< Back to Article

A New Framework for Cortico-Striatal Plasticity: Behavioural Theory Meets In Vitro Data at the Reinforcement-Action Interface

Figure 8

Extinction, renewal, and reacquisition.

(A) Summary of relevant data from Figure 3 in [49], (see Methods for interpretation of that data) showing renewal effects after sequence ABA, but not after control sequences AAA and ABB. The points labelled “acquis.” are the performance before extinction in the same context as the renewal test, giving a baseline for the performance change caused only by any switch in context after extinction. In all of (A–C), the blue/black symbols correspond to testing with contexts . (B) Summary of relevant data from [50] (see Figure 2 therein), showing reacquisition of responding in two contexts , after original acquisition in and extinction in . The symbols show endpoints of linear regressions through the original data, which include outcomes at several intermediate time points. (C) Behavioural responses of the basal ganglia model with MSNs initially trained with context . The acquisition (“acquis.”) is tested near the end of the intermission period for two contexts, , derived using different strong-afferent synaptic sets (see text for details). The renewal is tested at the end of 40 trials of extinction under both contexts, leading to the renewal sequences , , and . Reacquisition is measured after 40 learning trials, under each context. (D) Shows (for both D1- and D2-MSNs) the mean AMPA conductance of synaptic sets against trial number, during extinction (trials 1–40), and reacquisition (trials 41–80) under the behavioural protocols in (C). Trials are numbered from trial 80 near the end of the intermission period in the simulated experiment (Figure 7). The trajectory for under extinction with (pale blue line, dark blue symbols) is identical to the extinction shown in Figure 7.

doi: https://doi.org/10.1371/journal.pbio.1002034.g008