Du­el­ing with De­sire

The bet­ter we un­der­stand want/should con­flict, the more suc­cess­ful we will be at de­sign­ing ef­fec­tive in­ter­ven­tions that shift be­hav­iour.

Rotman Management Magazine - - FRONT PAGE - By T. Brad­ford Bit­terly, Robert Mislavsky, Hengchen Dai and Kather­ine L. Milk­man

IN OUR DAILY LIVES we fre­quently face a ten­sion be­tween what we want to do and what we be­lieve we should do. Af­ter a long week at work, we may want to share an ex­pen­sive din­ner and a few drinks with friends — when we know we should go home early and get a good night’s sleep; sim­i­larly, we might be tempted to get caught up on the cur­rent sea­son of Home­land when we know we should fo­cus on draft­ing that re­port we promised our col­league. For decades, re­searchers have ex­am­ined the bat­tle be­tween highly de­sir­able op­tions that pro­vide im­me­di­ate grat­i­fi­ca­tion (e.g., eat­ing junk food, pro­cras­ti­nat­ing, over­spend­ing) and op­tions that pro­vide more long-term ben­e­fits (e.g., eat­ing healthy food, meet­ing dead­lines and sav­ing for re­tire­ment).

Har­vard Pro­fes­sor Max Baz­er­man calls this com­mon strug­gle ‘want/should con­flict’. In this ar­ti­cle, we will sum­ma­rize the key find­ings on this per­va­sive con­flict and dis­cuss a se­ries of in­ter­ven­tions that pol­i­cy­mak­ers, or­ga­ni­za­tions, and in­di­vid­u­als can use to pro­mote more fu­ture-ori­ented choices.

A Strug­gle for the Ages

In­di­vid­u­als of­ten eval­u­ate de­ci­sions through two dif­fer­ent lenses: The want self fo­cuses my­opi­cally on the here and now, and thus strongly de­sires in­stant grat­i­fi­ca­tion; while the should self is more far-sighted, guided pri­mar­ily by long-term in­ter­ests.

In­dulging in the wants that we most de­sire can cause us to feel waste­ful, ir­re­spon­si­ble and im­moral, and as a re­sult of a dis­taste for such feel­ings, some in­di­vid­u­als un­der-con­sume want op­tions. How­ever, it is far more typ­i­cal for peo­ple to feel that they have made the op­po­site mis­take (i.e., over-in­dulging in wants at the ex­pense of shoulds) and to re­gret this ir­ra­tional be­hav­iour later on.

Over-in­dulging in want op­tions typ­i­cally has a much greater cost than overindulging in should op­tions. For ex­am­ple, fail­ures to con­trol one’s de­sires (e.g., choos­ing pizza over veg­eta­bles, watch­ing TV in­stead of ex­er­cis­ing, smok­ing rather than quit­ting, buy­ing an un­nec­es­sary de­signer hand­bag rather than de­posit­ing the money in a sav­ings ac­count) can con­trib­ute over time to

The want self fo­cuses my­opi­cally on the here and now, and thus strongly de­sires in­stant grat­i­fi­ca­tion.

se­ri­ous in­di­vid­ual and so­ci­etal prob­lems, such as obe­sity, higher can­cer rates, and un­der-sav­ing for re­tire­ment.

In many such cases, then, we can say that ob­served lev­els of in­dul­gence in want op­tions are sub-op­ti­mal (and thus not ra­tio­nal), as higher net util­ity would be ob­tained by se­lect­ing shoulds. Be­cause the mis­take of over-in­dulging in wants is gen­er­ally more com­mon and costly than the op­po­site er­ror, when we dis­cuss want/should con­flict through­out this ar­ti­cle, we will fo­cus our dis­cus­sion on how to in­crease the rate at which should op­tions are se­lected.

Fol­low­ing are eight fac­tors that have been shown to af­fect our choice of wants vs. shoulds.

1. CHOOS­ING FOR NOW OR LATER. A key find­ing is that peo­ple pre­fer should op­tions at a higher rate when mak­ing de­ci­sions for the more dis­tant fu­ture, but pre­fer want op­tions more of­ten the sooner a choice will take ef­fect. For ex­am­ple, de­cid­ing to go to the gym to­mor­row is eas­ier than de­cid­ing to go this very minute, and com­mit­ting to save more for re­tire­ment next year is eas­ier than com­mit­ting to forego a por­tion of to­day’s pay­cheque. This pat­tern has been demon­strated in de­ci­sion do­mains rang­ing from those in­volv­ing money to food and movie rentals.

Mul­ti­ple stud­ies in­ves­ti­gat­ing im­pul­sive­ness have con­firmed that peo­ple show ex­tremely high ‘dis­count rates’ for de­layed re­wards. In one study, the av­er­age par­tic­i­pant opted to re­ceive $50 im­me­di­ately (a want op­tion) rather than $100 in six months (a should op­tion) but pre­ferred to re­ceive $100 in 18 months rather than $50 in 12 months. These re­sults con­tra­dict the pre­dic­tions of stan­dard eco­nomic the­ory, which sug­gests that an in­di­vid­ual’s pref­er­ences be­tween two sure sums of money should de­pend only on the time de­lay that sep­a­rates their re­ceipt (six months in both cases).

These re­sults are con­sis­tent with hyper­bolic time dis­count­ing. In one ex­per­i­ment, sub­jects ran­domly as­signed to se­lect a film to watch that same day were more likely to se­lect low-brow films ( want choices) than sub­jects ran­domly as­signed to se­lect a film they would watch sev­eral days in the fu­ture. And in the do­main of on­line gro­cery shop­ping, one of the au­thors [ Kather­ine Milk­man] found that the per­cent­age of ex­treme should gro­ceries (e.g., fruits and veg­eta­bles) in a cus­tomer’s bas­ket tends to in­crease and the per­cent­age of ex­treme want gro­ceries (e.g., ice cream and cook­ies) tends to de­crease, the fur­ther in ad­vance of de­liv­ery a cus­tomer places her or­der.

2. COG­NI­TIVE LOAD. Daniel Kah­ne­man’s two-sys­tem model of think­ing sug­gests that the rel­a­tive strengths of Sys­tem 1 re­ac­tions (which are au­to­matic, emo­tional and in­stinc­tual) and Sys­tem 2 re­ac­tions (slower, more de­lib­er­a­tive and log­i­cal think­ing) in­flu­ence the out­comes of want/should con­flicts. Build­ing on this no­tion, want op­tions are ex­pected to be more likely to win out when the cog­ni­tive re­sources avail­able to make a de­ci­sion are lim­ited (i.e. when Sys­tem 2 is over-bur­dened), which al­lows Sys­tem 1 to dom­i­nate.

In one study de­signed to test this pre­dic­tion, re­searchers pre­sented par­tic­i­pants with two snack op­tions: A piece of choco­late cake or a cup of fruit salad. They found that in­di­vid­u­als who were ran­domly as­signed to mem­o­rize a seven digit num­ber (and who thus had re­duced cog­ni­tive re­sources) were more likely to choose cake over the fruit than those who were as­signed to mem­o­rize a two digit num­ber. This find­ing high­lights that the avail­abil­ity of cog­ni­tive re­sources is crit­i­cal to mak­ing far-sighted and de­lib­er­a­tive should de­ci­sions.

. CON­STRUAL LEVEL. Con­strual Level The­ory (CLT) sug­gests that we pre­fer shoulds over wants more of­ten when we are think­ing more ab­stractly and thus fo­cus­ing on the global fea­tures of op­tions, rather than when we are think­ing con­cretely and fo­cus­ing on the con­tex­tu­al­ized, sur­face-level and goal-ir­rel­e­vant fea­tures of op­tions. For in­stance, ab­stract rep­re­sen­ta­tions of ‘ex­er­cis­ing’ bring to mind its long-term ben­e­fits, while con­crete rep­re­sen­ta­tions re­mind us of its in-the-mo­ment pains and re­quired plan­ning.

Re­search has shown that the ten­dency to make should choices can be en­hanced by in­duc­ing ab­stract, high-level rep­re­sen­ta­tions of events — for ex­am­ple, by fo­cus­ing peo­ple on more dis­tal events in time and space, rather than more prox­i­mal events. In one study, re­searchers primed par­tic­i­pants to think ab­stractly by ask­ing them to de­scribe why they main­tain good phys­i­cal health, and primed oth­ers to think con­cretely by ask­ing them to de­scribe how they main­tain good phys­i­cal health. They found that when

a more ab­stract, high-level mind­set was ac­ti­vated, peo­ple ex­hib­ited stronger pref­er­ences for im­me­di­ate want out­comes over de­layed should re­wards. In­duc­ing peo­ple to adopt a higher-level ‘con­strual mind­set’ is there­fore one way to in­crease fu­ture ori­ented de­ci­sion mak­ing.

4. DE­PLE­TION. As in­di­cated, a grow­ing body of re­search sug­gests that ex­ert­ing willpower comes at a cost, and that cost is a re­duc­tion in avail­able self-con­trol re­sources for use in fu­ture choices. In other words, in­di­vid­u­als have lim­ited self-reg­u­la­tory re­sources, and ex­ert­ing self-con­trol to avoid wants in one sit­u­a­tion can de­crease one’s sub­se­quent abil­ity to ex­ert self-con­trol.

In one study de­signed to test this the­ory, re­searchers showed that par­tic­i­pants who re­sisted eat­ing choco­late chip cook­ies (an ob­vi­ous want for most peo­ple) quit work­ing on un­solv­able puz­zles ear­lier (where per­sis­tence is a should be­hav­iour) than did in­di­vid­u­als who re­sisted eat­ing radishes — an ac­tiv­ity that for most peo­ple, re­quires lit­tle self-con­trol. This and other stud­ies il­lus­trate the para­dox that by ex­er­cis­ing self-con­trol now, we in­crease the like­li­hood that we will give in to our de­sires to in­dulge later. For­tu­nately, al­though the self-con­trol mus­cle can be weak­ened through re­peated use, it can also be strength­ened through proper ex­er­cise: One study found ev­i­dence that prac­tic­ing small acts of self-con­trol greatly in­creased smok­ers’ chances of suc­cess­fully quit­ting.

5. JOINT VS. SEP­A­RATE EVAL­U­A­TIONS. The out­comes of want vs. should con­flicts are also in­flu­enced by whether we eval­u­ate op­tions one-at-a-time or si­mul­ta­ne­ously. Al­though want op­tions tend to be pre­ferred at a higher rate than should op­tions in iso­la­tion, we are more likely to think about the costs and ben­e­fits of each op­tion and make far-sighted choices when mul­ti­ple op­tions are eval­u­ated at the same time. For ex­am­ple, when viewed in iso­la­tion, a char­ity that saves baby po­lar bears may seem more al­lur­ing and re­ceive more do­na­tions than a char­ity that funds skin can­cer re­search. How­ever, when these choices are com­pared side-by-side, peo­ple tend to do­nate to the char­ity that helps peo­ple, view­ing its mis­sion as more im­por­tant, al­beit less emo­tion­ally res­o­nant. This re­search high­lights that pre­sent­ing want and should op­tions si­mul­ta­ne­ously rather than se­quen­tially is one way to pro­mote more should choices.

6. MOOD EF­FECTS. Pos­i­tive mood has been shown to fa­cil­i­tate fu­ture-ori­ented de­ci­sion mak­ing, and a num­ber of ex­pla­na­tions have been pro­posed for this. One ac­count is that ex­pe­ri­enc­ing pos­i­tive emo­tions sig­nals to de­ci­sion mak­ers that their cur­rent sit­u­a­tion is non-threat­en­ing, which re­duces dis­count­ing of the fu­ture and thus makes shoulds rel­a­tively more at­trac­tive.

Another rea­son is that pos­i­tive emo­tions can coun­ter­act ‘ego de­ple­tion’, restor­ing the de­pleted willpower re­sources nec­es­sary for se­lect­ing should op­tions. In one study, re­searchers demon­strated that af­ter ran­domly as­sign­ing par­tic­i­pants to watch clips of videos that in­duced ei­ther a happy or neu­tral mood, those placed in a pos­i­tive mood were more likely to se­lect a should op­tion (grapes) over a want op­tion (M&MS). How­ever, they found that the ten­dency for pos­i­tive moods to in­crease should choices is at­ten­u­ated by el­e­vated arousal be­cause arousal is de­plet­ing.

Re­search has also ex­plored the im­pact of neg­a­tive emo­tions on want/should con­flict. Re­cent stud­ies have demon­strated that these emo­tions can lead to self-con­trol break­downs. For in­stance, sad­ness in­creases de­ci­sion mak­ers’ ten­dency to fo­cus on im­me­di­ate grat­i­fi­ca­tion and to dra­mat­i­cally dis­count fu­ture out­comes. To­gether, this re­search shows that peo­ple who are re­laxed and happy are more likely to make should choices, whereas in­di­vid­u­als who are emo­tion­ally aroused or in a neg­a­tive mood are more likely to reach for in­stant grat­i­fi­ca­tion.

7. LI­CENS­ING EF­FECTS. In­ter­est­ingly, our choice be­tween a want and a should can also be af­fected by de­ci­sions we have made in the past, as well as those we an­tic­i­pate mak­ing in the fu­ture. Re­search shows that peo­ple feel ‘li­censed’ to make (or jus­ti­fied in mak­ing) want choices if they be­lieve they have pre­vi­ously en­gaged in should be­hav­iours or if they an­tic­i­pate hav­ing op­por­tu­ni­ties to en­gage in should be­hav­iours in the fu­ture.

For ex­am­ple, peo­ple who were asked to imag­ine they would par­take in a should be­hav­iour (e.g., do­nat­ing part of their tax re­bate to char­ity, vol­un­teer­ing for com­mu­nity ser­vice), rel­a­tive to a con­trol group who did not imag­ine any such fu­ture good be­hav­iour, were more likely to se­lect a de­sir­able want prod­uct (e.g., a pair of de­signer glasses) over a cog­ni­tively favourable should prod­uct (e.g., a less ex­pen­sive but more util­i­tar­ian pair of glasses).

Fur­ther­more, want prod­ucts are more likely to be se­lected when in­di­vid­u­als make what they be­lieve is the first of a se­ries of sim­i­lar de­ci­sions rather than a sin­gle, iso­lated choice, pre­sum­ably be­cause in­di­vid­u­als mak­ing re­peated de­ci­sions be­lieve that they will have the op­por­tu­nity to choose shoulds in the fu­ture to com­pen­sate for cur­rent in­dul­gences. These find­ings high­light that choos­ing be­tween wants and shoulds is of­ten not done in iso­la­tion, but in­stead, hinges on an in­di­vid­ual’s past choices and an­tic­i­pated fu­ture de­ci­sions.

8. FRESH STARTS. Re­cent re­search sug­gests that there are nat­u­rally-aris­ing points in time when peo­ple are par­tic­u­larly mo­ti­vated to pur­sue their long-term in­ter­ests and pre­fer shoulds. Tem­po­ral land­marks, which in­clude per­son­ally-rel­e­vant life events (e.g., an­niver­saries, birth­days) and ref­er­ence points on shared cal­en­dars (e.g., hol­i­days, the start of a new week, month, year or se­mes­ter), de­mar­cate the pas­sage of time and help us or­ga­nize our ac­tiv­i­ties and ex­pe­ri­ences. Field re­search has shown that tem­po­ral land­marks mag­nify peo­ple’s vir­tu­ous in­ten­tions and in­crease their en­gage­ment in should be­hav­iours. Re­searchers an­a­lyzed (a) daily Google search vol­ume for the term ‘diet’, (b) un­der­grad­u­ate stu­dents’ gym at­ten­dance records, and (c) a wide range of goals (e.g., per­tain­ing to ed­u­ca­tion, health, fi­nance, etc.) that In­ter­net users com­mit­ted to pur­su­ing on a goal-set­ting web­site ( stickk.com). These stud­ies re­vealed that peo­ple en­gage in should be­hav­iours (i.e., di­et­ing, ex­er­cis­ing and goal pur­suit) more fre­quently fol­low­ing tem­po­ral land­marks, in­clud­ing the start of the week, month, year and aca­demic se­mes­ter, as well as im­me­di­ately fol­low­ing a birth­day, a na­tional hol­i­day, or a school break. The au­thors re­fer to this phe­nom­e­non as ‘the fresh start ef­fect’.

Should Pre­scrip­tions

Faced with the above chal­lenges to our de­ci­sion mak­ing, what can be done? Fol­low­ing are three ‘nudges’ that have been shown to suc­cess­fully in­crease the rate at which peo­ple choose shoulds over wants.

PRE­SCRIP­TION 1: PROMPTS Prompt­ing peo­ple to stip­u­late when, where and how they will en­act their goals is one of the old­est pre­scrip­tions for in­creas­ing en­gage­ment in should be­hav­iours — dat­ing back to re­search con­ducted in the 1960s. Plan-mak­ing has been shown to im­prove our like­li­hood of achiev­ing goals in a di­verse ar­ray of do­mains, in­clud­ing ex­er­cise, di­et­ing, smok­ing ces­sa­tion, aca­demic per­for­mance, test prepa­ra­tion, re­cy­cling and vot­ing.

Plan­ning prompts are ef­fec­tive for a num­ber of rea­sons, one of which is that they re­duce for­get­ful­ness. When peo­ple take the time to cre­ate and even write down the when, where and how of a plan, they men­tally as­so­ciate their tar­get ac­tions with cues re­lat­ing to the when and where of ex­e­cu­tion.

Plan­ning also dis­cour­ages pro­cras­ti­na­tion by cre­at­ing ex­plicit com­mit­ments to one­self and some­times to oth­ers. For ex­am­ple, peo­ple feel in­ter­nal pres­sure to fol­low-through on their plans and seek to avoid break­ing ex­plicit com­mit­ments to them­selves be­cause be­hav­ing in­con­sis­tently with their past ac­tions, be­liefs, and at­ti­tudes cre­ates dis­com­fort. Fur­ther, some plans (e.g., to get a mam­mo­gram) may lit­er­ally re­quire mak­ing an ap­point­ment, which may be dif­fi­cult to can­cel or de­lay.

Re­cent large-scale stud­ies have demon­strated the ef­fec­tive­ness of plan-mak­ing as a means of in­creas­ing take-up of two im­por­tant should be­hav­iours: re­ceiv­ing flu shots and colono­scopies. In a study by one of the au­thors (Prof. Milk­man), plan­ning prompts were demon­strated to sig­nif­i­cantly in­crease take-up of colono­scopies. In this study, those pre­dicted to be the most likely to for­get to fol­low through (e.g., older adults, adults with chil­dren, and those who did not com­ply with pre­vi­ous re­minders) ben­e­fit­ted most from the plan­ning prompt.

PRE­SCRIP­TION 2: COMMITTMENT DE­VICES Many peo­ple are so­phis­ti­cated about pre­vent­ing their self-con­trol prob­lems from get­ting in the way of their good (or should) in­ten­tions. As a re­sult, another way to in­crease en­gage­ment in shoulds is by pro­vid­ing in­di­vid­u­als with ac­cess to com­mit­ment de­vices — or a means of vol­un­tar­ily (a) en­forc­ing re­stric­tions on them­selves un­til they have done what they know they should or (b) im­pos­ing penal­ties for fail­ing to do what they should. Com­mit­ment de­vices have ex­isted in many forms through­out the years. For in­stance, the piggy bank is a com­mit­ment de­vice that en­cour­ages us to com­mit to sav­ing by set­ting aside a cer­tain por­tion of earn­ings for fu­ture use.

More mod­ern forms of com­mit­ment de­vices in­clude Antabuse, a med­i­ca­tion that makes al­co­holics phys­i­cally ill af­ter con­sum­ing even a small amount of al­co­hol, and the afore­men­tioned stickk.com, a web­site that takes users’ money if they fail to achieve their goals. Ul­ti­mately, com­mit­ment de­vices are mech­a­nisms that al­low peo­ple to pre­vent them­selves from giv­ing in to

un­wise wants, and the re­search shows that they can be used to re­duce pro­cras­ti­na­tion, smok­ing, fail­ures to achieve work goals, and suc­cumb­ing to re­peated temp­ta­tions in a lab­o­ra­tory set­ting.

PRE­SCRIP­TION 3: TEMP­TA­TION BUNDLING Temp­ta­tion bundling seeks to in­crease should be­hav­iours by bundling them with tempt­ing wants — a strat­egy that can si­mul­ta­ne­ously re­duce en­gage­ment in wants and in­crease en­gage­ment in shoulds. For ex­am­ple, a doc­toral stu­dent may have the goal of spend­ing more time writ­ing a man­u­script (a should be­hav­iour) while rec­og­niz­ing that he has been con­sum­ing too many Star­bucks white choco­late mochas (a want be­hav­iour). Us­ing temp­ta­tion bundling, the stu­dent might com­mit to only con­sum­ing white choco­late mochas while work­ing on his man­u­script, thus in­creas­ing time spent writ­ing and re­duc­ing white choco­late mocha con­sump­tion. In ad­di­tion to si­mul­ta­ne­ously tack­ling two types of self-con­trol prob­lems, temp­ta­tion bundling has the po­ten­tial to har­ness con­sump­tion com­ple­men­tar­i­ties: Work­ing while drink­ing mochas may make work more en­joy­able and ef­fi­cient, as well as re­duc­ing the guilt (and there­fore over­all en­joy­ment) as­so­ci­ated with choco­late mocha con­sump­tion.

Temp­ta­tion bundling has proven an ef­fec­tive means of in­creas­ing en­gage­ment in one im­por­tant should be­hav­iour: ex­er­cise. Prof. Milk­man et al. demon­strated the ef­fec­tive­ness of temp­ta­tion bundling in a field ex­per­i­ment where study par­tic­i­pants were ran­domly as­signed to one of three con­di­tions:

(a) A full-treat­ment con­di­tion, in which ac­cess to tempt­ing, low­brow au­dio nov­els (wants) was re­stricted to the gym; (b) An in­ter­me­di­ate treat­ment con­di­tion, in which par­tic­i­pants were sim­ply en­cour­aged to self-re­strict their en­joy­ment of tempt­ing au­dio nov­els to the gym; or (c) A con­trol con­di­tion.

Their find­ing: Ini­tial gym at­ten­dance among in­di­vid­u­als in the full treat­ment con­di­tion was 51 per cent higher than at­ten­dance in the con­trol group, and par­tic­i­pants in the in­ter­me­di­ate treat­ment con­di­tion showed a marginally sig­nif­i­cant 29 per cent ini­tial in­crease in gym at­ten­dance. At the con­clu­sion of the study, 61 per cent of par­tic­i­pants were will­ing to pay to have their ac­cess to an ipod con­tain­ing tempt­ing au­dio nov­els re­stricted to the gym. In other words, peo­ple would pay to have ac­cess to a pos­ses­sion they could oth­er­wise use freely re­stricted so they could only en­joy this de­sir­able want while ex­er­cis­ing (i.e. en­gag­ing in a should be­hav­iour).

In clos­ing

The bet­ter we un­der­stand want/should con­flict, the more suc­cess­ful we will be­come at de­sign­ing ef­fec­tive in­ter­ven­tions that pro­mote should choices and help peo­ple avoid the temp­ta­tion to give in to harm­ful crav­ings and de­sires. As in­di­cated herein, even mi­nor in­ter­ven­tions can shift be­hav­iours in ben­e­fi­cial di­rec­tions.

T. Brad­ford Bit­terly and Robert Mislavsky are PHD can­di­dates in De­ci­sion Sci­ences at the Whar­ton School at the Univer­sity of Penn­syl­va­nia. Hengchen Dai is an As­sis­tant Pro­fes­sor of Or­ga­ni­za­tional Be­hav­iour at the Olin Busi­ness School, Wash­ing­ton...

Newspapers in English

Newspapers from Canada

© PressReader. All rights reserved.