DIY-NAMIC Behavior: A High-Throughput Method to Measure Complex Phenotypes in the Homecage

Abstract

Complex behavioral assessment is necessary to comprehensively assess in vivo manipulations in rodent models for neuropsychiatric disorders. Operant behavioral paradigms provide rich datasets and allow for the careful analysis of behavioral phenotypes. However, one major limitation in these studies is the expense and work-load that are required using traditional methods. The equipment for commercial operant boxes can be prohibitively expensive, and the daily experimenter effort and mouse costs required for these studies is extensive. Rodents are generally trained on task-specific paradigms for months, tested every day for 5-7 d/week. Additionally, appetitive paradigms usually require food restriction and are also commonly run in the non-active light phase of the rodent circadian rhythm. These limitations make operant behavioral testing especially difficult during adolescence, a time period of interest with regards to the development of adult-like phenotypes and a high-risk period for the development of neuropsychiatric disorders, including those which involve impulsive behavior. In order to address these issues, we developed an automated, inexpensive, open-source method which allows the implementation of most standard operant paradigms in the homecage of rodents in shorter time frames without food restriction, and with much less experimenter effort. All construction and code for the do-it-yourself Nautiyal Automated Modular Instrumental Conditioning (DIY-NAMIC) system are open source. We demonstrate their utility here by measuring impulsive behavior in a pharmacology experiment, as well as in adolescent mice.

Keywords: adolescence; automated behavioral testing; behavioral pharmacology; impulsivity; operant; serotonin.

Graphical abstract

Figure 1.

The DIY-NAMIC system is integrated into the homecage of standard mouse cages. A, The Arduino and all components of the DIY-NAMIC system are enclosed within the red encasing with the exception of a syringe for water or liquid reward, mounted on the outside. Modified cages remain compatible with most standard high-density ventilated racks. B, Inside homecage view of the DIY-NAMIC system including three noseports with blue LEDs illuminated, each of which contain a spout for liquid reward. C, The inside of the apparatus is shown including LEDs, solenoids, and IR head entry detectors wired to an Arduino UNO through an OpenMaze (OM1) shield. D, Schematic wiring diagram shows the wiring of one set of components (for one noseport of three total) to the OM1 shield. Blue rectangles indicate location of stacking headers for connection to Arduino. Red rectangles indicate locations of chip socket for H-bridge (h), push button for flush (b), and barrel jack (j). Green rectangles indicate location of female headers for connection to IR (LED and photodetector), cue LED, and solenoid components.

Figure 2.

Measuring operant behavior with the DIY-NAMIC system. A, Reward retrieval training (P1) occurred with presentations of reward available in trials with an ITI average of 45 s. Schematic shows the center reward port is illuminated when the reward is available. B, The number of rewards retrieved over 3 d is shown, separated by light and dark cycle. C, Number of pokes during the ITI when the reward is unavailable is shown over 3 d. D, Schematic for continuous reinforcement training (P2) shows that both side ports are illuminated during a trial, and pokes to either results in reward availability. E, The number of rewards received is shown across the 3 d of training. F, ITI responding is shown over 3 d for each noseport. G, Schematic for P3 which includes trial initiation shows how trials become self-initiated by a poke to the blinking center port. H, Number of self-initiated trials over 4 days of the paradigm. I, Schematic describes the operant cue discrimination paradigm (P4), in which the correct/rewarded port was illuminated by the LED. J, The group average shows that mice perform correctly on 85% of trials and have relatively few incorrect (14%) and omitted (1%) trials. K, The number of trials initiated and the number of correct trials are shown across 24 h, binned by 2 h. The shaded gray area indicates the dark phase of the light cycle. L, Schematic shows the modified two-choice serial reaction time task which was used to assay impulsive action. M, Three-day averages of premature responses per trial over three delay lengths, shown by 2-h bins over the dark cycle. N, Total number of correct, incorrect, and omitted trials on each of 3 d run on three delay lengths. All group averages are shown for N= 9 mice, with error bars representing SEM.

Figure 3.

Adolescents show higher levels of impulsive behavior. Premature responses as measured by total nosepokes during the delay period (A), total initiated trials (B), average number of omitted trials (C), and proportion of correct trials of those attempted (D) are shown for adults and adolescents for the three delay lengths. Group averages are shown for each delay averaged over 3 d, with error bars representing the SEM; *p < 0.05, #p = 0.05.

Figure 4.

5-HT_1B agonist CP 94253 reduces impulsivity. A, DIY-NAMIC boxes allow for continuous assessment of the behavioral effects of pharmacological manipulations on impulsive action. The average number of nosepokes during the 9-s delay window per trial is shown over the 12-h dark phase immediately following injection of vehicle or drug (CP 94253). B, Binned by 6 h, premature responses are shown for the first and second halves of the dark phase following vehicle or drug administration. Other measures including total number of trials initiated (C), ITI responding (D), proportion of omitted trials (E), and proportion of correct attempted trials (F) are shown for drug and saline conditions over first and second half of dark phase.