Incentive salience attribution, "sensation-seeking" and "novelty-seeking" are independent traits in a large sample of male and female heterogeneous stock rats

Abstract

There are a number of traits that are thought to increase susceptibility to addiction, and some of these are modeled in preclinical studies. For example, "sensation-seeking" is predictive of the initial propensity to take drugs; whereas "novelty-seeking" predicts compulsive drug-seeking behavior. In addition, the propensity to attribute incentive salience to reward cues can predict the propensity to approach drug cues, and reinstatement or relapse, even after relatively brief periods of drug exposure. The question addressed here is the extent to which these three 'vulnerability factors' are related; that is, predictive of one another. Some relationships have been reported in small samples, but here a large sample of 1,598 outbred male and female heterogeneous stock rats were screened for Pavlovian conditioned approach behavior (to obtain an index of incentive salience attribution; 'sign-tracking'), and subsequently tested for sensation-seeking and novelty-seeking. Despite the large N there were no significant correlations between these traits, in either males or females. There were, however, novel relationships between multiple measures of incentive salience attribution and, based on these findings, we generated a new metric that captures "incentive value". Furthermore, there were sex differences on measures of incentive salience attribution and sensation-seeking behavior that were not previously apparent.

Figure 1

Notched box plot summary. An example of a notched box plot with labels for each informational aspect as described in the text. SEM, standard error of the mean.

Figure 2

Pavlovian conditioned approach Index score in male and female HS rats. A Pavlovian conditioned approach (PCA) Index score was calculated for each rat as described in the text. A score of −1 indicates exclusive food cup-oriented behavior and a score of +1 indicates exclusive lever-oriented behavior. (a) PCA Index mean ± SEM for each of 5 training sessions for female (n = 799) and male (n = 799) rats. (b) Notched box plot of terminal PCA Index, calculated as the average of the PCA Index from sessions 4 and 5, for male and female rats. (c) Histogram of the distribution of terminal PCA Index for females and males. Horizontal hashed lines indicate the threshold of PCA Index used to define phenotype groups (GT, −5–0; intermediate responders −0.5– + 0.5; STs, +0.5–1).

Figure 3

Sign-tracking and goal-tracking behavior. Behavior displayed for the 5 sessions of PCA training for each phenotype of each sex (Female GT (n = 160), IN (n = 294), ST (n = 345); Male GT (n = 303), IN (n = 231), ST (n = 265)) as mean ± SEM of the following metrics: (a) probability to deflect the lever; (b) probability to enter the food cup; (c) number of lever deflections; (d) number of food cup entries; (e) latency to deflect the lever, and (f) latency to enter the food cup.

Figure 4

Conditioned reinforcement. (a) Notched box plot of the number of nose pokes into the active and inactive ports for both female (n = 795) and male (n = 795–796) rats. (b) Notched box plot of nose pokes into the active port minus nose pokes into the inactive port for females (GT (n = 160), IN (n = 293), ST (n = 341)) and males (GT (n = 302), IN (n = 227), ST (n = 264)). (c) Lever deflections during the conditioned reinforcement test for females (GT (n = 160), IN (n = 293), ST (n = 341)) and males (GT (n = 302), IN (n = 228), ST (n = 264)). (d) Notched box plot of Incentive Value Index for females (GT (n = 160), IN (n = 293), ST (n = 341)) and males (GT (n = 302), IN (n = 227), ST (n = 264)). (e) Plot of each individual rat’s data for Active minus Inactive (A-I) value as a function of terminal PCA Index with linear regression line. (f) Plot of each individual rat’s data for lever deflections as a function of terminal PCA Index with linear regression line. (g) Plot of each individual rat’s Incentive Value Index as a function of terminal PCA Index with linear regression line.

Figure 5

Locomotor response to novelty. Novelty-induced locomotor activity assessed on the first habituation day of the novelty-seeking paradigm. (a) Notched box plot of total distance (meters) travelled for females (GT (n = 152), IN (n = 280), ST (n = 332)) and males (GT (n = 292), IN (n = 222), ST (n = 252)). (b) Plot of each individual rat’s data for total distance travelled during habituation as a function of terminal PCA Index with linear regression line.

Figure 6

Novelty-seeking behavior. (a) Novelty-seeking paradigm timeline with schematics of the apparatus on habituation and test days. (b) Notched box plot of time spent (in seconds) in the familiar and novel zone for females (n = 761) and males (n = 760). (c) Notched box plot of the percent of time spent in the novel zone for females (GT (n = 152), IN (n = 278), ST (n = 330)) and males (GT (n = 287), IN (n = 221), ST (n = 250)). (d) Data for each individual rat’s percent time spent in the novel zone as a function of terminal PCA Index with linear regression line.

Figure 7

Principal components in rotated space. Principal components analysis of the relationship between: PCA Index Score, Incentive Value Index, Sensation Seeking, and Novelty Seeking. The two extracted factors cumulatively explained ~63% of the variance in behavior. PCA Index Score and Incentive Value Index loaded strongly (>0.7) onto Factor 1, which accounts for 38% of the total variance. Additionally, Sensation Seeking loaded onto this factor to a lesser degree (0.44). Factor 2, which accounts for 25% of the variance, is comprised largely of Novelty Seeking (0.98).