This data set was compiled for the experiments detailed in Enel, P., Wallis, J., & Rich, E. (2019). Stable and dynamic representations of value in the prefrontal cortex. Elife 9 (2020): e54313 10.7554/eLife.54313

Optimal decision-making requires that stimulus-value associations are kept up to date by constantly comparing the expected value of a stimulus with its experienced outcome. To do this, value information must be held in mind when a stimulus and outcome are separated in time. However, little is known about the neural mechanisms of working memory (WM) for value. Contradicting theories have suggested WM requires either persistent or transient neuronal activity, with stable or dynamic representations respectively. To test these hypotheses, we recorded neuronal activity in the orbitofrontal and anterior cingulate cortex of two monkeys performing a valuation task. We found that features of all hypotheses were simultaneously present in prefrontal activity, and no single hypothesis was exclusively supported. Instead, mixed dynamics supported robust, time invariant value representations while also encoding the information in a temporally specific manner. We suggest that this hybrid coding is a critical mechanism supporting flexible cognitive abilities.

Monkeys were first trained a value based decision making task in which a cue predicted a delayed reward. A detailed explanation of the task and first analysis of the data with decision trials can be found in Rich, E. L., & Wallis, J. D. (2016). Decoding subjective decisions from orbitofrontal cortex. Nature Neuroscience, 19(7), 973–980 10.1038/nn.4320. This data set only contains the data of single cue in which the animal was looking at a reward predicting cue, performed an unrelated joystick reponse task during a delay, and received the reward associated with the cue. Monkeys were head fixed, while single electrodes were lowered in the OFC (areas 11 and 13) and the dorsal bank of the ACC sulcus (area 24). All well isolated neurons were recorded without discrimination. Neurons were filtered out if there firing rate over the session was lower than 1Hz. This data set contains both single and multi units without distinction. The firing rate was estimated either with gaussian or square window, and with different parameters:

- type of smoothing: either gaussian or square

- the size of the smoothing window: either 100ms or 50ms

- the step between each bin: 25ms, 50ms or 200ms

- the alignment of the bins: either center to the original timing or to the end of the bin. When centered to the end of the bin, only data before the timing of the bin is included in the bin.

Pseudo-population data sets are generated with a specific seed to match the number of trials in each condition across monkey and recording session.

This data set contains the smoothed and binned firing rate of single and multi units of orbitofrontal and anterior cingulate cortex neurons of macaque monkeys performing a value based decision making task. Different methods were used to estimate the firing rate of neurons, aimed at specific analyses in Enel, P., Wallis, J., & Rich, E. (2019). Stable and dynamic representations of value in the prefrontal cortex. Elife 9 (2020): e54313 10.7554/eLife.54313. The data set is saved in three python pickle files. Pseudo-randomly generated pseudo-population data sets can be obtained from these unit data set files with an accompanying python script.

Enel, Pierre, Joni D. Wallis, and Erin L. Rich. "Stable and dynamic representations of value in the prefrontal cortex." Elife 9 (2020): e54313.

Each file contains the unit data set in nested python containers (python dictionary or list). The details of the data structure can be found in the ReadMe file.

Pseudo-population data sets can be generated with the same seeds used in Enel et al. (2019) by using the generate_population_dataset.py script. A list of python package dependency can be found in the ReadMe file.