Third-party punishment-like behavior in a rat model

Abstract

Third-party punishment (TPP) is an altruistic behavior or sense willing to punish transgressors as a third party not directly involved in the transgression. TPP is observed worldwide, regardless of tradition and culture, and is essential for morality in human society. Moreover, even preverbal infants display TPP-like judgement, suggesting that TPP is evolutionarily conserved and innate. Thus, it is possible that non-human animals display TPP-like behavior, although TPP has been said to be human-specific. We investigated whether or not male mature Wistar rats displayed TPP-like behaviors when they witnessed deadly aggression by an unknown aggressive mouse toward another unknown victim mouse. Normally reared rats did not display TPP-like behaviors, but rats reared with extensive affectionate handling by human caretakers as beloved pets contacted the unknown aggressive mice in a gentle manner leading to reduced aggression toward the unknown victim mice, even when the aggressive mice fought back. Furthermore, the handled rats touched unknown rat pups that were drowning in water and anesthesia-induced comatose rats more frequently than control rats. These findings suggest a possibility that TPP is not in fact human-specific and innate but rather an acquired behavior that flourishes in affectionate circumstances.

Keywords: Behavior; Morality; Prefrontal cortex; Triage; Wistar.

Fig. 1

Extensive affectionate handling (EAH). (a) In EAH, rats were allowed to move freely across the caretaker’s body. (b–d) EAH rats showed familiarity to humans with their palms outstretched in front of a cage while simultaneously seeming to distinguish the caretaker (c) from unknown experimenters (b: male; d: female). (e) CNT rats did not approach the caretaker. (f–i) The attachment test (AtT): the caretaker gently placed an EAH rat on one end of the open arm of the elevated plus maze (EPM) (f), moved to the other end, and held out both hands, calling “Come!” or “Rick (the name given to all of the rats)” (g); if all 4 of the rat's legs were on the hands within 90 s, the session was considered to have been successfully completed (h). All 16 EAH rats but only 1 CNT rat reached the hands of the caretaker; ‘yes’ and ‘no’ denote succeeded and failed in the AtT, respectively (i). χ² test and Fisher’s exact test. (j, k) The open-field test (OFT): total moved distance (j) and the duration staying in the center zone of OF arena (k). Unpaired two-tailed t-test. (l, m) An EPM apparatus (l). EAH rats stayed longer on the open arms than CNT rats (m). Unpaired two-tailed t-test. n = 16. *, ***, **** indicate statistical significance at p < 0.05, 0.001, and 0.0001, respectively.

Fig. 2

Apparatus and procedures of third-party punishment test (TPPT). (a, b) The TPPT apparatus. Values are in millimeters (a). Two video cameras were set in the front and the side of the apparatus. (c–n) The observation procedure of TPPT in the case of an aggressive mice pair. All of the steps were videorecorded. (c, i) Confirming that the ICR (white) and BL6 (black) mice pairs showed aggressive behaviors in the apparatus. The time (min) shown in figure, (c–h) denotes observation duration. (d, j) An actor rat was released for habituation to the apparatus and the brush attached to the acrylic rod. (e, k) A transparent partition was inserted to divide the mouse and rat compartments, and the mice and rats were placed in each compartment. The rat was allowed to watch the aggression by the ICR mouse toward the BL6 mouse for 3 min. (f, l) Inserting a partition between the two mouse compartments to stop the aggressive behavior by the ICR mouse and waiting for 2 min for the animals’ excitement to subside. (g, m) All partitions were removed, and the behaviors of the animals were observed for 3 min. (h, n) After the rat was removed, the behaviors of the mice were observed for 1 min.

Fig. 3

Third-party punishment test (TPPT) results. (a, b) Behaviors of two representative CNT rats in the TPPT. The CNT rats avoid approaching the mice. The two video captures show instants when ICR mice attacked BL6 mice. (c, d) Behaviors of an EAH rat. The rat suffered a retaliatory strike by an ICR mouse (c), at which point the rat nonviolently contacted the aggressive mouse (d). (e–g) Another EAH rat pressed down a violent ICR mouse with its forepaws but did not bite the mouse (e). The ICR mouse decreased its aggressive behaviors toward the BL6 mouse in the presence of the EAH rat (f). The rat approached the BL6 mouse but did not frequently touch it (g). (h) EAH rats showed more frequent contact with the aggressive ICR mice than with the BL6 mice compared to CNT rats. (i, j) When presented with a peaceful pair of mice, there were no fights at all in the absence (i) or presence (j) of an EAH rat. (k) When presented with peaceful pairs, the number of contacts did not significantly differ between the CNT and EAH rats. (l) Most EAH rats made contact more frequently with aggressive ICR mice than with BL6 mice; however, there was no significant difference regarding frequency of contact made by CNT rats. (m) The ICR mice reduced the number of attacks made in the presence of the EAH rats. (n) The number of ICR mice that attacked the BL6 mice within 1 min after rat removal; ‘yes’ and ‘no’ denote attacked and not-attacked in the TPPT, respectively. CNT, n = 16; EAH, n = 16; only the EAH group data in h, n = 15. (h, k) A two-way ANOVA and Tukey’s post hoc test; paired (l) or unpaired (m) two-tailed t-test; (n), χ² and Fisher’s exact test.

Fig. 4

The actors’ attitudes toward the mice after witnessing the aggression in TPPT. (a) Definition of ICR and BL6 zones in the rat compartment in the TPPT apparatus; the former was adjacent to the ICR compartment, and the latter the BL6 compartment. Values are in millimeters. The ICR and BL6 zones were switched for each experiment. (b) The partition between the rat and mice compartments with holes. (c, d) The actor rats could sniff the mice through the holes drilled the partition set between the rat and the mice compartments by inserting their nose tips. Arrowheads denote the sniffing nose tip through the holes. (e) Sniffing frequency by actors toward ICR or BL6 mice was counted. (f, g) Paired t analysis on the sniffing frequency by EAH (f) and CNT (g) rats toward attacker ICR or victim BL6 mice. (h) Correlation between sniffing frequency and number of contacts with ICR mice by EAH rats. (i) Duration stayed in either ICR or BL6 zones. (j, k) Analysis on the duration by EAH (f) and CNT (g) rats in the ICR or BL6 zones with paired t-test. (l) n = 16; except for the data in h and i (n = 15). (e, i) A two-way ANOVA and Tukey’s post hoc test; (f, g, j, k) paired two-tailed t-test; (h, l) Pearson correlation coefficient and two-tailed Student's t-distribution.

Fig. 5

The drowning test (DT), 4-room test (4RT), triage test (TT), and rescuing-restrained-rat test (RarT). (a–d) DT: the apparatus (a). Values are in millimeters. A drowning PND16 pup and an actor rat (b). Response of a CNT rat (c) and an EAH rat (d; see SV5). The number of rats that touched the partition between the left and center chambers during habituation; ‘yes’ and ‘no’ denote touched and not-touched the partition, respectively (e) and those that made contact with the drowning pups; ‘yes’ and ‘no’ denote contacted and not-contacted the pups, respectively (f). (g–i) 4RT: the 4RT box (g). Colocalization of an actor and a PND11 pup (h). Duration (sec) of the colocalization (i). Mother (Moth) rats but not Father (Fath) rats stayed with their pups for longer periods than CNT and EAH rats. (j–o) Triage test (TT): setting (j). Video captures showing behaviors of a CNT rat (k) and an EAH rat (l) in the live zone. Duration (sec) in the live and dead zones (m). Number of contacts with the head (n) and the body (o). EAH rats stayed longer in the live zone than in the dead zone and made contact with the head of comatose rats more frequently than with the head of euthanized rats. (p–u) RarT: setting (p). A restrained rat was placed in a live zone of the TT. A rescued rat (q, r; Fig. S8). Setting of the rescue zone corresponding to the live zone in the TT and a representative heatmap showing movements of an EAH rat over 5 min (s). Duration (sec) in the rescue zone (t) and time until the first entry (sec) into the rescue zone (u). n = 16. (e, f) χ² test and Fisher’s exact test; (i) a one-way ANOVA and Tukey’s multiple comparison test; (m–o) a two-way ANOVA and Sidak’s multiple comparison test; (t, u), unpaired two-tailed t-test. *, **, *** indicate statistical significance at p < 0.05, 0.01, and 0.001, respectively.

Fig. 6

Various rearing methods and behavioral changes in the AtT, GOFT, and TT. (a) Group rearing of 12 rats in an enriched environment (EE) prepared in a complex housing. (b) Rearing with gentle stroking (GS): rats gently stroked on the back with a cosmetic brush. (c) Attachment test (AtT). Rats reared with 5 different methods were subjected to AtT; ‘yes’ and ‘no’ denote succeeded and failed in the AtT, respectively. F1: offspring of EAH rat parents. (d, e) Grouped OFT (GOFT): four cage mates were released to the OF, and the mean distance between the rats was measured for 5 min (d). (e) GOFT with an unknown rat: mean distance between the cage mate and the unknown rat. (f) Duration spent in the live or dead zone in the triage test (TT). (g) The comparison of the duration spent in the live and dead zones by individual rats. (h) Number of contacts with the heads of comatose and euthanized rats in TT. (i) The comparison of the number of contacts with the heads of the comatose and euthanized rats by individual rats. n = 24, except for GS group (n = 16). (c) χ² test and Fisher’s exact test; (d, e) a one-way ANOVA and Tukey’s multiple comparison test; (f, h) a two-way ANOVA and Sidak’s multiple comparison test; (g, i) paired two-tailed t-test. *, **, ***, **** indicate statistical significance at p < 0.05, 0.01, 0.001, and 0.0001, respectively.

Fig. 7

EAH-induced changes in gene expression in the mPFC. (a) Significant results from comparison of the gene expression in the medial prefrontal cortex (mPFC) between CNT and EAH rats were obtained by RNAseq; Rps20_3, Fos, Egr1, and Ier2. (b, c) Increased Rps20 (b) and Fos (c) expression in EAH brains as revealed by qPCR. (d) Gene expression encoding oxytocin (Oxt) in the hypothalamus did not significantly change. (e, f) Rps20 protein expression was found in neurons (e, CNT; f, EAH brains) (green; Rps20, white; a neuronal marker NeuN, red; a microglia marker CD11b). Unpaired two-tailed t-test. *, **, ***, **** indicate statistical significance at p < 0.05, 0.01, 0.001, and 0.0001, respectively.