Figure 3. Aversive conditioning rapidly induces excitatory responses to aversion-predicting cues and omitting an unconditioned aversive stimulus slowly extinguishes previously-conditioned responses.

(A) Heatmap representation of LHb Ca²⁺ transients within a session of cue-quinine Pavlovian conditioning. The conditioning session consisted of 20 trials. The dashed lines and timeline below indicate the timing of an auditory cue (2 s), delay (2 s), and intra-oral infusion of quinine (0.5 s). (B) The peri-event plot of the average Ca²⁺ transient from the same mouse shown in (A) during the first five trials (black) and last 5 trials of the conditioning session. (C) Mean Ca²⁺ transient for the entire test group (n = 9 mice). (D) Sum of Ca²⁺ transients for cues (0–2 s; blue line) and quinine infusion (4.0–4.5 s; red line) throughout the conditioning process. (E–H) LHb neurons rapidly gained responses to an auditory cue after its coupling to footshock. (E) Heatmap representation of Ca²⁺ transients during a conditioning session (n = 20 trials). (F) Mean Ca²⁺ transients across the conditioning trials for the same mouse shown in (E). (G) Mean Ca²⁺ transients for the entire test group (n = 9 mice). (H) Ca²⁺ responses to the auditory cue (0–2 s, blue line) increase, whereas those to the footshock (4.0–4.5 s, red line) remain largely stable during the conditioning phase (n = 9 mice). (I–L) The effects of omitting footshock on previously conditioned responses to the footshock-predicting cue. (I) Heatmap representation of Ca²⁺ transients in an extinction session (30 trials), within which we repetitively presented 30 CS cues but omitted footshock. (J) Mean Ca²⁺ transients for one extinction session. (I and J) correspond to the same mouse in (E and F). (K) Population mean of Ca²⁺ transients (n = 9 mice). Thick lines indicate the mean and shaded areas indicate the SEM. Red segments indicate statistically-significant increases from the baseline (p<0.05; multivariate permutation test). (L) Sum of Ca²⁺ transients during cue presentation (0–2 s; blue line) and footshock omission (4–4.5 s; red line). Each data point represents the average of 5 consecutive trials. (In D, H, L), *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001; n.s., not significant; nonparametric one-way ANOVA with Geisser-Greenhouse correction for the difference between the first data point and those of the following trials.

DOI: http://dx.doi.org/10.7554/eLife.23045.007

Figure 3—figure supplement 1. Cue-footshock associative learning changes animal locomotor behavior.

(A) Heatmap representation of the locomotor speed of a mouse shown during a conditioning session. Same mouse as shown in Figure 3E and Figure 3I. (B) At the group level, animal locomotor speed changed during the footshock-predicting period (0–4 s after cue onset) across conditioning trials. Each data point represents the average of 5 consecutive trials. (C) Heatmap representation of the locomotor speed of the same mouse shown in (A) in an extinction session. (D) Population locomotor speed following cue presentation (0–4 s following cue onset) across extinction trials. Each data point represents the average of 5 consecutive trials. The speed was measured as the rate of body position change. (In B and D), *p<0.05; **p<0.01; n.s., not significant; nonparametric one-way ANOVA with Geisser-Greenhouse correction for the difference between the first data point and those in the following trials; n = 5 mice.

DOI: http://dx.doi.org/10.7554/eLife.23045.008