Urai AE, Braun A, Donner THD. (2017) Pupil-linked arousal is driven by decision uncertainty and alters serial choice bias. Nature Communications, 8: 14637.

This is the data to replicate all analyses from Urai AE, Braun A, Donner THD. (2017) Pupil-linked arousal is driven by decision uncertainty and alters serial choice bias. Nature Communications, 8: 14637.

See https://github.com/anne-urai/pupilUncertainty for the analysis code, including detailed instructions for replicating the figures from the manuscript.

You probably only want to use 2ifc_data_allsj.csv. Column names:

• stim: generative stimulus identity, i.e. 'weaker' (-1) vs 'stronger' (+1)
• coherence: generative coherence difference from reference
• • Note: a column stim * coherence will get you the signed motion coherence, that is most useful for generating regular choice-based psychometric functions.
• difficulty: level of difficulty within the session; see 'Task and procedure' in the Methods section.
• motionstrength: the estimated motion energy in each generated video. Correlates highly with stim*coherence, as shown in Figure 2c
• resp: response of the subject
• rt: reaction time, after offset of the second stimulus
• correct: feedback given
• correctM: feedback that should have been given, had the exact stimulus identity been based on the motion energy filtered values (rather than the generative values)
• trialnr within each block
• blocknr within each session
• sessionnr within each sj
• subjnr participant ID
• baseline_pupil, decision_pupil, feedback_pupil, trialend_pupil, response_pupil: pupil dilation at various epochs during the trial
• int1motion, int2motion: filtered motion energy values during the reference and test interval. Their difference generates the motionstrength column.
• rtNorm: z-score(log(rt))

correctM: