A different recruitment of the lateral and basolateral amygdala promotes contextual or elemental conditioned association in Pavlovian fear conditioning

Abstract

Convergent data suggest dissociated roles for the lateral (LA) and basolateral (BLA) amygdaloid nuclei in fear conditioning, depending on whether a discrete conditioned stimulus (CS)-unconditional stimulus (US) or context-US association is considered. Here, we show that pretraining inactivation of the BLA selectively impaired conditioning to context. In contrast, inactivation of the LA disrupted conditioning to the discrete tone CS, but also either impaired or enhanced contextual conditioning, depending on whether the context was in the foreground or in the background. Hence, these findings refine the current model of the amygdala function in emotional learning by showing that the BLA and the LA not only differentially contribute to elemental and context-US association, but also promote, through their interaction, the most relevant of these two associations.

Figure 1.

Histological controls. (A) Illustration of the injectors' tip location within the LA (left) and the BLA (right). Numbers indicate relative position of the coronal sections in millimeters posterior to bregma (adapted with permission from Elsevier © 1997, Franklin and Paxinos 1997). (BLA) Basolateral nucleus; (BMA) basomedial nucleus, anterior; (BMP) basomedial nucleus, posterior; (LA) lateral nucleus; (CE) central nucleus. (B) Representative photomicrograph depicting the cannulae placement into the LA (left) and the BLA (right). (C) Schematic drawing of coronal sections of a mouse brain hemisphere (–1.94 [left] and –1.58 [right] posterior to bregma), showing the sphere of diffusion of a 0.15-μL solution of India ink into the LA (left) or the BLA (right) under the conditions as detailed in Materials and Methods. For each coronal section, the gray area (right part of the section) represents the superimposed sphere of diffusion in a sample of four animals that were either LA- (left) or BLA- (right) infused mice.

Figure 2.

Behavioral procedure. For acquisition of fear conditioning, animals were either submitted to an auditory-cue conditioning, which consists in tone CS-footshock–US pairing (background contextual conditioning) (A), or to a CS–US unpairing (foreground contextual conditioning) (B). Then, animals were replaced in their home cages. Twenty-four hours later, each animal was first submitted to the auditory-cue test while maintained in its home cage, then (2 h later) was re-exposed to the conditioning context for the context test. During each of these tests, the amounts of freezing were assessed by 2-min blocks.

Figure 3.

Auditory-cue test. Mean percentage freezing (±SEM) on successive 2-min blocks before and during tone presentation in animals that were submitted to lidocaine infusion (n = 32), NaCl (n = 31), or pseudo-infusion (n = 33) into the BLA (A) or into the LA (B), before either CS–US pairing or CS–US unpairing.

Figure 4.

Context test. Mean percentage freezing (±SEM) on successive 2-min blocks during re-exposure to the conditioning context in animals that were submitted to lidocaine, NaCl, or pseudo-infusion into the BLA (A) or into the LA (B), before either CS–US pairing or CS–US unpairing.