We have made available some repositories of example markdown scripts, as well as a ‘cookbook’ of useful tips here: Hopefully this will help others to develop the necessary skills 4/6.

All of our contributing authors were extremely cooperative, and clearly saw the benefit of their work being reproducible. We think this is how all scientific results should be reported. However we recognise several barriers to wider adoption, chiefly expertise and time 3/6.

Blazing-fast image creation – using just your voice. Try Grok Imagine.

Using #RMarkdown we converted standard manuscripts into a fully reproducible format, with all analyses computed from the raw data and results input directly into the text. We learnt a lot along the way, and developed some new methods, including using Docker and Github Actions 2/6.

The question now is what we should do next. We have a really nice pipeline for automatically building markdown files on Github. But the process of making papers reproducible is time consuming.

A bit more about this, in case anyone is interested. Such a cool technique.

Also available from @AmazonUK.

Also available (currently with a 12% discount) from: @blackwellbooks.

Available direct from @OxUniPress .

The book can be preordered from a variety of suppliers (links below), or why not ask your University or public library to get some copies? Plenty of online resources available (code examples, MCQs) for anyone planning to use this in their own teaching 8/8.

Available both in physical and electronic formats, we have deliberately set a low price point (<£30) to help prevent cost being a barrier to access. That said, it is over 350 pages (19 chapters), so definitely no shortage of content! 7/8.

Much of the material has grown from a module I teach to some brilliant third year undergraduates at @YorkPsychology, but I think it would also be very suitable for MSc and PhD students across a wide range of disciplines. It introduces a core research toolkit of useful methods 6/8.

The writing aims to avoid overly technical language and equations, and to be as accessible as possible. Examples are in #rstats, with the code to generate all figures posted to a @github repository. That means the entire book is computationally reproducible 5/8.

There are also chapters on data cleaning, graph plotting, correcting for multiple comparisons, reproducible data analysis, and basic programming skills in #rstats. So quite a lot of #OpenScience relevant material too 4/8.

Topics covered include: meta analysis, power analysis, mixed effects models, multivariate statistics (including machine learning methods like MVPA), Fourier analysis, function optimization, stochastic methods, Bayesian statistics, and signal detection theory 3/8.

I've always felt there was a huge gap between introductory stats texts teaching the basics, and advanced books on a single topic that assume high levels of technical ability. This book aims to bridge that gap, giving a gentle introduction to some advanced statistical methods 2/8.

Next we ran two new experiments, directly comparing four types of mask in the same participants. We used a similar model to estimate group posterior parameter distributions, and found clear evidence for contrast gain control at the occipital pole (Oz) for all masks [6/9]

However for all four types of mask the group posteriors overlapped with a value of 1. This means we lack convincing evidence for an effect of either type of suppression (though there was more evidence for contrast gain than response gain) [5/9].

First we ran a computational meta-analysis of 16 published data sets. Our model had free parameters governing the strength of response gain and contrast gain, both with priors peaking at 1 (no suppression). Posterior parameter intervals for individual studies were often >1 [4/9]

Previous neurophysiology work indicated this might be the case, but the picture was mixed, with results dependent on layer and stimulus properties. Here we used steady-state EEG to compare suppression quantitatively using a Bayesian modelling approach [3/9].

Suppression between visual neurons is ubiquitous, occuring between features like orientation, location and eye of origin. Do the distinct anatomical pathways underlying these processes favour either contrast gain or response gain? [2/9]