Aversive control of Betta splendens behavior using water disturbances: Effects of signaled and unsignaled free-operant avoidance and escape contingencies.

The vast majority of research on aversive control of behavior using animal models employs electric-shock based procedures with avian and mammalian species. Notwithstanding that pragmatic aspects explain this prominence in the field, there is still a need for testing well-established facts about shock-based aversive control using other, perhaps more ecologically valid or biologically relevant, forms of stimulation with other species. Following up on an observation that water disturbances appear to be aversive to Betta splendens (bettas), we developed a preparation for studying free-operant avoidance with this species in which water flows (WFs) replaced electric shocks. Fish changed compartments in a shuttle tank to escape or avoid 10-s WFs, which were delivered with 30-s flow-flow and response-flow intervals. We tested the effect of adding a warning stimulus (curtains of air bubbles) to the last 5 s of the response-flow interval (i.e., signaled avoidance) on the bettas' temporal distribution of crossings and swimming patterns. Escape was the predominant response, which importantly reduced the exposure to the WFs. Avoidance responses rarely exceeded the frequency of escape. The warning stimulus did not produce the expected postponement of responses to the last segment of the response-flow interval. Distinctive swimming patterns emerged, dissipated, and reappeared during baseline and treatment conditions. These findings confirm the aversive function of WFs for bettas and expand the generality of negative reinforcement phenomena using non-shock-based procedures with a less-frequently studied species. Further developments of the WFs paradigm show promise for replicating other aversive control phenomena (e.g., punishment), and analyzing spatiotemporal patterns produced by aversive contingencies.