[ACT-R-users] Updated ACT-R software

[Dan Bothell]

A new version of the ACT-R software is now available on the prerelease
page of the ACT-R web site:

<http://act-r.psy.cmu.edu/next/>

It will be moved to the current version soon.
Here are the notable changes and additions:
- The remote interface now has the option to switch from a TCP/IP socket
   to Named Pipes when using macOS or Linux.  The Python interface
   provided with the tutorial was updated to use that, and on average,
   running the tutorial tasks are about 26% faster with the Named Pipes
   on my mac.

- The speech module has a new parameter called :hear-self.  If it is set
   to nil then the model's vocal action output is not relayed to the audio
   module.  If a model doesn't need to hear itself, turning this off can
   simplify the need for a model to filter its own speech acts from the
   task audio. It also provides a performance improvement for models that
   have a lot of vocal output (the Zbrodoff task in the tutorial runs about
   10% faster with this turned off).

- The trace information for some buffer setting and modifying actions now
   provide more information. Actions that looked like this:

     0.435   IMAGINAL     SET-BUFFER-CHUNK-FROM-SPEC IMAGINAL
   Now look like this:
     0.435   IMAGINAL     SET-BUFFER-CHUNK IMAGINAL (LETTER "V")
If you have any questions or problems with this version let me know.

Dan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/act-r-users/attachments/20260617/04316700/attachment.html>