A Characterization of Approach and Avoidance Learning in Alcohol-Preferring and Alcohol-Nonpreferring Rats

Evaluating habit formation across pairs of female and male selectively bred alcohol-preferring and non-preferring rats

Some individuals with alcohol use disorder (AUD) continue to drink because they have developed a habit where they do not consider the consequences of their actions. Genetically selected lines of alcohol-preferring and non-preferring rats allow for exploration of how specific endophenotypes, such as tendency to form habits, may be risk factors that interact with a genetic predisposition of AUD. While high alcohol drinking (HAD) and alcohol-preferring (P) rats were selectively bred to consume high amounts of freely available ethanol, they exhibit differences in alcohol-seeking behaviors as well as impulsive behaviors, and may represent different behavioral models of AUD. The goal of the current study was to compare the tendency to develop habitual behaviors across female and male HAD1, HAD2, and P rats and their respective alcohol non-preferring counterparts. Alcohol-naïve rats were trained on a variable interval schedule using a non-ethanol reinforcer and were then tested in two extinction sessions, one prior to a reinforcer devaluation (conditioned taste aversion) procedure and one after. There were no differences in total lever presses between P and alcohol non-preferring (NP) rats, but there were differences between HAD and low-alcohol drinking (LAD) rats. All six strains decreased lever pressing after reinforcer devaluation. However, P and NP females did not increase latency to first lever press after devaluation, suggesting some inclination toward habitual behavior that was not apparent in either the HAD or LAD lines. Selective breeding for alcohol preference does not seem to influence the tendency to form habits, whereas background strain and sex may have an influence on this behavior.

Alcohol-Preferring P Rats Exhibit Elevated Motor Impulsivity Concomitant with Operant Responding and Self-Administration of Alcohol

BACKGROUND

Increased levels of impulsivity are associated with increased illicit drug use and alcoholism. Previous research in our laboratory has shown that increased levels of delay discounting (a decision-making form of impulsivity) are related to appetitive processes governing alcohol self-administration as opposed to purely consummatory processes. Specifically, the high-seeking/high-drinking alcohol-preferring P rats showed increased delay discounting compared to nonselected Long Evans rats (LE) whereas the high-drinking/moderate-seeking HAD2 rats did not. The P rats also displayed a perseverative pattern of behavior such that during operant alcohol self-administration they exhibited greater resistance to extinction.

METHODS

One explanation for the previous findings is that P rats have a deficit in response inhibition. This study followed up on this possibility by utilizing a countermanding paradigm (stop signal reaction time [SSRT] task) followed by operant self-administration of alcohol across increasing fixed ratio requirements (FR; 1, 2, 5, 10, and 15 responses). In separate animals, 24-hour access 2-bottle choice (10% EtOH vs. water) drinking was assessed.

RESULTS

In the SSRT task, P rats exhibited an increased SSRT compared to both LE and HAD2 rats indicating a decrease in behavioral inhibition in the P rats. Also, P rats showed increased operant self-administration across all FRs and the greatest increase in responding with increasing FR requirements. Conversely, the HAD2 and LE had shorter SSRTs and lower levels of operant alcohol self-administration. However, for 2-bottle choice drinking HAD2s and P rats consumed more EtOH and had a greater preference for EtOH compared to LE.

CONCLUSIONS

These data extend previous findings showing the P rats to have increased delay discounting (decision-making impulsivity) and suggest that P rats also have a lack of behavioral inhibition (motor impulsivity). This supports the notion that P rats are a highly impulsive as well as "high-seeking" model of alcoholism, and that the HAD2s' elevated levels of alcohol consumption are not mediated via appetitive processes or impulsivity.

Amygdalar unit activity during three learning tasks: eyeblink classical conditioning, Pavlovian fear conditioning, and signaled avoidance conditioning

Neural activity in central and basolateral amygdala nuclei (CeA and BLA, respectively) was recorded during delay eyeblink conditioning, Pavlovian fear conditioning, and signaled barpress avoidance. During paired training, the CeA exhibited robust learning-related excitatory activity during all 3 tasks. By contrast, the BLA exhibited minimal activity during eyeblink conditioning, while demonstrating pronounced increases in learning-related excitatory responsiveness during fear conditioning and barpress avoidance. In addition, the relative amount of amygdalar activation observed appeared to be related to the relative intensity of the unconditioned stimulus and somatic requirements of the task. Results suggest the CeA mediates the Pavlovian association between sensory stimuli and the BLA mediates the modulation of instrumental responding through the assignment of motivational value to the unconditioned stimulus.

Pain thresholds in alcohol preferring and non-preferring rats: diurnal and repeated trial line differences

BACKGROUND

Studies in humans and animals suggest that those with a genetic predisposition toward high levels of alcohol consumption have decreased pain thresholds. The present study explores this association in rat lines selectively bred for alcohol preference or nonpreference.

METHODS

13 HAD1, 11 LAD1, 16 iP5C, and 16 iNP1 rats were tested in the Ugo Basile Plantar Analgesiometer and the IITC Tail Flick device. Five trials were conducted in each test session with one session per device conducted in the light and one in the dark. Trial one latency represents baseline latency. Decline in Latency over Repeated Trials (DLRT) was analyzed using linear regression.

RESULTS

Testing in the Plantar Analgesiometer revealed no significant line differences in baseline latency (iP5C versus iNP1; HAD1 vs. LAD1). The alcohol preferring lines (iP5C and HAD1), however, both demonstrated an increase in baseline latency in the dark phase compared to the light phase (p < 0.05). The iNP1 line demonstrated highly significant DLRT in both the light phase (p < 0.001) and dark phase (p < 0.01) while the iP5C line demonstrated a significant DLRT only at night (p < 0.01). In the tail flick apparatus, the HAD1 line demonstrated a significantly increased baseline latency compared to the LAD1 line in both the light (p < 0.05) and dark (p < 0.01) phases. The HAD1 line also demonstrated a significant increase in baseline latency in the dark compared to the light phase (p < 0.05) and a significant DLRT in the light (p < 0.01) and the dark phase (p < 0.001).

CONCLUSIONS

The alcohol preferring HAD1 line demonstrates a significantly increased baseline tail latency compared to the nonpreferring LAD1 line in both light and dark phases. There is a significant diurnal rhythm of paw and tail latency in both of the alcohol preferring lines (iP5C and HAD1) but not in the nonpreferring lines (iNP1 and LAD1). A novel finding of Decline in Latency with Repeated Trials (DLRT) was found in both alcohol preferring and nonpreferring lines; degree of DLRT differed as a function of line and circadian period.

High-alcohol-drinking rats exhibit persistent freezing responses to discrete cues following Pavlovian fear conditioning

We previously reported that high-alcohol-drinking (HAD) rats exhibited selective deficits in active avoidance learning and that those deficits were partially reversed by moderate doses of ethanol under certain training conditions [Pharmacol. Biochem. Behav. 75 (2003) 89]. In that study, we hypothesized that HAD deficits resulted from exaggerated fear in the conditioning context and that the anxiolytic properties of ethanol, along with prior exposure to the conditioning apparatus, were responsible for the facilitated avoidance learning that was observed in HAD rats following moderate doses of ethanol. The current study was designed to test whether HAD rats exhibit behaviors consistent with increased fear in aversive learning contexts. We used a standard Pavlovian fear conditioning paradigm to assess behavioral freezing in HAD (HAD-1 and HAD-2) and low-alcohol-drinking (LAD; LAD-1 and LAD-2) rats. No significant differences were observed between HAD-1 and HAD-2 or between LAD-1 and LAD-2 rats, indicating that the replicate lines performed similarly in this study. Both HAD and LAD rats exhibited robust fear conditioning during training. Although no differences were observed between HAD and LAD rats during fear training, HAD rats failed to extinguish freezing behavior in response to the discrete tone conditional stimulus during subsequent fear retention tests. Thus, HAD rats demonstrated prolonged cue-elicited fear that was resistant to extinction.

Moderate doses of ethanol partially reverse avoidance learning deficits in high-alcohol-drinking rats

We previously reported that ethanol-naive high-alcohol-drinking (HAD1 and HAD2) rats exhibited selective deficits in active avoidance learning, as compared to low-alcohol-drinking (LAD1 and LAD2) rats, in a signaled bar-pressing task [Alcohol. Clin. Exp. Res. 24 (2000) 1778]. In the current study, we used appetitive and aversive learning tasks to assess whether administration of ethanol influences approach and avoidance learning in HAD and LAD rats. Rats were administered 0.0, 0.5, 1.0, or 1.5 g ethanol/kg body weight during appetitive and aversive conditioning sessions. We found that ethanol impaired acquisition of the appetitive conditioned response in a dose-dependent manner in both HAD and LAD rats, with 1.5 g/kg ethanol producing the greatest deficits. Notably, moderate doses of ethanol (0.5 and 1.0 g/kg) partially reversed avoidance learning deficits in HAD rats, but only when appetitive conditioning preceded aversive conditioning. The highest dose (1.5 g/kg EtOH) abolished avoidance responding altogether in HAD rats. Avoidance responding in LAD rats was not affected by any dose of ethanol. These results are consistent with previous studies suggesting that alcohol preference may be associated with increased fear or anxiety, but the conditions under which ethanol produces a reduction of fear and anxiety in HAD rats appear to be relatively complex.

Evaluation of behavioral disinhibition in P/NP and HAD1/LAD1 rats

1. Two lines of rats specifically bred for alcohol preference were exposed to two different behavioral tasks that required behavioral inhibition to successfully solve. 2. Learning and performance of a step-down passive avoidance task and a differential reinforcement of low-rate responding task were studied in P/NP and HAD1/LAD1 rats. 3. While the P rats had difficulty in learning both tasks, HAD1, LAD1 and NP rats performed at control levels. 4. These data suggest that P rats, but not HAD1 rats, may have problems learning tasks that require inhibition of ongoing or previously learned behaviors.

What is inherited in the predisposition toward alcoholism? A proposed model

BACKGROUND

The etiological factors associated with the predisposition to develop alcohol dependence remain largely unknown. In recent years, neurophysiological anomalies have been identified in young and adult offspring of alcoholic probands. These neuroelectric features have been replicated in several laboratories across many different countries and are observed in male and female alcoholics and some of their relatives and offspring. Moreover, these electrophysiological abnormalities are heritable and predictive of future alcohol abuse or dependence.

METHODS

A model is presented which hypothesizes that the genetic predisposition to develop alcoholism involves an initial state of central nervous system (CNS) disinhibition/hyperexcitability. We propose that the event-related brain potential (ERP) anomalies reflect CNS disinhibition. This homeostatic imbalance results in excess levels of CNS excitability which are temporarily alleviated by the ingestion of alcohol. It is hypothesized that this hyperexcitability is heritable, and is critically involved in the predisposition toward alcoholism and the development of dependence. A brief review of the relevant literature is presented.

RESULTS

Neurophysiological, neurochemical, and genetic evidence support the proposed model. In addition, strikingly similar observations between animal research and the human condition are identified. Finally, it is asserted that the proposed model is primarily biological in nature, and therefore does not account for the entire clinical variance.

CONCLUSION

A putative CNS homeostatic imbalance is noted as a critical state of hyperexcitability. This hyperexcitability represents a parsimonious model of what is inherited in the predisposition to develop alcoholism. It is our hope that this model will have heuristic value, resulting in the elucidation of etiological factors involved in alcohol dependence.