Delayed reinforcement hinders subsequent extinction

In operant conditioning, animals associate their own behavior with a reinforcer, and the probability of the behavioral responses is increased. This form of learning is called reinforcement. In contrast, when the previously reinforced responses are no longer paired with a reinforcer, these responses are eventually extinguished. The effectiveness of reinforcement depends primarily on time intervals between reinforcers and responses, but it is not fully understood how the intervals affect subsequent extinction. To address this question, we performed electrical stimulation of the rat medial forebrain bundle (MFB), a part of the brain reward system, and an operant task in which the MFB was electrically stimulated 0.1 s (immediate condition) or 1 s (delayed condition) after the rat's nose was poked. During the first half of the task period (a reinforcement period), nose pokes were associated with MFB stimulation. In contrast, during the second half (an extinction period), we did not stimulate the MFB irrespective of nose pokes. We found that rats exhibited increased nose-poke behaviors during the reinforcement period under both conditions, whereas during the extinction period, nose pokes were more persistent in the delayed condition than in the immediate condition. The persistent responses in the extinction period were independent of responses in the reinforcement period. Therefore, reinforcement and extinction are driven by independent neural mechanisms.

Design and construction of a low-cost nose poke system for rodents

Operant behavioral tasks for animals have long been used to probe the function of multiple brain regions (i.e., understanding the role of dopamine in electrical brain stimulation reward [1], or determining the rewarding properties of feeding oriented brain pathways [2]). The recent development of tools and techniques has opened the door to refine the answer to these same questions with a much higher degree of specificity and accuracy, both in biological and spatial-temporal terms [3], [4]. A variety of systems designed to test operant behavior are now commercially available, but have prohibitive costs. Here, we provide a low-cost alternative to a nose poke system for mice. Adapting a freely available sketch for ARDUINO boards, in combination with an in-house built PVC box and inexpensive electronic material we constructed a four-port nose poke system that detects and counts port entries. To verify the applicability and validity of our system we tested the behavior of DAT-CRE transgenic mice injected with an adeno-associated virus to express ChannelRhodopsin 2 in the Ventral tegmental area (VTA) and used the BNC output to drive a blue laser coupled to a fiber implanted above the VTA. Over 6 days, mice perform as it has been reported previously [5] exhibiting a remarkable preference for the port that triggers optogenetic stimulation of VTA dopamine neurons.

•

We provide a low cost alternative to commercially available nose poke system.

•

Our custom made apparatus is open source and TTL compatible.

•

We validate our system with optogenetic self-stimulation of dopamine neurons.

On the time course of rapid and repeatable response induction

As reinforcers are delivered for a particular response, that response increases. Other similar responses also increase, with the extent to which this increase is seen being positively related to the similarity between the responses. This process, called induction, is pronounced during initial response acquisition, and dissipates as the target response is established. The present experiment examined this process in a paradigm wherein responses to one of five nose poke receptacles results in reinforcement for six sessions in each condition and the target location changes between conditions. Consistent patterns of induction were seen across conditions, with patterns of induction dissipating within each condition.