Figure 2.

CPP Memory Requires Both dCA1 Representations and a Functional dCA1→NAc Pathway

(A) dCA1 PYRs of CamKII-Cre mice were transduced with ArchT-GFP, and their firing activity was monitored with tetrodes. Delivery of yellow light to dCA1 somata (ON^dCA1) or their axons in NAc (ON^dCA1→NAc) was performed using optic fibers inserted above dCA1 or NAc, respectively.

(B) Each day, mice performed a 1-day CPP task and explored an additional enclosure (other). In light ON days, yellow light was delivered during CPP test.

(C) Example paths (top) and 3D occupancy maps (bottom) of a mouse performing the CPP task. CPP score equals time in sucrose-paired enclosure (+Suc.), minus that in water-paired enclosure (+Wat.), divided by the sum. A negative score indicates that the mouse spent less time in the enclosure paired with sucrose during conditioning. Note that during test, the mouse successfully changed its preference for the enclosure that was paired with sucrose, as indicated by the positive score.

(D) Examples of dCA1 assembly representations of the CPP apparatus. For clarity, each assembly is depicted in the form of a weight vector where PYRs are ordered and color-coded in the lollipop plots to highlight those neurons with high coincidental spiking. Shown are the weight vectors (top-left) of the detected assembly patterns, the corresponding assembly maps (bottom-left; numbers indicate assembly activation rate), a sample raster plot of the spike trains (one PYR per row; top-right) and the corresponding activation strength time course of the detected assembly patterns (bottom-right; highlighting activation peaks of the red assembly). See also Figure S2.

(E, H, and K) CPP score (mean ± SEM; n = 81 days from 21 mice) during pre-test and test (E: n = 48 light OFF; H: 12 ON^dCA1; K: 21 ON^dCA1→NAc). Light was delivered during CPP test (H: 20 × 30-s pulses over 15 min, K: one 10-min pulse).

(F, I, and L) Example dCA1 assembly similarity matrices from 3 recording days. For clarity, the same number of assemblies is represented across those matrices. Note that in both light OFF (F) and ON^dCA1→NAc (L) conditions, the CPP assemblies were reinstated during test, but not during the exposure to the other (CPP-unrelated) enclosure, thus showing the context-dependent reinstatement of dCA1 representations. Reinstatement of dCA1 CPP representations was suppressed during ON^dCA1 test (I; see also Figure S3).

(G, J, and M) Activation strength of dCA1 CPP assemblies (mean ± SEM; n = 209 total detected assembly patterns; 11.61 ± 1.36 patterns per day; 1,176 PYRs total recorded from eight mice; 65.33 ± 7.91 PYRs per mouse recording day). Each pair of connected data points shows the average strength of a CPP assembly pattern tracked during other and test (G: n = 95 light OFF assemblies from 493 PYRs; J: 37 ON^dCA1 assemblies from 227 PYRs; M: 77 ON^dCA1→NAc assemblies from 456 PYRs). ^∗∗∗p < 0.001, paired t tests.

See also Figure S4.