The shared files contain learning and degradation challenge data from a contingency degradation task performed by marmosets with excitotoxic lesions of lesions of perigenual anterior cingulate cortex (pgACC), orbitofrontal cortex (OFC) or with sham control lesions (sham). Lesioned marmosets were trained on variable interval schedules in separate sessions to develop two independent action/stimulus-outcome associations in a touchscreen apparatus. The contingency was then partially degraded for one of the action–outcome pairings, but not the other. For one of the actions, the juice that was available noncontingently was the same as the contingent juice and thus, for those sessions the action–outcome contingency was partially degraded (“degraded session”). For the other action, the juice that was available noncontingently was not the same as the contingent juice, and thus by continuing to respond the animal could gain access to two different juices (“nondegraded session”). Data are provided as both Excel (.xlsx) and machine-readable (.csv) files. Word (.docx) and text (.txt) files are provided that describe the data in detail. Data include response numbers for late stage variable interval training. The primary data set includes raw response and reward numbers for each subject for the twenty sessions that comprise the contingency degradation test. Independent variables include subject name, lesion group, session and session pair IDs, contingent and noncontingent reward IDs, and degradation condition. Intermediate calculations, such as natural log transformation of response numbers and ratio score are also included. Additional files contain summary data used in analysis and figures presented in the published manuscript.

This research data supports “Role of the Perigenual Anterior Cingulate and Orbitofrontal Cortex in Contingency Learning in the Marmoset” published in “Cerebral Cortex”.