Concrete next steps for ageing-based welfare measures

Re­lated: Assess­ing bio­mark­ers of age­ing as mea­sures of cu­mu­la­tive an­i­mal welfare

Sev­eral peo­ple have asked me how they or oth­ers can take the ideas from my pre­vi­ous post on age­ing-based mea­sures of an­i­mal welfare for­ward and ap­ply them to helping an­i­mals. Since I’ve now left WAI and won’t be work­ing on this fur­ther, I thought I’d draw a line un­der my in­volve­ment by quickly sketch­ing out some ways I can see fur­ther work on this be­ing valuable. This is not an or­ganised re­search agenda; rather, it is a rel­a­tively un­pri­ori­tised and in­for­mal list of things I would like to see hap­pen in this space.

I would ex­pect use­ful fu­ture work on age­ing-based welfare mea­sures (hence­forth AWMs) to fall into four broad cat­e­gories:

  1. Fur­ther de­vel­op­ment of biolog­i­cal age­ing mea­sures that are suit­able for use as AWMs

  2. The­o­ret­i­cal or ex­per­i­men­tal work in­ves­ti­gat­ing the scope and limi­ta­tions of AWMs

  3. Ap­ply­ing AWMs to in­ves­ti­gat­ing ac­tive un­cer­tain­ties in an­i­mal welfare

  4. Outreach to spread aware­ness of AWMs among groups that could make use of them or de­velop them further

I’ll ad­dress each of these sep­a­rately, but I ex­pect many pro­jects could at­tack two or more of these cat­e­gories si­mul­ta­neously.

1. Devel­op­ment of new age­ing mea­sures in animals

What is it?

There are many differ­ent ways of mea­sur­ing biolog­i­cal age. Of these, telomere length seems to be by far the most widely used in non­hu­man an­i­mals. As far as I know, telomere length and hip­pocam­pal vol­ume (which I know much less about) are the only age­ing met­rics that have been ap­plied speci­fi­cally as mea­sures of cu­mu­la­tive welfare.

I ex­pect that other mea­sures of biolog­i­cal age­ing may turn out to be at least as use­ful as telomere length, and pos­si­bly more so in con­texts where telomere length performs rel­a­tively poorly. It would there­fore be good to see more work done to eval­u­ate other bio­mark­ers of age­ing as po­ten­tial welfare mea­sures (in terms of com­pre­hen­sive­ness, cost, trans­fer­abil­ity, etc.), and to de­velop new, bet­ter ways of com­bin­ing differ­ent bio­mark­ers into effec­tive com­pos­ite mea­sures.

One rea­son I think it would be par­tic­u­larly valuable for peo­ple with a spe­cific in­ter­est in an­i­mal welfare to work in this space is be­cause they will have differ­ent pri­ori­ties from re­searchers whose pri­mary goal is ad­vanc­ing hu­man biomed­i­cal sci­ence. In par­tic­u­lar, they will place greater weight on the cost and trans­fer­abil­ity of po­ten­tial AWMs, as these will be of par­tic­u­lar im­por­tance in fund­ing-con­strained an­i­mal con­texts. Hence, state-of-the-art AWMs may well differ sub­stan­tially from the meth­ods that are cur­rently best re­garded in the age­ing field.

Why is it valuable?

  • I ex­pect differ­ent bio­mark­ers of age­ing to have differ­ent pros and cons as mea­sures of welfare. Fur­ther ex­plo­ra­tion of this space could un­cover bio­mark­ers that are com­ple­men­tary or su­pe­rior to telomere length in this con­text.

  • Bet­ter meth­ods of com­bin­ing bio­mark­ers could al­low re­searchers to ob­tain bet­ter mea­sures of biolog­i­cal age at lower costs.

  • Bet­ter un­der­stand­ing of the cost and trans­fer­abil­ity of these mea­sures would help welfare sci­en­tists and or­gani­sa­tions bet­ter se­lect which mea­sures they want to use in their species of in­ter­est, low­er­ing the cost of en­try to us­ing these meth­ods.

  • Dis­cov­ery and val­i­da­tion of new age­ing-based welfare mea­sures would help to val­i­date the ap­proach as a whole and en­courage its use in more species and con­texts.

Who could do it?

A lot of the work in this area prob­a­bly re­quires ac­cess to biolog­i­cal lab­o­ra­tory fa­cil­ities and so would be re­stricted to re­searchers with that ac­cess, pri­mar­ily aca­demics. Nev­er­the­less, ini­tial work to iden­tify promis­ing bio­mark­ers could be done through liter­a­ture re­views and other ap­proaches that do not re­quire ac­cess to labs, and I ex­pect de­vel­op­ment of new com­pos­ite mea­sures to benefit from data-sci­ence and ma­chine-learn­ing skills that are rare in tra­di­tional biolog­i­cal re­search groups, so there is definitely scope for oth­ers to con­tribute here as well.

2. Ex­plor­ing the age­ing-welfare connection

What is it?

The idea of mea­sur­ing cu­mu­la­tive welfare us­ing biolog­i­cal age­ing is new, and the liter­a­ture on the topic is small. It is still un­clear ex­actly how broad the scope of these meth­ods is or which ap­proaches to mea­sur­ing age­ing for welfare are most promis­ing. There is there­fore broad scope for both the­o­ret­i­cal and em­piri­cal re­search to provide valuable in­for­ma­tion on the po­ten­tial of AWMs in gen­eral.

Some ques­tions I would like to see ad­dressed in the fu­ture in­clude:

  • Can the the­o­ret­i­cal ar­gu­ment for the age­ing-welfare con­nec­tion be de­vel­oped into some­thing more for­mal­ised (and there­fore po­ten­tially more testable)?

  • Can the the­o­ret­i­cal ar­gu­ment be ex­tended to con­vinc­ingly cover so­cial stresses (foot­note 15 in the origi­nal re­port)?

  • Is it pos­si­ble to adapt age­ing-based welfare meth­ods to ap­ply be­tween groups that differ in ge­netic com­po­si­tion, such as differ­ent strains of boiler hens?

  • Sev­eral peo­ple, in­clud­ing me, have wor­ried that these meth­ods might un­der­weight the im­por­tance of rare acute events. Is there a way to work out if these wor­ries are well-founded?

  • What meth­ods might be best for mea­sur­ing the cu­mu­la­tive welfare of ju­ve­niles ver­sus adults?

  • How might age­ing-based welfare meth­ods be ap­plied in in­sects or other in­ver­te­brates? How might we go about val­i­dat­ing these ap­pli­ca­tions?

  • Can age­ing-based welfare meth­ods be ap­plied to an­i­mals with un­usual age­ing tra­jec­to­ries?

  • How well do age­ing-based meth­ods in­cor­po­rate pos­i­tive ex­pe­riences? Is it even the­o­ret­i­cally pos­si­ble to ad­dress this ques­tion?

Which of these ques­tions is of high­est pri­or­ity de­pends on one’s fo­cus; the val­idity of AWMs in in­ver­te­brates and an­i­mals with un­usual age­ing tra­jec­to­ries is much more im­por­tant for those in­ter­ested in wild-an­i­mal welfare, for ex­am­ple, while is­sues of ge­netic com­po­si­tion will be es­pe­cially im­por­tant for those in­ter­ested in com­par­ing the welfare of differ­ent do­mes­ti­cated strains.

Why is it valuable?

  • An­swer­ing these ques­tions would help give us a much bet­ter idea of the scope of AWMs in gen­eral: when they do and don’t ap­ply and when they might be bet­ter or worse than al­ter­na­tive meth­ods. This would help re­searchers and ac­tivists make more in­formed choices about when to use them.

  • Work on these is­sues could also help clar­ify what pro­duc­tive ap­proaches to the other two ar­eas of re­search dis­cussed here (de­vel­op­ing new mea­sures and ap­ply­ing them in prac­tice) are likely to be.

Who could do it?

Some of these ques­tions could be best ad­dressed by a the­o­ret­i­cal anal­y­sis or liter­a­ture re­view, while oth­ers would re­quire novel biolog­i­cal ex­per­i­ments or a com­bi­na­tion of differ­ent ap­proaches. In­di­vi­d­ual re­searchers in an aca­demic con­text, such as Masters stu­dents, would be well-placed to ad­dress many of these ques­tions. An­i­mal-fo­cused EA re­searchers could also likely make head­way in this space if they are an­a­lyt­i­cally minded and have (or can ac­quire) a rea­son­able ground­ing in evolu­tion­ary biol­ogy.

3. Ap­ply­ing AWMs to welfare questions

What is it?

It would be good to start see­ing ap­pli­ca­tions of ex­ist­ing age­ing-based welfare mea­sures to un­cer­tain­ties in an­i­mal welfare. WAI is in the pro­cess of de­vel­op­ing pro­jects that ap­ply these tech­niques to wild-an­i­mal welfare, but I ex­pect many of the most valuable early ap­pli­ca­tions to re­late to the welfare of cap­tive an­i­mal pop­u­la­tions.

Some key un­cer­tain­ties I am aware of in this space in­clude the rel­a­tive welfare of caged ver­sus cage-free chick­ens, the value of differ­ent ap­proaches to im­prov­ing the lives of farmed fish, and the rel­a­tive welfare im­pacts of differ­ent forms of tag­ging and sam­pling in farmed and lab­o­ra­tory pop­u­la­tions. I am sure peo­ple more ac­tively in­volved in the an­i­mal-welfare move­ment would be aware of oth­ers.

The de­sign of these ex­per­i­ments could be quite sim­ple. Many could sim­ply fol­low the tem­plate I set out in the origi­nal post: get two or more groups of an­i­mals liv­ing un­der differ­ent con­di­tions, mea­sure the biolog­i­cal age of in­di­vi­d­u­als of differ­ent chronolog­i­cal ages, and plot the biolog­i­cal age­ing curve for each pop­u­la­tion.

Why is it valuable?

  • The pur­pose of de­vel­op­ing new welfare mea­sures is to let us bet­ter as­sess the welfare con­di­tions of an­i­mals, so ap­ply­ing age­ing-based meth­ods in this way is di­rectly valuable to the cause of an­i­mal welfare.

  • Ac­tu­ally do­ing stud­ies us­ing these meth­ods would help take them from the realm of the­ory into ac­tual prac­ti­cal use­ful­ness, and so per­haps at­tract more at­ten­tion and ca­pac­ity from other welfare sci­en­tists and welfare or­gani­sa­tions.

  • It is difficult to pre­dict how a method will fare in prac­tice un­til it has ac­tu­ally been tried. Or­ganis­ing these ex­per­i­ments would gen­er­ate im­por­tant prac­ti­cal knowl­edge that could then be shared with other in­ter­ested par­ties.

Who could do it?

Perform­ing these ex­per­i­ments would re­quire a mix­ture of skills in­clud­ing an­i­mal han­dling, molec­u­lar biol­ogy and statis­ti­cal anal­y­sis, as well as ac­cess to an­i­mal pop­u­la­tions of in­ter­est and suit­able lab­o­ra­tory fa­cil­ities. Ac­tive col­lab­o­ra­tions with co-op­er­a­tive farms (or other an­i­mal fa­cil­ities) and aca­demic groups with the req­ui­site ex­per­tise are likely to be crit­i­cal.

Though I ex­pect rel­a­tively few an­i­mal welfare or­gani­sa­tions will have the nec­es­sary lab­o­ra­tory ex­per­tise or fa­cil­ities to perform these ex­per­i­ments in-house, I can cer­tainly en­vi­sion a non­profit or­gani­sa­tion with an em­piri­cal fo­cus and in-house data-anal­y­sis skills part­ner­ing with aca­demic or in­dus­try col­lab­o­ra­tors to perform this work.

4. Spread­ing aware­ness of AWMs

What is it?

Cur­rently, the the gen­eral level of aware­ness around AWMs is very low. I’m aware of a very small num­ber of aca­demics (and one non­profit, WAI) do­ing re­search ex­plic­itly in this area, and a small num­ber of other or­gani­sa­tions have ex­pressed ten­ta­tive in­ter­est. This is a pre­car­i­ous situ­a­tion; given this level of aware­ness, it would not be too sur­pris­ing if these meth­ods fail to be taken up and used widely in fu­ture.

High-qual­ity out­reach about these meth­ods, which ex­plains why they are po­ten­tially so valuable and how they could be of use to welfare sci­en­tists and or­gani­sa­tions, there­fore has the po­ten­tial to be highly im­pact­ful, es­pe­cially if fur­ther re­search sug­gests AWMs will live up to their cur­rent promise. This is what I am try­ing to do with these posts, and it is my hope that the word will spread fur­ther.

Why is it valuable?

  • The more re­searchers, or­gani­sa­tions and fun­ders are aware of the po­ten­tial of AWMs, the more likely it is that fur­ther ex­plo­ra­tion and ap­pli­ca­tion of these meth­ods will be funded and performed.

Who could do it?

This may be an area where EA-al­igned an­i­mal-welfare or­gani­sa­tions could be par­tic­u­larly im­pact­ful, as they already have ex­pe­rience with out­reach and con­nec­tions with other or­gani­sa­tions who could po­ten­tially make great use out of AWMs. Fund­ing bod­ies in­ter­ested in this area could also have an im­pact by speci­fi­cally ad­ver­tis­ing their in­ter­est in pro­pos­als to ad­vance this area (or other promis­ing new meth­ods for mea­sur­ing welfare), there­fore si­mul­ta­neously rais­ing aware­ness and cre­at­ing the op­por­tu­nity for fur­ther progress to be made.

What can a gen­er­al­ist do?

EA is long on gen­er­al­ist re­searchers and short on biol­o­gists. Many of the promis­ing di­rec­tions I’ve dis­cussed here re­quire ac­cess to a lab­o­ra­tory, or at least to biolog­i­cal ex­per­tise. How­ever, there are sev­eral top­ics I think could benefit from im­me­di­ate at­ten­tion from an­i­mal-fo­cused gen­er­al­ist re­searchers and or­gani­sa­tions.

Firstly, as a re­searcher at WAI, my think­ing about AWMs has been fo­cused on their ap­pli­ca­tions to wild-an­i­mal welfare. While I have dis­cussed some ex­am­ples of where I think these meth­ods could be ap­plied to do­mes­ti­cated an­i­mals, a more thor­ough re­view by some­one with more knowl­edge of that area could be valuable, both to at­tract at­ten­tion to the topic in that space and to help tar­get fu­ture re­sources more effec­tively.

Se­condly, given some read­ing up on evolu­tion­ary biol­ogy and com­par­a­tive psy­chol­ogy, I think progress could be made on many of the ques­tions in sec­tion 2 with­out ac­cess to a lab­o­ra­tory, both through fo­cused liter­a­ture re­views and greater the­o­ret­i­cal scrutiny. A bet­ter idea of the strength and scope of the the­o­ret­i­cal un­der­pin­nings of AWMs would be very valuable for di­rect­ing re­search in the fu­ture. Topics I think could es­pe­cially benefit from this in­clude age­ing in ju­ve­niles, the welfare-mea­sure­ment po­ten­tial of age­ing bio­mark­ers in in­sects and other in­ver­te­brates, and whether or not it is pos­si­ble to use these meth­ods to as­sess the welfare effects of ge­netic differ­ences be­tween groups. Progress could prob­a­bly also be made on the rel­a­tive po­ten­tial of differ­ent cur­rently-ex­ist­ing age­ing met­rics for use as AWMs in an­i­mals.

Fi­nally, many an­i­mal-fo­cused EA orgs have strong con­nec­tions with other an­i­mal-welfare or­gani­sa­tions, re­searchers and fun­ders. Ac­tively spread­ing the word about the po­ten­tial of AWMs within their net­work rep­re­sents one of best routes I can see to achiev­ing more gen­eral aware­ness of, and in­ves­ti­ga­tion into, these promis­ing new meth­ods.