Thanks, that's a good find.

I'm unsure about the methodology of that experiment - it seems that the survey asked people to list details about their pet first, which could potentially 'pre-load' a fond mental image of a particular childhood pet and affect later answers, even though such a mental image would never arise when someone was going about their day-to-day meat shopping.

Nevertheless, the results are interesting and I think that more work needs to be done in this area.

Can I also suggest though that this tendency should be balanced by the potential positive impact of owning a pet. Jonathan Safran Foer quotes Oxford historian Sir Keith Thomas:

the spread of pet-keeping... created the psychological foundation for the view that some animals at least were entitled to moral consideration.

I've no idea about whether studies of this have been done, but it seems plausible at least that a child in a pet-owning family is more likely to develop some level of empathy for animals, and more likely to question the hypocrisy of treating cats and dogs so well and other animals so abysmally.

In pre- domestic pet societies such as most of the second and third worlds today, empathetic attitudes towards animals are often non-existent (except when there are religions focussed on alleviation of suffering such as in India). There's just little basis at all for caring about the experience of animals. Thus pets may help train moral consideration for animals, and may be worth the meat-consumption tradeoff that they require.

But that's a very tentative may... the likelihood of children in pet-owning families becoming vegetarian or making pro-animal interventions is probably so marginal that it doesn't counteract all the extra chickens that their cat or dog needs.

Interesting point. I suppose that Peter did pre-emptively respond to it when he noted that "it is still premature to do lots of studies on the relative effectiveness of certain vegetarian messaging ... when we don’t even know if the absolute effectiveness is there yet."

Furthermore, this would probably be really difficult to detect, as ads which aim to reduce animal product consumption now might actually be the most potent vector for effecting long-term value shifts - people who start integrating some vegetarianism into their diets are more likely to come across and even spread information showing the intelligence of factory farmed animals.

Given the problems faced by this study, I doubt we'll have a clear answer to the real long-term effectiveness of various interventions any time soon. The best we can do at the moment is to try a combination of methods that appeal to diverse moral intuitions and interests.