Limits...
Rats prefer mutual rewards in a prosocial choice task.

Hernandez-Lallement J, van Wingerden M, Marx C, Srejic M, Kalenscher T - Front Neurosci (2015)

Bottom Line: Pro-sociality, i.e., the preference for outcomes that produce benefits for other individuals, is ubiquitous in humans.Here, we present a rodent analog of the Pro-social Choice Task that controls strategic components, de-confounds other-regarding choice motives from the animals' natural tendencies to maximize own food access and directly tests the effect of social context on choice allocation.We found that actors chose "both reward" at levels above chance and more often in the partner than in the toy condition.

View Article: PubMed Central - PubMed

Affiliation: Comparative Psychology, Institute of Experimental Psychology, Heinrich-Heine University of Düsseldorf Düsseldorf, Germany.

ABSTRACT
Pro-sociality, i.e., the preference for outcomes that produce benefits for other individuals, is ubiquitous in humans. Recently, cross-species comparisons of social behavior have offered important new insights into the evolution of pro-sociality. Here, we present a rodent analog of the Pro-social Choice Task that controls strategic components, de-confounds other-regarding choice motives from the animals' natural tendencies to maximize own food access and directly tests the effect of social context on choice allocation. We trained pairs of rats-an actor and a partner rat-in a double T-maze task where actors decided between two alternatives only differing in the reward delivered to the partner. The "own reward" choice yielded a reward only accessible to the actor whereas the "both reward" choice produced an additional reward for a partner (partner condition) or an inanimate toy (toy Condition), located in an adjacent compartment. We found that actors chose "both reward" at levels above chance and more often in the partner than in the toy condition. Moreover, we show that this choice pattern adapts to the current social context and that the observed behavior is stable over time.

No MeSH data available.


Individual differences in pro-social behavior. (A) Cumulative choice plots illustrate individual differences in pro-social behavior: for each trial, the running pro-social tally is incremented by +1 for each BR choice, and decremented by −1 for each OR choice. Thus, a monotonous upward slope indicates consistent BR choices across trials and sessions, neutral slopes indicate indifference, and negative slopes indicate consistent OR choices. Gray areas represent sessions where the BR compartment was on the left side. The two left panels show the cumulative choice plots of a rat classified as pro-social (performance in the toy condition indicated in red, upper left panel, performance in the partner condition indicated in blue in the 2nd down left panel). The two right panels show the choice data from a rat classified as non-pro-social. (B) Social bias scores of all rats: colors represent rats classified as pro-social (green) and non-pro-social (gray). The vertical red line represents the upper 95% confidence interval threshold obtained from the permutation analysis. (C) Correlation between partner weight index and social bias scores: we found a negative correlation between the social bias score and the normalized partner weight.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4296215&req=5

Figure 3: Individual differences in pro-social behavior. (A) Cumulative choice plots illustrate individual differences in pro-social behavior: for each trial, the running pro-social tally is incremented by +1 for each BR choice, and decremented by −1 for each OR choice. Thus, a monotonous upward slope indicates consistent BR choices across trials and sessions, neutral slopes indicate indifference, and negative slopes indicate consistent OR choices. Gray areas represent sessions where the BR compartment was on the left side. The two left panels show the cumulative choice plots of a rat classified as pro-social (performance in the toy condition indicated in red, upper left panel, performance in the partner condition indicated in blue in the 2nd down left panel). The two right panels show the choice data from a rat classified as non-pro-social. (B) Social bias scores of all rats: colors represent rats classified as pro-social (green) and non-pro-social (gray). The vertical red line represents the upper 95% confidence interval threshold obtained from the permutation analysis. (C) Correlation between partner weight index and social bias scores: we found a negative correlation between the social bias score and the normalized partner weight.

Mentions: Overall, the above analysis, in which we pooled BR choices across all rats, showed that the rats' frequency of choosing the BR compartment was significantly above chance in the partner condition, but the effect was relatively small (55% BR choices on average). However, the preference for the BR compartment greatly varied across rats, i.e., some rats showed substantially higher preference for the BR alternative in the partner condition compared to the toy condition (Figure 3A—left), whereas others neither developed a preference for the BR alternative nor showed a condition-dependent choice pattern (Figure 3A—right). Thus, the overall mean fraction of BR choices may be diluted by the data from rats that did not display condition-dependent preferences. To determine the extent to which rats differed in their BR-preferences, we calculated a social bias score for each rat (SB, See Section Analysis) reflecting the percent difference in BR-choices in the partner compared to the toy condition. Thus, SB scores can be interpreted as estimates of how much more (or less) a rat preferred the BR-alternative in the partner relative to the toy condition. Furthermore, we compared each rats' SB score to a benchmark SB score distribution obtained through a bootstrapped permutation analysis (see Supplementary Data and Figure 3B; the red vertical line indicates the 95% confidence interval limit). We categorized all rats showing significantly higher SB scores than the upper confidence interval bound as pro-social (N = 29). All remaining animals were categorized as non-pro-social (N = 19). This analysis revealed a substantial degree of heterogeneity in preferences across rats, with SB scores ranging from −14.8 (14.8% more BR choices in the toy than in the partner condition) to 45.6 (45.6% more BR choices in the partner than in the toy condition).


Rats prefer mutual rewards in a prosocial choice task.

Hernandez-Lallement J, van Wingerden M, Marx C, Srejic M, Kalenscher T - Front Neurosci (2015)

Individual differences in pro-social behavior. (A) Cumulative choice plots illustrate individual differences in pro-social behavior: for each trial, the running pro-social tally is incremented by +1 for each BR choice, and decremented by −1 for each OR choice. Thus, a monotonous upward slope indicates consistent BR choices across trials and sessions, neutral slopes indicate indifference, and negative slopes indicate consistent OR choices. Gray areas represent sessions where the BR compartment was on the left side. The two left panels show the cumulative choice plots of a rat classified as pro-social (performance in the toy condition indicated in red, upper left panel, performance in the partner condition indicated in blue in the 2nd down left panel). The two right panels show the choice data from a rat classified as non-pro-social. (B) Social bias scores of all rats: colors represent rats classified as pro-social (green) and non-pro-social (gray). The vertical red line represents the upper 95% confidence interval threshold obtained from the permutation analysis. (C) Correlation between partner weight index and social bias scores: we found a negative correlation between the social bias score and the normalized partner weight.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4296215&req=5

Figure 3: Individual differences in pro-social behavior. (A) Cumulative choice plots illustrate individual differences in pro-social behavior: for each trial, the running pro-social tally is incremented by +1 for each BR choice, and decremented by −1 for each OR choice. Thus, a monotonous upward slope indicates consistent BR choices across trials and sessions, neutral slopes indicate indifference, and negative slopes indicate consistent OR choices. Gray areas represent sessions where the BR compartment was on the left side. The two left panels show the cumulative choice plots of a rat classified as pro-social (performance in the toy condition indicated in red, upper left panel, performance in the partner condition indicated in blue in the 2nd down left panel). The two right panels show the choice data from a rat classified as non-pro-social. (B) Social bias scores of all rats: colors represent rats classified as pro-social (green) and non-pro-social (gray). The vertical red line represents the upper 95% confidence interval threshold obtained from the permutation analysis. (C) Correlation between partner weight index and social bias scores: we found a negative correlation between the social bias score and the normalized partner weight.
Mentions: Overall, the above analysis, in which we pooled BR choices across all rats, showed that the rats' frequency of choosing the BR compartment was significantly above chance in the partner condition, but the effect was relatively small (55% BR choices on average). However, the preference for the BR compartment greatly varied across rats, i.e., some rats showed substantially higher preference for the BR alternative in the partner condition compared to the toy condition (Figure 3A—left), whereas others neither developed a preference for the BR alternative nor showed a condition-dependent choice pattern (Figure 3A—right). Thus, the overall mean fraction of BR choices may be diluted by the data from rats that did not display condition-dependent preferences. To determine the extent to which rats differed in their BR-preferences, we calculated a social bias score for each rat (SB, See Section Analysis) reflecting the percent difference in BR-choices in the partner compared to the toy condition. Thus, SB scores can be interpreted as estimates of how much more (or less) a rat preferred the BR-alternative in the partner relative to the toy condition. Furthermore, we compared each rats' SB score to a benchmark SB score distribution obtained through a bootstrapped permutation analysis (see Supplementary Data and Figure 3B; the red vertical line indicates the 95% confidence interval limit). We categorized all rats showing significantly higher SB scores than the upper confidence interval bound as pro-social (N = 29). All remaining animals were categorized as non-pro-social (N = 19). This analysis revealed a substantial degree of heterogeneity in preferences across rats, with SB scores ranging from −14.8 (14.8% more BR choices in the toy than in the partner condition) to 45.6 (45.6% more BR choices in the partner than in the toy condition).

Bottom Line: Pro-sociality, i.e., the preference for outcomes that produce benefits for other individuals, is ubiquitous in humans.Here, we present a rodent analog of the Pro-social Choice Task that controls strategic components, de-confounds other-regarding choice motives from the animals' natural tendencies to maximize own food access and directly tests the effect of social context on choice allocation.We found that actors chose "both reward" at levels above chance and more often in the partner than in the toy condition.

View Article: PubMed Central - PubMed

Affiliation: Comparative Psychology, Institute of Experimental Psychology, Heinrich-Heine University of Düsseldorf Düsseldorf, Germany.

ABSTRACT
Pro-sociality, i.e., the preference for outcomes that produce benefits for other individuals, is ubiquitous in humans. Recently, cross-species comparisons of social behavior have offered important new insights into the evolution of pro-sociality. Here, we present a rodent analog of the Pro-social Choice Task that controls strategic components, de-confounds other-regarding choice motives from the animals' natural tendencies to maximize own food access and directly tests the effect of social context on choice allocation. We trained pairs of rats-an actor and a partner rat-in a double T-maze task where actors decided between two alternatives only differing in the reward delivered to the partner. The "own reward" choice yielded a reward only accessible to the actor whereas the "both reward" choice produced an additional reward for a partner (partner condition) or an inanimate toy (toy Condition), located in an adjacent compartment. We found that actors chose "both reward" at levels above chance and more often in the partner than in the toy condition. Moreover, we show that this choice pattern adapts to the current social context and that the observed behavior is stable over time.

No MeSH data available.