Characterizing Different Strategies for Resolving Approach-Avoidance Conflict

Abstract

The ability of animals to maximize benefits and minimize costs during approach-avoidance conflicts is an important evolutionary tool, but little is known about the emergence of specific strategies for conflict resolution. Accordingly, we developed a simple approach-avoidance conflict task in rats that pits the motivation to press a lever for sucrose against the motivation to step onto a distant platform to avoid a footshock delivered at the end of a 30 s tone (sucrose is available only during the tone). Rats received conflict training for 16 days to give them a chance to optimize their strategy by learning to properly time the expression of both behaviors across the tone. Rats unexpectedly separated into three distinct subgroups: those pressing early in the tone and avoiding later (Timers, 49%); those avoiding throughout the tone (Avoidance-preferring, 32%); and those pressing throughout the tone (Approach-preferring, 19%). The immediate early gene cFos revealed that Timers showed increased activity in the ventral striatum and midline thalamus relative to the other two subgroups, Avoidance-preferring rats showed increased activity in the amygdala, and Approach-preferring rats showed decreased activity in the prefrontal cortex. This pattern is consistent with low fear and high behavioral flexibility in Timers, suggesting the potential of this task to reveal the neural mechanisms of conflict resolution.

Keywords: PVT; accumbens; amygdala; individual differences; prefrontal.

FIGURE 1

Approach-avoidance conflict training. (A) Rats were given 10 days of avoidance conditioning followed by 3 days of reward conditioning and 16 days of conflict training. (B) As rats learned avoidance conditioning, they increased the time spent on the platform (blue line), while reducing pressing during the tone (relative to the baseline established during the first tone) (green line). The pressing during the intertrial interval (ITI) (ITI = 1 min prior to each trial) (orange line) returned to preconditioning baseline (BL) by the end of avoidance conditioning. During reward conditioning, rats decreased their pressing during the ITI and increased their pressing during the light (green line). When the tone-light were co-presented in conflict training, rats gradually reduced the time spent on platform and gradually increased their pressing.

FIGURE 2

Rats showed three distinct strategies for resolving approach-avoidance conflict. (A) Criteria for separating rats into different conflict strategies. Rats were separated into three subgroups: Avoidance-preferring (22/69 32%, red), Approach-preferring (13/69, 19%, green), and Timers (34/69, 49%, yellow). (B) Heat maps showing the location of rats during the conflict test (n = 7 for each subgroup). (C) Presses/min vs. % time pressing for individual rats at conflict test trial 1, showing that the three subgroups were distinct from each other. Inset shows averaged data for all three trials of conflict test. (D) Actogram depicting each rat’s behavior during conflict test (bin = 1 s) across the conflict test. Subgroups are separated by horizontal dashed lines. (E) Average time spent on platform for each subgroup. (F) Averaged behaviors for each subgroup prior to, and following, conflict training. Subgroup differences were not apparent prior to conflict training. (G) Average time spent avoiding (square) and pressing (circle) during conflict training, for each subgroup. **P < 0.01.

FIGURE 3

Approach-preferring rats showed reduced social exploration. (A) The three subgroups showed no differences in the % of time exploring the center of the open field, a measure of innate fear. (B) Experimental setup for the exploration task in which rats could explore the outside of a novel cage that was first empty and later contained a demonstrator rat. (C) Placing the empty cage in the center of the open field increased the time spent exploring near the cage in all three subgroups. Adding a demonstrator rat to the empty cage further increased the time spent exploring for Avoid-pref. and Timer subgroups, but not in the Approach-pref. subgroup. Avoid-pref., n = 23; Timers, n = 24; Approach-pref., n = 10 (Student’s t-tests *p < 0.05, Bonferroni corrected).

FIGURE 4

The three subgroups showed distinct neuronal activity patterns. (A) Average presses/min vs. % time on platform for the three trials at conflict test in rats included in the cFos analysis. Only rats with consistent subgroup behavior were selected for cFos analysis. (B) Rats were euthanized 90 min after the conflict test and processed for cFos. Avoid-pref. Rats showed elevated activity in the lateral (LA) and centromedial (CeM) subregions of amygdala, relative to the other two subgroups. (C) Timer rats showed elevated activity in the nucleus accumbens-core (NAcC), and reduced activity in the paraventricularthalamus (PVT), relative to the other two groups. (D) Approach-pref. rats showed reduced activity in the prelimbic (PL) and infralimbic (IL) subregions of medial prefrontal cortex, relative to the other two subgroups (one-way ANOVA*p < 0.05, **p < 0.01, ***p < 0.001 Tukey post-hoc test). (E) Activity within CeM and PVT separated the three subgroups, consistent with low levels of fear and high levels of behavioral flexibility in Timer rats.