Macaque monkeys learn and perform a non-match-to-goal task using an automated home cage training procedure

Abstract

In neurophysiology, nonhuman primates represent an important model for studying the brain. Typically, monkeys are moved from their home cage to an experimental room daily, where they sit in a primate chair and interact with electronic devices. Refining this procedure would make the researchers' work easier and improve the animals' welfare. To address this issue, we used home-cage training to train two macaque monkeys in a non-match-to-goal task, where each trial required a switch from the choice made in the previous trial to obtain a reward. The monkeys were tested in two versions of the task, one in which they acted as the agent in every trial and one in which some trials were completed by a "ghost agent". We evaluated their involvement in terms of their performance and their interaction with the apparatus. Both monkeys were able to maintain a constant involvement in the task with good, stable performance within sessions in both versions of the task. Our study confirms the feasibility of home-cage training and demonstrates that even with challenging tasks, monkeys can complete a large number of trials at a high performance level, which is a prerequisite for electrophysiological studies of monkey behavior.

Figure 1

Schematic illustration of the home-cage training system and tasks. (a) Home-cage training system. The photograph on the left shows the cage interface (CI) connected through the guillotine door to the experimental cage. The monkey sits in front of the opening of the CI (1) where a stainless steel tube (2) releases the reward via a reward system (top right image). A touchscreen (3) allows the monkey to perform the task. A wide-angle camera (4) monitors the monkey during sessions. The lower image shows the behavioral control unit, which is connected to the touchscreen, reward system, and camera and is used to run the tasks and record the monkey’s responses. (b) The 19 steps presented during phase 1 of experiment 1. The monkey learns to touch a central stimulus (CS) of progressively smaller size (steps 1–5); to keep touching the CS while it is on the screen (steps 6–11); to remove its hand after the CS is turned off within a specific time window (steps 12–15), and to touch and continue toughing a target when it is presented on the screen (steps 16–19). (c) The sequence of events of one trial of the non-match-to-goal task. The upper section shows a trial performed by the monkeys in phase 2 of experiment 1 and in experiment 2. The lower section shows a trial performed by the ghost agent in experiment 2. The black rectangle represents the touchscreen. The white or red circle is the CS, and the disappearance of the CS represents the go signal for starting the movement. The targets (here, a pink rhombus and a yellow cross) are the stimuli for the choice. The gray rectangle in the computer trial simulates the sequence of actions necessary to carry out a correct complete trial. The first black rectangle on the left shows the target chosen in the previous trial that was, according to the task design, presented again in the current trial, alongside a different stimulus. On the right the four targets and the four types of visual feedback for both experiments are shown. Under the two example trials, the duration of each is shown.

Figure 2

Behavioral results of experiment 1. (a) Results of phase 1 and correct response rate of phase 2. Bars represent the number of trials performed to complete a step that was presented for the first time during phase 1. Dashed lines indicate the minimum number of trials required to reach the next step. (b) Learning curves for both monkeys. The curves show the percentage of correct choices for each session during phase 2 (44 sessions for monkey L and 31 for monkey N). The thick lines represent the 10 consecutive sessions in which the monkeys reached a correct response rate of above 70% (where a correct response rate of 50% corresponds to chance). (c) Working blocks. Each row represents a session during phase 2. Colored lines indicate working blocks, which comprise completed trials (see “Experimental procedures” section). The color bar indicates the correct response rate calculated for each working block. Black lines indicate the non-interactive blocks, which comprise trials in which the monkeys did not interact with the touchscreen. The white area indicates the end of the sessions. (d) Stability of interaction within phase 2 sessions. The curves indicate the mean percentage of complete trials (see experimental procedures) calculated using a moving window of 100 trials in sessions in which the learning criterion was achieved and which included at least 800 trials. Color-shaded regions represent the standard error of the mean. (e) Stability of performance within phase 2 sessions. Correct response rate for sessions in which the learning criterion was achieved and which included at least 800 trials. The sessions are split into two halves: first half (FH) and second half (SH). The central line represents the median and the top and bottom borders of each box represent the 25th (p25) and 75th (p75) percentiles of the two groups of trials. The whiskers represent the range from 1.5 times the interquartile range above p75 (p75 + 1.5 [p75 − p25]) and below p25 (p25 − 1.5 [p75 − p25]). Data points outside this range are plotted individually (red crosses). The dashed line indicates 50%, the level associated with responses dictated by chance alone.

Figure 3

Behavioral results of experiment 2. (a) Overall performance. Bars represent the percentage of correct choices in after-monkey and after-computer trials, based on all the sessions in experiment 2 (29 and 21 sessions for monkeys L and N, respectively). (b) Stability of the monkeys’ performance within sessions in after-monkey and after-computer trials. The structure of the boxplots and calculation of the percentages of correct choices are as described in Fig. 2e, except they apply to the after-monkey and after-computer trial categories. In both graphs, the dashed line indicates 50%, the level associated with responses dictated by chance alone. Asterisks indicate significant differences (***p < 0.001, Wilcoxon signed-rank test). (c) Stability of the monkeys’ interaction with the apparatus. The curves represent the mean percentage of completed trials (see experimental procedures) calculated in moving windows of 100 trials, using the sessions presented in (a).