VISUAL-REINFORCER COLOR, AND OPERANT BEHAVIOR IN SIAMESE FIGHTING FISH

Rage Against the Machine: Advancing the study of aggression ethology via machine learning

RATIONALE

Aggression, comorbid with neuropsychiatric disorders, exhibits with diverse clinical presentations and places a significant burden on patients, caregivers, and society. This diversity is observed because aggression is a complex behavior that can be ethologically demarcated as either appetitive (rewarding) or reactive (defensive), each with its own behavioral characteristics, functionality, and neural basis that may transition from adaptive to maladaptive depending on genetic and environmental factors. There has been a recent surge in the development of preclinical animal models for studying appetitive aggression-related behaviors and identifying the neural mechanisms guiding their progression and expression. However, adoption of these procedures is often impeded by the arduous task of manually scoring complex social interactions. Manual observations are generally susceptible to observer drift, long analysis times, and poor inter-rater reliability, and are further incompatible with the sampling frequencies required of modern neuroscience methods.

OBJECTIVES

In this review, we discuss recent advances in the preclinical study of appetitive aggression in mice, paired with our perspective on the potential for machine learning techniques in producing automated, robust scoring of aggressive social behavior. We discuss critical considerations for implementing valid computer classifications within behavioral pharmacological studies.

KEY RESULTS

Open-source automated classification platforms can match or exceed the performance of human observers while removing the confounds of observer drift, bias, and inter-rater reliability. Furthermore, unsupervised approaches can identify previously uncharacterized aggression-related behavioral repertoires in model species.

DISCUSSION AND CONCLUSIONS

Advances in open-source computational approaches hold promise for overcoming current manual annotation caveats while also introducing and generalizing computational neuroethology to the greater behavioral neuroscience community. We propose that currently available open-source approaches are sufficient for overcoming the main limitations preventing wide adoption of machine learning within the context of preclinical aggression behavioral research.

Animal Models of (or for) Aggression Reward, Addiction, and Relapse: Behavior and Circuits

Inappropriate and pathological aggression plays a leading role in the suffering and death of millions of people, and further places an untenable strain on the caregivers and families of those afflicted. In some cases, such as addictive drugs, aggression can be highly rewarding (appetitive) and continually pursued despite short- and long-term negative consequences. Similarly, recidivism (relapse) rates for repeat violent offenders are as high as relapse rates for drug addicts. Appetitive aggression and relapse to aggression seeking can be modeled in mice studies using conditioned place preference and self-administration procedures followed by a period of abstinence and subsequent tests for relapse to aggression preference and aggression seeking. These procedures allow for the study of the mechanisms that control the appetitive versus the consummatory (attack) phases of aggressive behavior. In this review, we first discuss the behavioral procedures developed to probe appetitive aggression in mouse models, spanning from Pavlovian to operant tasks, and we also describe the recently proposed phenomenon of "aggression addiction." Next, we discuss the pharmacological and circuit mechanisms of aggression conditioned place preference and aggression self-administration, seeking, and relapse, highlighting mechanistic congruence and divergence between appetitive and consummatory phases of aggression. We conclude by discussing clinical implications of the studies reviewed.

Effect of reward type on object discrimination learning in socially monogamous coppery titi monkeys (Callicebus cupreus)

Highly valued food items are often used as rewards to reinforce an animal's behavior. For social species, social interaction is rewarding and can drive an individual's behavior as well. In the currently study, we wanted to compare the efficacy of a food reward and a social reward on object discrimination learning in socially monogamous titi monkeys. We hypothesized that titi monkeys would perform more accurately for a social reward (their pair mate) than for a food reward (a highly desired food item). Eleven adult titi monkeys were tested with a two-object visual discrimination task for both types of reward. The colors and shapes of the objects in the two-object discrimination task were counterbalanced across subjects. During each trial, subjects were shown two objects, and the trial ended when the subject touched the reinforced shape (S+) or after 5 min. A correct trial was defined as one when the subject touched S+ first. We found that 45.5% of subjects were able to learn the task with a social reward, and 83.3% were able to learn the task with a food reward. We found that subjects balked more often and had fewer correct trials for the social reward. Finally, subjects took longer to approach the shapes for a social reward, possibly indicating lower motivation to engage in the task when a social reward is used compared to a food reward. Although significantly fewer subjects met criteria of success with the social reward than with the food reward, our results show that titi monkeys can learn a visual discrimination task with either type of reward.

Behavioral effects of fluoxetine on aggression and associative learning in Siamese fighting fish (Betta splendens)

Past research has implicated serotonin as an important neurotransmitter in the facilitation of aggressive behavior. In Siamese fighting fish (Betta splendens), the SSRI fluoxetine has been demonstrated to reduce both frequency and duration of aggressive displays across a variety of concentration exposure procedures. While this multi-method approach has provided strong evidence for fluoxetine's impact on aggression, no study has sought to examine the behavioral mechanism by which fluoxetine exerts its anti-aggressive effect. To address this question, a Go-No Go discrimination task utilizing mirror presentations as a reinforcer was designed. Consistent with previous reports, the results indicated that fluoxetine may exert a sedative effect upon aggressive behavior via decreased arousal to external stimuli.

Response acquisition by Siamese fighting fish (Betta splendens) with delayed visual reinforcement

Male Siamese fighting fish, Betta splendens, swam through a ring in an aquarium, breaking a photocell beam and initiating an unsignaled, resetting delay interval. Following delays of 0 s, 10 s, or 25 s, a 15-s mirror presentation released an aggressive display by the fish. Swimming through the ring increased in the absence of either a period of acclimatization to the reinforcer (analogous to magazine training when appetitive reinforcers are used) or explicit training of the response by the experimenters. Response rates were a decreasing function of delay duration. Other fish exposed to a schedule of response-independent mirror presentations failed to acquire and maintain the response. The results demonstrate the robustness and generality of the phenomenon of response acquisition with delayed reinforcement. They further qualify earlier observations about behavioral mechanisms involved in the phenomenon.

Aggression as positive reinforcement in mice under various ratio- and time-based reinforcement schedules

There is evidence suggesting aggression may be a positive reinforcer in many species. However, only a few studies have examined the characteristics of aggression as a positive reinforcer in mice. Four types of reinforcement schedules were examined in the current experiment using male Swiss CFW albino mice in a resident-intruder model of aggression as a positive reinforcer. A nose poke response on an operant conditioning panel was reinforced under fixed-ratio (FR 8), fixed-interval (FI 5-min), progressive ratio (PR 2), or differential reinforcement of low rate behavior reinforcement schedules (DRL 40-s and DRL 80-s). In the FR conditions, nose pokes were maintained by aggression and extinguished when the aggression contingency was removed. There were long postreinforcement pauses followed by bursts of responses with short interresponse times (IRTs). In the FI conditions, nose pokes were maintained by aggression, occurred more frequently as the interval elapsed, and extinguished when the contingency was removed. In the PR conditions, nose pokes were maintained by aggression, postreinforcement pauses increased as the ratio requirement increased, and responding was extinguished when the aggression contingency was removed. In the DRL conditions, the nose poke rate decreased, while the proportional distributions of IRTs and postreinforcement pauses shifted toward longer durations as the DRL interval increased. However, most responses occurred before the minimum IRT interval elapsed, suggesting weak temporal control of behavior. Overall, the findings suggest aggression can be a positive reinforcer for nose poke responses in mice on ratio- and time-based reinforcement schedules.

Visual reinforcement in the female Siamese fighting fish, Betta splendens

Operant conditioning with Betta splendens (Bettas) has been investigated extensively using males of the species. Ethological studies of female Bettas have revealed aggressive interactions that qualitatively parallel those between male Bettas. Given these similarities, four experiments were conducted with female Bettas to examine the generality of a widely reported finding with males: mirror-image reinforcement. Swimming through a ring was reinforced by a 10-s mirror presentation. As with males, ring swimming was acquired and maintained when mirror presentations were immediate (Experiments 1, 2, and 3) and delayed (Experiment 4). The failure of conventional extinction (Experiments 1 and 2) and response-independent mirror presentations (Experiment 3) to maintain responding confirmed the reinforcing properties of mirror presentation. These results extend previous findings of mirror images as reinforcers in males of the same species and illustrate a complementarity between behavioral ecology and the experimental analysis of behavior.

Fish on Prozac: a simple, noninvasive physiology laboratory investigating the mechanisms of aggressive behavior in Betta splendens

The neuromodulator serotonin is an important regulator of aggressive behavior in vertebrates. Experimentally increasing synaptic levels of serotonin with fluoxetine, a selective serotonin reuptake inhibitor, has been shown to reliably decrease the expression of aggressive behavior. Here, we describe a method by which fluoxetine can be noninvasively administered to male Betta splendens (an attractive model for the study of aggressive behavior) and describe a simple laboratory exercise that allows students to experimentally investigate the physiological mechanisms of aggressive behavior. We demonstrate that relatively short-term exposure (3 h) of male bettas to as little as 3 microg/ml of fluoxetine-treated aquarium water is sufficient to reduce the expression of specific aggressive behaviors. We emphasize the physiological concepts that can be addressed with this exercise, including the role of the serotonergic system in regulating aggression, and the interplay of environmental contaminants and physiology in regulating the expression of behavior. We also highlight important aspects of experimental design. This exercise can be flexibly altered to accommodate one or several laboratory periods. It is also low cost, is low impact to the animals, and requires minimal preparation time for instructors.

Responses of crayfish to a reflective environment depend on dominance status

Previous work has shown that the reflection from aquarium walls can modify several aspects of crayfish behaviour, but only if the crayfish are socialized. Such socialization is known to generate a dominance hierarchy composed of dominant and subordinate animals. The present study was conducted to determine whether or not behavioural responses to reflection depend on dominance status. Procambarus clarkii (Girard, 1852) were maintained in pairs for 2 weeks to establish dominance ranks and were subsequently observed in an aquarium in which half of the walls were lined with mirrors and the other half were lined with nonreflective plastic. Dominant crayfish spent more time on the reflective side of the aquarium than on the nonreflective side, and they spent more time in reflective corners than in nonreflective corners. Dominant crayfish also turned more frequently on the reflective side and crossed the tank more frequently toward reflective walls. Such differences in behaviour were not observed for subordinate crayfish or for crayfish that had been isolated for 2 weeks. Subordinate crayfish exhibited more reverse walking on the reflective side of the aquarium than on the nonreflective side, but crayfish in the other groups did not. Thus, responses of crayfish to reflection depend on dominance status.

Using visual reinforcement to establish stimulus control of responding of Siamese fighting fish (Betta splendens)

Stimulus control of ring swimming was studied with male Siamese fighting fish (Betta splendens) using 2-component multiple schedules in which the components were correlated with the presence or absence of air bubbles in the water. In Experiment 1, either response-independent mirror presentations or extinction was juxtaposed with immediate response-dependent mirror presentations. Rates of ring swimming generally were higher with immediate reinforcement than with either response-independent mirror presentations or extinction. In Experiment 2, different durations of response-dependent mirror presentations were juxtaposed. Generally, higher rates of ring swimming occurred with 15-s than with 0-, 1-, or 3-s durations. Results demonstrate that stimulus control of responding can be established with these fish under several conditions of differential reinforcement.

Chemoreception in Betta splendens

4 Ss were kept visually and chemically isolated for 21 days in individual glass beakers with no aereation or drainage. They were then transplanted to each other's beakers (their “habitats”), which retained the water “contaminated” by the original S. They were observed for signs of aggressive display. No sign of displaying occurred.

Operant responding in Siamese fighting fish (Betta splendens) as a function of schedule of reinforcement and visual reinforcers

Siamese fighting fish were trained to emit an operant response that was reinforced by the opportunity to view a motion picture film image of another fish. Performance under various schedules of reinforcement was examined. When reinforcement followed every response and when reinforcement was delivered after every second response, the number of responses per session was higher than during operant level or during extinction. Reinforcement delivered following intervals of no responding (differential reinforcement of other behavior) markedly decreased responding. Light from a projector without film was found to be as effective a reinforcer as film reinforcement. Responding when projector light reinforcement followed every response was maintained at approximately the same level as that obtained under film reinforcement. Responses per session decreased when only the light was delivered on a differential-reinforcement-of-other-behavior schedule. The behavior of the fish during presentation of the film was markedly different from their behavior while the projector light was being presented.

Acquired stimulus control of drug-induced changes in aggressive display in betta splendens

Male Siamese fighting fish exhibit stereotyped aggression reactions to their mirror reflections. When distinctive neutral stimuli (flickering colored lights) were repeatedly associated with drug-potentiated aggression (morphine sulfate) and drug-depressed aggression (phenergan), the stimuli came to exert specific stimulus control over aggressive display even after the drugs were discontinued.

Classical conditioning of aggressive display in Siamese fighting fish

Unconditioned aggressive-display behavior elicited by the mirror image of a male Siamese Fighting Fish was brought under the control of a previously ineffective stimulus by classical conditioning. A stimulus light repeatedly paired with mirror presentation came to elicit the complex aggressive-behavior sequence. Relative rates of acquisition of four components of the display were compared. Fin erection and undulating movements were acquired most rapidly while gill-cover erection and frontal approach were acquired most slowly. A discriminative conditioning procedure revealed that the response was specifically elicited by the conditioned stimulus, and not a sensitization artifact.