"Long before it's in the papers"
January 27, 2015

RETURN TO THE WORLD SCIENCE HOME PAGE


Putting the squeeze on suspects: scientists refine methods to catch deception

May 11, 2011
Courtesy of UCLA
and World Science staff

When some­one is act­ing sus­pi­ciously at an air­port, sub­way sta­t­ion or oth­er pub­lic space, how can po­lice fig­ure out wheth­er he’s up to no good?

A group of re­search­ers has been an­a­lyz­ing years of past stud­ies—in­cluding their own—on de­cep­tion de­tec­tion, and this week are re­port­ing what they de­scribe as some of the most up-to-date guide­lines on catch­ing liars in the act with­out pol­y­graph or oth­er phys­i­cal tests. The guide­lines are geared to­ward use in law en­force­ment and interro­gation situations in which sus­pects or wit­nesses are be­ing ques­tioned.

A group of re­search­ers has been an­a­lyz­ing years of past stud­ies—in­cluding their own—on de­cep­tion de­tec­tion, and this week are re­port­ing what they de­scribe as some of the most up-to-date guide­lines on catch­ing liars in the act with­out pol­y­graph or oth­er phys­i­cal tests. (Courtesy FBI)


Univers­ity of Cal­i­for­nia Los An­ge­les psy­chol­o­gist R. Ed­ward Geisel­man, who has taught in­ves­ti­ga­tive in­ter­view­ing tech­niques to law en­forcers for years, and col­leagues re­port the find­ings in the April is­sue of the Amer­i­can Jour­nal of Fo­ren­sic Psy­chi­a­try.

The more re­li­a­ble signs of de­ceit, Gei­sel­man said, in­clude:
  • De­cep­tive peo­ple gen­er­ally want to say as lit­tle as pos­si­ble un­der ques­tion­ing. Geisel­man in­i­tially thought they would tell an elab­o­rate sto­ry, but dis­cov­ered, he said, that the vast ma­jor­ity give only the bare-bones. Stud­ies with col­lege stu­dents, as well as pris­on­ers, show this, he added. Geisel­man’s in­ves­ti­ga­tive in­ter­view­ing tech­niques are de­signed to get peo­ple to talk.

  • Al­though de­cep­tive peo­ple don’t say much, they tend to spon­ta­ne­ously give a jus­tifica­t­ion for what lit­tle they are say­ing.

  • They tend to re­peat ques­tions be­fore an­swer­ing them, per­haps to give them­selves time to con­coct an an­swer.

  • They of­ten mon­i­tor the lis­ten­er’s re­ac­tion to what they are say­ing. “They try to read you to see if you are buy­ing their sto­ry,” Geisel­man said.

  • They of­ten in­i­tially slow down their speech be­cause they have to cre­ate their sto­ry and mon­i­tor your re­ac­tion, and when they have it straight “will spew it out faster,” Geisel­man said. Truth­ful peo­ple are not both­ered if they speak slow­ly, but de­cep­tive peo­ple of­ten think slow­ing their speech down may look sus­pi­cious. “Truth­ful peo­ple will not dra­mat­ic­ally al­ter their speech rate with­in a sin­gle sen­tence,” he said.

  • They tend to use sen­tence frag­ments more fre­quently than truth­ful peo­ple; of­ten, they will start an an­swer, back up and not com­plete the sen­tence.

  • They’re more likely to press their lips when asked a sen­si­tive ques­tion and are more likely to play with their hair or en­gage in oth­er “groom­ing” be­hav­iors. Ges­tur­ing to­ward one’s self with the hands tends to be a sign of de­cep­tion; ges­tur­ing out­wardly is not.

  • Truth­ful peo­ple, if chal­lenged about de­tails, will of­ten de­ny that they are ly­ing and ex­plain even more, while de­cep­tive peo­ple gen­er­ally will not pro­vide more spe­cif­ics.

  • When asked a hard ques­tion, truth­ful peo­ple will of­ten look away be­cause the ques­tion re­quires con­centra­t­ion, while dis­hon­est peo­ple will look away only brief­ly, if at all, un­less it is a ques­tion that should re­quire in­tense con­centra­t­ion.

If dis­hon­est peo­ple try to mask these nor­mal re­ac­tions to ly­ing, they would be even more ob­vi­ous, Geisel­man said. Among the tech­niques he teaches to en­a­ble de­tec­tives to tell the truth from lies are:

  • Have peo­ple tell their sto­ry back­wards, start­ing at the end and sys­tem­at­ic­ally work­ing their way back. In­struct them to be as com­plete and de­tailed as they can. This tech­nique, part of a “cog­ni­tive in­ter­view” Geisel­man co-de­vel­oped with Ronald Fish­er, a form­er UCLA psy­chol­o­gist now at Flor­i­da In­terna­t­ional Un­ivers­ity, “in­creases the cog­ni­tive load to push them over the edge.” A de­cep­tive per­son, even a “pro­fes­sional liar,” is “un­der a heavy cog­ni­tive load” as he tries to stick to his sto­ry while mon­i­toring your re­ac­tion.

  • Ask open-end­ed ques­tions to get them to pro­vide as many de­tails and as much com­plete in­forma­t­ion as pos­si­ble (“Can you tell me more about...?” “Tell me ex­act­ly...”). First ask gen­er­al ques­tions, and only then get more spe­cif­ic.

  • Don’t in­ter­rupt, let them talk and use si­lent pauses to en­cour­age them to talk.

If some­one in an air­port or oth­er pub­lic space is be­hav­ing sus­pi­ciously and when ap­proached ex­hibits a ma­jor­ity of the more re­li­a­ble red flags, Geisel­man rec­om­mends pulling him or her aside for more ques­tion­ing. If there are only one or two red flags, he would probably let them go.

Geisel­man tested tech­niques for tell­ing the truth from de­cep­tion with hun­dreds of un­ivers­ity stu­dents, and the stud­ies he and his co-au­thors an­a­lyzed in­volved thou­sands of peo­ple.

De­tect­ing de­cep­tion is hard, Geisel­man said, but train­ing pro­grams can be ef­fec­tive. Pro­grams must be ex­ten­sive, with an educa­t­ion phase fol­lowed by nu­mer­ous vi­deo ex­am­ples, and a phase in which those be­ing trained judge vi­deo clips and sim­u­late real-world in­ter­view­ing. Train­ing should be con­ducted on mul­ti­ple days over a pe­ri­od of a week or two.

“Peo­ple can learn to per­form bet­ter at de­tect­ing de­cep­tion,” Geisel­man said. “How­ever, po­lice de­part­ments usu­ally do not pro­vide more than a day of train­ing for their de­tec­tives, if that, and the avail­a­ble re­search shows that you can’t im­prove much in just a day.”

When Geisel­man con­ducted train­ing with Ma­rine in­tel­li­gence of­fi­cers, he found that they were im­pres­sively ac­cu­rate in de­tect­ing de­cep­tion even be­fore the train­ing be­gan. In con­trast, the av­er­age col­lege stu­dent is only 53 per­cent ac­cu­rate with­out train­ing, and with ab­bre­vi­at­ed train­ing, “we of­ten make them worse,” he said.

“With­out train­ing, many peo­ple think they can de­tect de­cep­tion, but their per­cep­tions are un­re­lat­ed to their ac­tu­al abil­ity. Quick, in­ad­e­quate train­ing ses­sions lead peo­ple to over-analyze and to do worse than if they go with their gut re­ac­tions.”

Geisel­man is de­vel­op­ing a train­ing pro­gram that he hopes will ef­fec­tively com­press the learn­ing curve and thus will serve to rep­li­cate years of ex­pe­ri­ence.

In the next year, Geisel­man plans to teach po­lice de­tec­tives tech­niques for in­ves­ti­ga­tive in­ter­view­ing and spot­ting de­cep­tion through the U.S. De­part­ment of Home­land Se­cur­ity’s Ru­ral Polic­ing In­sti­tute for un­der­served po­lice de­part­ments. He said this will be a per­fect fit for him be­cause he comes from Cul­ver, Ind., a small town that has few­er res­i­dents than UCLA has psy­chol­o­gy ma­jors.

A course Geisel­man taught on in­ves­ti­ga­tive in­ter­view­ing was used to in­ter­dict some in­sur­gent ac­ti­vity in the current Iraq war, he said, adding that mili­tary sources later told him the techniques may have saved many lives.

Geisel­man al­so has worked with the Los An­ge­les Coun­ty Sher­if­f’s De­part­ment on ef­fec­tive tech­niques for in­ter­view­ing chil­dren who may have been mo­lested and has in­ter­viewed crime vic­tims for po­lice de­part­ments around the coun­try in mur­der cases gone cold. His re­search has been funded by the U.S. De­part­ment of Jus­tice and the U.S. De­part­ment of Home­land Se­cur­ity.


* * *

Send us a comment on this story, or send it to a friend









 

Sign up for
e-newsletter
   
 
subscribe
 
cancel

On Home Page         

LATEST

  • St­ar found to have lit­tle plan­ets over twice as old as our own

  • “Kind­ness curricu­lum” may bo­ost suc­cess in pre­schoolers

EXCLUSIVES

  • Smart­er mice with a “hum­anized” gene?

  • Was black­mail essen­tial for marr­iage to evolve?

  • Plu­to has even cold­er “twin” of sim­ilar size, studies find

  • Could simple an­ger have taught people to coop­erate?

MORE NEWS

  • F­rog said to de­scribe its home through song

  • Even r­ats will lend a help­ing paw: study

  • D­rug may undo aging-assoc­iated brain changes in ani­mals

When someone is acting suspiciously at an airport, subway station or other public space, how can authorities find out whether he’s up to no good? A group of researchers has been analyzing years of past studies—including their own—on deception detection, and this week are reporting what they describe as some of the most up-to-date guidelines on catching liars in the act without polygraph or other physical tests. University of California Los Angeles psychologist R. Edward Geiselman, who has taught investigative interviewing techniques to law enforcers for years, and colleagues report the findings in the April issue of the American Journal of Forensic Psychiatry. The more reliable signs of deceit, Geiselman said, include: Deceptive people generally want to say as little as possible under questioning. Geiselman initially thought they would tell an elaborate story, but discovered, he said, that the vast majority give only the bare-bones. Studies with college students, as well as prisoners, show this, he added. Geiselman’s investigative interviewing techniques are designed to get people to talk. Although deceptive people don’t say much, they tend to spontaneously give a justification for what little they are saying. They tend to repeat questions before answering them, perhaps to give themselves time to concoct an answer. They often monitor the listener’s reaction to what they are saying. “They try to read you to see if you are buying their story,” Geiselman said. They often initially slow down their speech because they have to create their story and monitor your reaction, and when they have it straight “will spew it out faster,” Geiselman said. Truthful people are not bothered if they speak slowly, but deceptive people often think slowing their speech down may look suspicious. “Truthful people will not dramatically alter their speech rate within a single sentence,” he said. They tend to use sentence fragments more frequently than truthful people; often, they will start an answer, back up and not complete the sentence. They’re more likely to press their lips when asked a sensitive question and are more likely to play with their hair or engage in other “grooming” behaviors. Gesturing toward one’s self with the hands tends to be a sign of deception; gesturing outwardly is not. Truthful people, if challenged about details, will often deny that they are lying and explain even more, while deceptive people generally will not provide more specifics. When asked a hard question, truthful people will often look away because the question requires concentration, while dishonest people will look away only briefly, if at all, unless it is a question that should require intense concentration. If dishonest people try to mask these normal reactions to lying, they would be even more obvious, Geiselman said. Among the techniques he teaches to enable detectives to tell the truth from lies are: Have people tell their story backwards, starting at the end and systematically working their way back. Instruct them to be as complete and detailed as they can. This technique, part of a “cognitive interview” Geiselman co-developed with Ronald Fisher, a former UCLA psychologist now at Florida International University, “increases the cognitive load to push them over the edge.” A deceptive person, even a “professional liar,” is “under a heavy cognitive load” as he tries to stick to his story while monitoring your reaction. Ask open-ended questions to get them to provide as many details and as much complete information as possible (“Can you tell me more about...?” “Tell me exactly...”). First ask general questions, and only then get more specific. Don’t interrupt, let them talk and use silent pauses to encourage them to talk. If someone in an airport or other public space is behaving suspiciously and when approached exhibits a majority of the more reliable red flags, Geiselman recommends pulling him or her aside for more questioning. If there are only one or two red flags, he would probably let them go. Geiselman tested techniques for telling the truth from deception with hundreds of university students, and the studies he and his co-authors analyzed involved thousands of people. Detecting deception is hard, Geiselman said, but training programs can be effective. Programs must be extensive, with an education phase followed by numerous video examples, and a phase in which those being trained judge video clips and simulate real-world interviewing. Training should be conducted on multiple days over a period of a week or two. “People can learn to perform better at detecting deception,” Geiselman said. “However, police departments usually do not provide more than a day of training for their detectives, if that, and the available research shows that you can’t improve much in just a day.” When Geiselman conducted training with Marine intelligence officers, he found that they were impressively accurate in detecting deception even before the training began. In contrast, the average college student is only 53 percent accurate without training, and with abbreviated training, “we often make them worse,” he said. “Without training, many people think they can detect deception, but their perceptions are unrelated to their actual ability. Quick, inadequate training sessions lead people to over-analyze and to do worse than if they go with their gut reactions.” Geiselman is developing a training program that he hopes will effectively compress the learning curve and thus will serve to replicate years of experience. The cognitive interview that Geiselman and Fisher developed works well with both criminal suspects and eyewitnesses of crimes. Geiselman thinks these techniques are likely to work in non-crime settings as well, but said additional research should be done in this area. In the next year, Geiselman plans to teach police detectives techniques for investigative interviewing and spotting deception through the U.S. Department of Homeland Security’s Rural Policing Institute for underserved police departments. He said this will be a perfect fit for him because he comes from Culver, Ind., a small town that has fewer residents than UCLA has psychology majors. Later this month, Geiselman will travel to Hong Kong to provide training in investigative interviewing to the Independent Commission Against Corruption. An instructional course Geiselman taught on investigative interviewing before the second Iraq war resulted in cognitive interviewing techniques that were used to interdict some insurgent activity in Iraq, perhaps saving many lives, he was later informed. Geiselman also has worked with the Los Angeles County Sheriff’s Department on effective techniques for interviewing children who may have been molested and has interviewed crime victims for police departments around the country in murder cases gone cold. His research has been funded by the U.S. Department of Justice and the U.S. Department of Homeland Security.