The developmental inter-relationships between activity, novelty preferences, and delay discounting in male and female rats

Abstract

Increased locomotion, novelty-seeking, and impulsivity are risk factors associated with substance use. In this study, the inter-relationships between activity, novelty preferences, and delay discounting, a measure of impulsivity, were examined across three stages: juvenile/early adolescence (postnatal Day [P] 15, 19, and 42 for activity, novelty, and impulsivity, respectively), adolescent/late adolescent (P28, 32, 73), and adult (P90, 94, 137) in male and female rats. Our estimates of impulsive choice, where animals were trained to criterion, revealed an age × sex interaction where early adolescent females had the lowest levels of impulsivity. The relationships of activity and novelty to impulsivity significantly changed across age within each sex. Early adolescent males with high activity, but low novelty preferences, were more impulsive; however, low activity and high novelty preferences were related to high impulsivity in adult males. Female activity gradually increased across age, but did not show a strong relationship with impulsivity. Novelty preferences are moderately related to impulsivity into adulthood in females. These data show that males and females have different developmental trajectories for these behaviors. Males show greater sensation-seeking (e.g., activity) and risky behavior (e.g., novelty preferences) earlier in life, whereas these behaviors emerge during adolescence in females.

Keywords: delay discounting; development; impulsivity; novelty; rat; sex differences.

FIGURE 1

Timeline showing the ages of testing of subjects. Activity (Act) and novelty preferences (Nov) were single day tests, whereas the Mean ± SE is presented for the final age of testing for impulsivity. The solid arrow represents males and the open arrow represents females. Age and Sex differences were present (statistics in the text).

FIGURE 2

The training and testing paradigm for delay discounting. Three different training phases were conducted, with different symbol presentation and pellet reinforcers dispensed as shown. The fourth phase tested subjects for 3 consecutive days or until responding stabilized (<15% variability across all delays).

FIGURE 3

Age differences between activity levels and novelty preferences for males and females. Mean ± SE presented. *P28 subjects significantly differed from P15 rats (p = .003) and P90 (p = .02)

FIGURE 4

Correlations between activity, a measure of sensation-seeking behavior, and novelty preferences, a measure of risk-taking behavior, with all subjects included on the left; males only on the right. A significant correlation was observed for the male only analysis.

FIGURE 5

The amount of locomotor activity in response to a novel environment versus the delay indifference point when 50% of the large reinforcers are received (DDT₅₀). Each age group and both sexes are presented with their individual Pearson R-values showing the strength of the relationship and significance level. Cross-age comparisons of the different relationships within each sex revealed that P15 males differ from P28 to 90 (p < .05), and P28 differs from P90 (p < .001). The female activity/DDT50 relationship differed between P15 and P28 (p < .05) and between P28 and P90 (p < .05).

FIGURE 6

Time spent on the novel side versus DDT₅₀. Each age group and both sexes are presented with their individual Pearson R-values showing the strength of the relationship and significance level. Cross-age comparisons of the different relationships within each sex revealed that P19 males differ from P32 to 90 (p < .05), and P32 differs from P90 (p < .001). Female P32 rats only differed from P19 subjects (p < .05).

FIGURE 7

Delay discounting curves as a function of age. Following a significant interaction of Age × Sex × [delay], males and females were analyzed separated. Data presented here are covaried for activity levels for males, but not for females as activity was not a significant covariate. Means ± SE presented. *Shows that early adolescents (both sexes) were significantly different from late adolescents and adults at all delays in males, but only at the 40 s delay in females; p < .05, Bonferroni corrected.