Rapid learning of the 5-choice serial reaction time task in an automated rodent training system

Abstract

11Experiments aiming to understand sensory-motor systems, cognition and behavior often require animals 12 trained to perform complex tasks. Traditional training protocols require lab personnel to move the animals 13 between home cages and training chambers, to start and end training sessions, and in some cases, to 14 hand-control each training trial. Human labor not only limits the amount of training per day, but also 15 introduces several sources of variability and may increase animal stress. Here we present an automated 16 training system for the 5-choice serial reaction time task (5CSRTT), a classic rodent task often used to test 17 sensory detection, sustained attention and impulsivity. We found that fully automated training without 18 human intervention greatly increased the speed and efficiency of learning, and decreased stress as 19 measured by corticosterone levels. Introducing training breaks did not cancel these beneficial effects of 20 automated training, and mice readily generalized across training systems when transferred from 21 automated to manual protocols. Additionally, we validated our automated training system with mice 22 implanted with wireless optogenetic stimulators, expanding the breadth of experimental needs our 23 system may fulfill. Our automated 5CSRTT system can serve as a prototype for fully automated behavioral 24 training, with methods and principles transferrable to a range of rodent tasks. 25 A few automated training systems have been developed for rodent behavioral tasks 8-15 , including 5-choice 40 serial reaction time task (5CSRTT) 16,17 , in order to standardize the training and reduce the effects of human 41 factors and other random variables. While these systems provide means for large capacity automated 42 training of rodents, most of them are customized to train a specific task variant, and/or contain expensive, 43proprietary components. For these reasons, automated behavioral training of the 5CSRTT task has not yet 44 become widespread. Here we developed an affordable, open source, high-throughput automated training 45 system for mice and demonstrate its use on an automated protocol of the widely used 5CSRTT assay 18-21 . 46 We show that use of this Automated Training System (ATS) allows faster training of mice, and that 47

In vivo localization of chronically implanted electrodes and optic fibers in mice

“…five-choice serial reaction time task 36 . Before the surgery, the animals were pretrained for 5-10 days in an automated system 37 (see "Methods"). By the end of the pretraining period, all animals learned the task and performed at around 80% accuracy.…”

“…Poking during the ITI, choosing an incorrect port, or omitting a response was punished by a 5-s-long timeout, during which the house light was turned off and no new trial was started. All mice were pretrained before the surgery in a custom-built automated training system 37 , where they were allowed to enter the training setup from their home cages and perform the task regularly every 2 h for 15 min (12 times a day), until they reached 80% performance (5-10 days). After 1 week of recovery period following the surgery, postoperative CT scans were performed in the case of the test group (n = 7, allocated randomly), and the behavioral training was resumed.…”

Investigating gene-environment interaction on attention in a double-hit model for Autism Spectrum Disorder

“…The 5-CSRTT quantifies attention, impulsivity, and cognitive flexibility in preclinical models for ASD. Several previous studies have shown the construct validity of the 5-CSRTT as a model for attention [ 41 , 44 ]. Due to the equalized effects of genetic and environmental risk factors, it was initially hypothesized that Cntnap2 deficiency would interact with poly I:C MIA to exacerbate ASD-related attentional alterations seen in each model individually.…”

Dissociating motor impairment from five-choice serial reaction time task performance in a mouse model of Angelman syndrome

“…A 20-30 s intertrial interval would allow mice to finish eating rewards on most trials (Figure 5B). Alternatively, water rewards could be used instead of pellet rewards (Birtalan et al, 2020). Additionally, the 5CSRTT can be automated to allow ad libitium 24 h access to the task via a tube connecting the homecage to the testing chamber (Remmelink et al, 2017;Bruinsma et al, 2019).…”