Posts

Seeking a criminal partner is not unlike searching for a dance partner on a Saturday night. Humans are inherently social creatures, driven by the innate need to connect with others. We interact with other people almost every day, forming a so-called “social network”. Interestingly, regardless of the end goal - whether it is dancing or robbery - the mechanisms underlying these social connections have remarkable similarities.

The ability to read the mind of criminals for sure sounds exciting. However, as exciting as it may sound, assessing anyone’s decision-making processes is not the most straightforward task. This post will give a brief overview of the behavioral economics, psychology, and criminology literature on offenders’ decision-making. It will be discussed whether actions of crime can be considered rational and whether social influence plays a role in criminal behavior.

The post aims to give a brief review of psychopathy as a neuropsychiatric disorder. It starts with a discussion of features that define the psychopath. Then it takes a closer look at some of the traits, such as empathy, to show the general beliefs and controversy across the studies with the provided overview of the dysfunctions in the brain and impairments that stem from them. Lastly, the post includes an overview of research on “successful” and “unsuccessful” psychopaths and assesses the literature-based opinion on whether psychopathy can be treated.

How to fit reinforcement learning models to behavioral data using Bayesian inference. This part is focused on the hierarchical Bayesian modeling and particularly on the usage of hBayesDM package. Approaches for the model diagnostic, selection, validation are discussed. The post also goes over groups comparison using posterior distributions of model parameters. Additionally, a brief results comparison between Bayesian inference and Maximum Likelihood Estimation is provided.

How to fit reinforcement learning models to behavioral data using Maximum Likelihood Estimation (MLE). The main goal is to show how to answer research questions using modeling. Post goes over important steps of modeling, such as model selection, model validation, and data generation.

This post serves as an introduction to the EEG data processing and particularly the usage of MNE-Python package. The post goes over such preprocessing steps as labeling bad channels and trials, artifacts removal, and data epoching. Additionally, the event-related potential is calculated and compared between groups and conditions. Sample of EEG data is taken from Cavanagh et al. (2019) experiment.

Frequentist vs Bayesian battle is one of the “hot” topics in the statistics world. In this post, we will go over the differences between Frequentist and Bayesian approaches using the hypothesis testing for a population proportion. Note that the goal is to introduce the idea of both approaches, rather than selecting “the best” one.

Overview of the math behind an Ordinary Least Squares method, and what is needed to be taken into consideration when fitting a regression line.

Literature review for the final project at 2020 International Youth Neuroscience Association Summer Course.

Overview of the power and the effect size of the test and why low p-value is not always enough to make a decision.