You can't teach an old dog new tricks

From Dogs (Canis familiaris) stick to what they have learned rather than conform to their conspecifics’ behavior by Markus Germar, Amira Sultan, Juliane Kaminski, and Andreas Mojzisch. From the Abstract:

In recent years, an increasing number of studies has investigated majority influence in nonhuman animals. However, due to both terminological and methodological issues, evidence for conformity in nonhuman animals is scarce and controversial. Preliminary evidence suggests that wild birds, wild monkeys, and fish show conformity, that is, forgoing personal information in order to copy the majority. By contrast, chimpanzees seem to lack this tendency. The present study is the first to examine whether dogs (Canis familiaris) show conformity. Specifically, we tested whether dogs conform to a majority of conspecifics rather than stick to what they have previously learned. After dogs had acquired a behavioral preference via training (i.e., shaping), they were confronted with counter-preferential behavior of either no, one or three conspecifics. Traditional frequentist analyses show that the dogs’ behavior did not differ significantly between the three conditions. Complementary Bayesian analyses suggest that our data provide moderate evidence for the null hypothesis. In conclusion, our results suggest that dogs stick to what they have learned rather than conform to the counter-preferential behavior of others. We discuss the possible statistical and methodological limitations of this finding. Furthermore, we take a functional perspective on conformity and discuss under which circumstances dogs might show conformity after all.

I think what they are saying is that you can't teach an old dog new tricks.