Operant Conditioning in Indian Free-Ranging Dogs: Effects of Positive and Threatening Cues on Sociability

Abstract: Sociability toward humans is a key adaptive trait in free-ranging dogs, enabling them to access resources while navigating risks associated with human interactions. In this study, we investigated whether operant conditioning shapes sociability in Indian free-ranging dogs and whether learned responses generalize to unfamiliar individuals. We experimentally exposed 58 dog groups to either positive or a threatening cue over five consecutive days and assessed their behaviour using approach proportion, approach latency, and demeanor across repeated interactions with a familiar experimenter, followed by a test with an unfamiliar individual. Using Bayesian generalized linear mixed models, we found that cue type and repeated exposure significantly influenced sociability. Dogs exposed to a positive cue showed increased approach behaviour and reduced approach latency over time, along with increased affiliative demeanor. In contrast, dogs exposed to threatening cues exhibited reduced approach behaviour, increased approach latency, and a shift toward neutral and less affiliative responses across days. Importantly, positive cues partially generalized across individuals, as dogs showed increased approach toward an unfamiliar experimenter, although this was accompanied by hesitation to approach. In contrast, threatening cues did not generalize in the same way; dogs did not reduce their approach toward unfamiliar individuals but displayed increased approach latency, indicating heightened caution. Our findings demonstrate that operant conditioning plays a crucial role in shaping dog-human interactions, with asymmetric generalization of positive and threatening experiences.

• The paper demonstrates that repeated positive cues significantly increased dog approach frequency while reducing latency.
• Operational tests showed that threatening cues led to increased hesitation and neutral behavior, contrasting with positive cue effects.
• Bayesian hierarchical analysis revealed asymmetric generalization, where positive reinforcement generalized better across humans than threats.

Operant Conditioning and Sociability in Indian Free-Ranging Dogs

Experimental Context and Aims

This study provides a rigorous assessment of how operant conditioning—via repeated positive (food reward) and threatening (stick display) human cues—modifies the sociability of Indian free-ranging dogs and evaluates the extent of generalization of these learned responses across different human individuals (2607.02056). The experimental design involved systematically exposing 58 dog groups to either cue across five consecutive days, followed by cross-individual tests with a novel human. Key behavioral metrics included proportion and latency of approach, and coded demeanor, analyzed using Bayesian hierarchical models.

Major Findings and Quantitative Results

Differential Impacts of Positive and Threatening Cues

• Learning Curve and Behavioral Plasticity: Repeated exposure to positive cues yielded a statistically credible increase in approach frequency and lowered approach latency over time, alongside a shift toward affiliative demeanor. By contrast, repeated threatening cues reduced approach proportion, increased approach latency, and resulted in a notable transition toward neutral, less-affiliative behaviors.
• Temporal Dynamics: For positive cues, approach proportion was significantly higher on Day 6A (unfamiliar experimenter) than Day 1 (median estimate = -0.12; 95% HPD: -0.23, -0.02). Approach latency decreased notably over repeated positive exposures (e.g., Day 1 vs. Day 4, median estimate = 0.82; 95% HPD: 0.35, 1.30).
• Negativity Bias: Dogs exposed to threatening cues exhibited a more rapid decrease in approach motivation (elevated latency and declining approach proportion), but generalized avoidance to unfamiliar humans was not observed. Approach latency increased across days post-exposure to threatening cues; e.g., Day 1 vs. Day 5, median estimate = -1.18; 95% HPD: -1.76, -0.61.

Asymmetric Generalization

• Positive Cue Generalization: There was partial generalization of positive reinforcement—exposure to food rewards by one individual modestly facilitated approach behavior to unfamiliar experimenters. This was statistically robust for approach proportion but not for approach latency, indicating persistence of caution in new contexts.
• Threatening Cue Non-Generalization: Conversely, threatening cues did not broadly suppress approach behavior toward unfamiliar humans. Instead, prior exposure led to increased hesitation (latency), not outright avoidance, when interacting with novel humans.

Demeanor Modulation and Sex Effects

• Behavioral Profiles: Affiliative demeanor increased significantly with repeated positive cues; neutral and less-affiliative demeanor became prominent with repeated threatening cues (e.g., neutral demeanor was more likely under threatening cue on Day 6B, median estimate = 0.21; 95% HPD: 0.06, 0.41). Aggressive and anxious behaviors decreased over time, especially under the threatening cue protocol.
• Sex Differences: Females exhibited higher baseline approach latency and were more likely to display neutral, less-affiliative behaviors compared to males, an effect consistent across conditions.

Implications for Canine Behavior and Urban Management

These findings substantiate that operant conditioning—mediated by human behavior—strongly shapes the behavioral plasticity and risk/resource calculus of free-ranging dog populations. The partial cross-individual generalization of positive reinforcement, but not of threat, aligns with adaptive strategies for negotiating unpredictable and heterogeneous human-dominated landscapes, supporting the theoretical framework that animals balance fine-grained social learning with risk management.

Of particular significance, the absence of strong negativity bias in cross-individual generalization contradicts the canonical expectation from mammalian learning theory that aversive events produce rapid, indiscriminate avoidance behaviors. Instead, avoidance and caution were nuanced and context-specific, suggesting an evolved strategy for persistent resource acquisition in uncertain anthropogenic environments.

Practical applications include leveraging positive human-dog engagements in stray management and urban welfare interventions, as systematic positive reinforcement may enhance sociability and minimize conflict, while punitive/aversive strategies are unlikely to generate population-level general avoidance and instead may increase hesitancy or stress without deterring approach behaviors toward other humans.

Theoretical and Future Research Relevance

This work provides empirical support for the concept that behavioral plasticity in urban free-ranging canids can be robustly measured and modulated by repeated, cue-specific operant conditioning. The asymmetric generalization observed—positive affect more broadly generalizing across humans than negative affect—has implications for models of social learning, risk assessment, and the maintenance of interspecies cohabitation in cities.

Further investigations could address the persistence of these conditioned behaviors over longer timescales, the interaction of genetic and social learning factors, and expand the analysis to other convergently urbanized taxa. The Bayesian hierarchical modeling approach can serve as a benchmark for future complex behavioral ecology research in heterogeneous social settings.

Conclusion

Indian free-ranging dogs exhibit substantial operant learning in response to repeated positive and threatening human cues. Sociability is enhanced with repeated positive engagement and only moderately attenuated by repeated threats, with learning that is both cue-specific and context-contingent. The asymmetric generalization between valences underscores adaptive behavioral strategies facilitating resource access amid anthropogenic unpredictability, with significant ramifications for animal welfare policy and urban ecological theory.