At the end of the day, if they want to leave many/all cog modellers "behind" and we can just be let alone to do modelling work that is OK with me. I have been doing modelling replications ALL my scientific career. I know what state my (sub*)field is in. *Kleene star
-
-
Thanks! I've learnt so much from twitter that I just wouldn't be exposed to in my small university, I'm grateful to everyone that talks about these issues or I'd still be p hacking having no idea....
-
In modelling we don't have p-values to hack. But we do replicate our work a lot for other reasons — implementation, model, specification, and theory evaluation — that are not discussed in the empirical discussions on replication.
-
This is something I've been explicitly taking about for years, and trying to engage on with empirical people, so I'm glad to see some interest.

-
That sounds kind of cool, so you have 'built in' replications, compared with what seems to happen in my field of needing 'novelty' so moving on to the next study. Thankyou for engaging

-
I'm not sure about built-in. It depends on what you/we mean. The concept of replication is different, arguably quite a bit more complex, in modelling. In addition, a failure to replicate a model is again dealt with in a more complex way too.
-
If I could be so bold at to suggest this as a starting point for more on exactly this: Guest, O. & Rougier, N. P. (2016). Dialogue: What is Computational Reproducibility?. IEEE CIS Newsletter on Cognitive and Developmental Systems. 13 (2). http://oliviaguest.com/doc/guest_rougier_2016.pdf …
- End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.