm_kalman_filter, banditNarm_lapse, banditNarm_lapse_decay, and banditNarm_singleA_lapse.
• Fixed bart_ewmv to avoid dividing by zero.

hBayesDM 1.1.12021-05-03

• Fix symbolic link errors for stan files and example data.

hBayesDM 1.1.0

• Added the cumulative model for the Cambridge gambling task: cgt_cm.
• Added two new models for aversive learning tasks: alt_delta and alt_gamma.
• Added exponential-weight mean-variance model for BART task: bart_ewmv.
• Added simple Q learning model for the probabilistic selection task: prl_Q.
• Added signal detection theory model for 2-alternative forced choice task: task2AFC_sdt.

hBayesDM 1.0.22019-11-13

• Fixed an error on using data.frame objects as data (#112).

hBayesDM 1.0.12019-09-01

• Minor fix on the plotting function.

hBayesDM 1.0.02019-08-30

hBayesDM 0.7.22019-02-12

• Add three new models for the bandit4arm task: bandit4arm_2par_lapse, bandit4arm_lapse_decay and bandit4arm_singleA_lapse.
• Fix various (minor) errors.

hBayesDM 0.7.12019-01-21

• Make it usable without manually loading rstan.
• Remove an annoying warning about using ..insensitive_data_columns.

hBayesDM 0.7.02018-12-13

• Now, in default, you should build a Stan file into a binary for the first time to use it. To build all the models on installation, you should set an environmental variable BUILD_ALL to true before installation.
• Now all the implemented models are refactored using hBayesDM_model function. You don’t have to change anything to use them, but developers can easily implement new models now!
• We added a Kalman filter model for 4-armed bandit task (bandit4arm2_kalman_filter; Daw et al., 2006) and a probability weighting function for general description-based tasks (dbdm_prob_weight; Erev et al., 2010; Hertwig et al., 2004; Jessup et al., 2008).
• Initial values of parameter estimation for some models are updated as plausible values, and the parameter boundaries of several models are fixed (see more on issue #63 and #64 in Github).
• Exponential and linear models for choice under risk and ambiguity task now have four model regressors: sv, sv_fix, sv_var, and p_var.
• Fix the Travix CI settings and related codes to be properly passed.

hBayesDM 0.6.3

• Update the dependencies on rstan (>= 2.18.1)
• No changes on model files, as same as the version 0.6.2

hBayesDM 0.6.2

• Fix an error on choiceRT_ddm (#44)

hBayesDM 0.6.1

• Solve an issue with built binary files.
• Fix an error on peer_ocu with misplaced parentheses.

hBayesDM 0.6.02018-09-11

• Add new tasks (Balloon Analogue Risk Task, Choice under Risk and Ambiguity Task, Probabilistic Selection Task, Risky Decision Task (a.k.a. Happiness task), Wisconsin Card Sorting Task)
• Add a new model for the Iowa Gambling Task (igt_orl)
• Change priors (Half-Cauchy(0, 5) –> Half-Cauchy(0, 1) or Half-Normal(0, 0.2)
• printFit function now provides LOOIC weights and/or WAIC weights

hBayesDM 0.5.1

• Add models for the Two Step task
• Add models without indecision point parameter (alpha) for the PRL task (prl_*_woa.stan)
• Model-based regressors for the PRL task are now available
• For the PRL task & prl_fictitious.stan & prl_fictitious_rp.stan –> change the range of alpha (indecision point) from [0, 1] to [-Inf, Inf]

hBayesDM 0.5.02018-01-03

• Support variational Bayesian methods (vb=TRUE)
• Allow posterior predictive checks, except for drift-diffusion models (inc_postpred=TRUE)
• Add the peer influence task (Chung et al., 2015, USE WITH CAUTION for now and PLEASE GIVE US FEEDBACK!)
• Add ‘prl_fictitious_rp’ model
• Made changes to be compatible with the newest Stan version (e.g., // instead of # for commenting).
• In ’prl_*’ models, ‘rewlos’ is replaced by ‘outcome’ so that column names and labels would be consistent across tasks as much as possible.
• Email feature is disabled as R mail package does not allow users to send anonymous emails anymore.
• When outputs are saved as a file (*.RData), the file name now contains the name of the data file.

hBayesDM 0.4.02017-05-23

• Add a choice reaction time task and evidence accumulation models
  • Drift diffusion model (both hierarchical and single-subject)
  • Linear Ballistic Accumulator (LBA) model (both hierarchical and single-subject)
• Add PRL models that can fit multiple blocks
• Add single-subject versions for the delay discounting task (dd_hyperbolic_single and dd_cs_single).
• Standardize variable names across all models (e.g., rewlos –> outcome for all models)
• Separate versions for CRAN and GitHub. All models/features are identical but the GitHub version contains precompilled models.

hBayesDM 0.3.1

• Remove dependence on the modeest package. Now use a built-in function to estimate the mode of a posterior distribution.
• Rewrite the “printFit” function.

hBayesDM 0.3.02017-01-22

• Made several changes following the guidelines for R packages providing interfaces to Stan.
• Stan models are precompiled and models will run immediately when called.
• The default number of chains is set to 4.
• The default value of adapt_delta is set to 0.95 to reduce the potential for divergences.
• The “printFit” function uses LOOIC by default. Users can select WAIC or both (LOOIC & WAIC) if needed.

hBayesDM 0.2.3.32016-12-28

• Add help files
• Add a function for checking Rhat values (rhat).
• Change a link to its tutorial website

hBayesDM 0.2.3.22016-12-21

• Use wide normal distributions for unbounded parameters (gng_* models).
• Automatic removal of rows (trials) containing NAs.

hBayesDM 0.2.3.12016-09-30

• Add a function for plotting individual parameters (plotInd)

hBayesDM 0.2.32016-07-17

• Add a new task: the Ultimatum Game
• Add new models for the Probabilistic Reversal Learning and Risk Aversion tasks
• ‘bandit2arm’ -> change its name to ‘bandit2arm_delta’. Now all model names are in the same format (i.e., TASK_MODEL).
• Users can extract model-based regressors from gng_m* models
• Include the option of customizing control parameters (adapt_delta, max_treedepth, stepsize)
• ‘plotHDI’ function -> add ‘fontSize’ argument & change the color of histogram

hBayesDM 0.2.12016-04-03

hBayesDM 0.2.02016-03-25

Initially released.