(function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); skip to primary navigationskip to content

PR Task

Probabilistic Reversal Task

Download a Windows XP version of the Probabilistic Reversal task (neuropsych version with a single reversal) here

Unpack to a folder on your Desktop (e.g. ProbRevXP) containing the directories "timing", "stimuli", "data", "modules" and "instructions". You can start the task by running Prob_Rev.exe from the main directory.

If you haven't already installed Rhodri's Scannersync files, you will need to do this (by running Scannersync Setup in the timing folder). You might also have to copy the file 'perfpres.dll' into c:\Windows\System

There is currently one parallel version in the zip file (Prob_Rev_parallel.exe) which presents different coloured stimuli. The feedback sequence is currently identical to the Prob_rev.exe task, but you can tinker with the files 'trials2.txt' in Notepad to alter this sequence. [The columns in the text file refer to Trial number, stimulus 1 position (0 top, 1 right, 2 bottom, 3 left), stimulus 2 position, stimulus 1 feedback (1=correct, 2=wrong), stimulus 2 feedback, stimulus 1 identity (pictureX.bmp where X=1-4), stimulus 2 identity; NB stimulus 1 feedback is the feedback to whichever stimulus they select on the first trial, so the first trial is always right]. I can make more parallel versions very quickly if they are useful.

Task instructions here

A revised macro for the XP version (July 2008) here, which extracts trials to criterion as well as errors. This macro also enables you to calculate the 'Maintenance Failure Score' and the 'Probability Matching Score' from Murphy et al (2003), although you have to compute these by hand, to avoid 'divide by zero' errors when snaffling.

An identical macro for the Advantech version (June 2006) with trials to criterion extraction here.

NB. The task macros might not work if your Windows settings are non-British. We have had trouble with German-set computers which use commas differently, and this causes problems with comma-delimited output files.


The pre-scan Practice session for the fMRI task is available here; this is sized for a paceblade.

A macro for behavioural analysis of the scandata from the fMRI task here


Key papers: Cools R, Barker RA, Sahakian BJ, Robbins TW. Enhanced or impaired cognitive function in Parkinson's disease as a function of dopaminergic medication and task demands. Cereb Cortex. 2001 Dec;11(12):1136-43.

Chamberlain SR, Muller U, Blackwell AD, Clark L, Robbins TW, Sahakian BJ. Double dissociation between noradrenergic and serotonergic modulation of response inhibition and probabilistic learning in man. Science 2006, 311: 861-3.

Murphy FC, Michael A, Robbins TW, Sahakian BJ (2003). Neuropsychological impairment in patients with major depressive disorder: the effects of feedback on task performance. Psychol Med 33(3): 455-467.

Taylor Tavares JV, Clark L, Furey M, Williams GB, Sahakian BJ, Drevets W (2008). Neural basis of abnormal response to negative feedback in mood disorders. Neuroimage in press.