R v Singh[1]  has attracted renewed attention to the protection that the right to silence is supposed to afford.  There can be little doubt that some sort of safeguard is necessary, in light of the many DNA-exonerated innocent defendants who confessed prior to their trials[2].  Over the last 10 years a substantial body of research[3] on the psychology of confession evidence has been produced.   Recent laboratory studies have revealed the disturbing finding that, depending on how they are interrogated, actual innocence may put innocent people at risk [4].  Research in the U. S. has shown that Miranda cautions are imperfectly understood in the first place, especially by young people or adults with cognitive impairments[5].  The standard caution in use in Canada fares no better[6].  Some innocent suspects waive their right to silence because they perceive their actual innocence to be protective and believe that their blamelessness will soon be self-evident[7].  Unrealistically, they anticipate that they will be able to explain to investigators the error of their ways.  Regrettably, the ensuing interrogation is at risk for eliciting a false confession from an innocent person, and (possibly) contributing to a false conviction.  In what follows, I spell out some of the psychological forces at play in this process.  

            While it is understandable that an innocent person would honestly believe that their candour will (ultimately) remove them from suspicion, naivety in the interrogation room is not restricted to the suspect.  Criminal Interrogations and Confessions (4^th ed.)[8], describes the ‘Reid technique’, one of the most influential and widely used police interrogation procedures in current usage. It is not my purpose here to provide an in depth analysis of the tactics outlined in this 600-page treatise.  Comprehensive critiques are available elsewhere[9].   While there is little doubt that the recommended devices are psychologically manipulative, if not coercive, this fact alone is hardly damning.  After all, the techniques are designed to extract confessions from lawbreakers.  There are however, two fundamental and inter-related assumptions that drive the interrogation process whose validity is sufficiently dubious so as to compromise the reliability of the entire enterprise.   

Mistaken Assumptions in the Interrogation Room

The first false assumption concerns the screening exercise that precedes the interrogation stage of the investigation.  It is assumed that innocent suspects can be distinguished from their guilty counterparts on the basis of an assortment of behavioral cues that are displayed during the initial interview.  Simply put, guilty suspects appear deceptive.  Their duplicity becomes apparent to the investigator who uses indicators from facial expressions, eye contact, posture, hand gestures, particular phrases, and various other “symptoms” of deceit.  Suspects who are guarded, who appear insincere, who are uncooperative, etc. are assumed to be guilty.  The problem with this “symptom analysis” approach is that it doesn’t work.  Innocent suspects may say and do things that are incorrectly attributed to their assumed culpability.  There is no compelling evidence that police (or anybody else) can detect deception with a high degree of accuracy[10] .  One study[11] presented videotaped denials of mock suspects who were truly guilty or innocent of a mock crime to college students who had been trained in the Reid technique.  They performed at chance when distinguishing liars from truth-tellers.  Those who were trained were less accurate than untrained controls.  Another study[12] engaged trained, experienced investigators in the same task.  They displayed chance-level accuracy, high confidence, and a response bias toward deception. The procedures that investigators use to (purportedly) sort out the liars from the truth-tellers have no empirical foundation.  The interviewers would be no less accurate if they simply flipped a coin: 

“To trust the information that these manuals provide in terms of cues to deception might result in misinterpretations of the verbal and nonverbal behaviours that a suspect shows.  Such misinterpretations might, in turn, fuel suspect-driven investigations that might ultimately result in miscarriages of justice.”[13]

     The training manual encourages investigators to trust their intuitions and to rely on a smorgasbord of behaviours, statements, and reactions that are not, in fact, reliable cues to deceit.  Because there is such a wide range of potential criteria to choose from, it is not surprising that the research reveals a mismatch between the accuracy of deceit judgements (little better than chance) and the confidence with which they are made (very high).    The outcome of the initial screening is a defining moment because if deceit is “detected”, the suspect is assumed guilty and the subsequent interrogation is designed to extract a confession.  The interrogation is not investigative in nature, rather it is guilt presumptive.  Its purpose is to elicit a confession from a suspect who is ‘known’ to be guilty.  Because the initial screening exercise is error prone, many innocent suspects will be subjected to highly intimidating interrogation techniques based on little more than a vague impression or a misinterpretation of the suspect’s demeanor. 

             The second false assumption underlying the interrogation process is the belief that innocent suspects are immune to the tactics of the Reid technique.  The manual is replete with unsupported pronouncements regarding what an innocent (compared to a guilty) suspect “would do”.  The reader is repeatedly assured that false confessions are highly improbable:  “…none of the steps is apt to make an innocent person confess” (p. 212); “... the prescribed efforts (…) would, in no way, be apt to cause an innocent person to confess”. (p. 365); “It is our clear position that merely introducing fictitious evidence during an interrogation would not cause an innocent person to confess.” (p. 429).[14]   

The Reid procedure consists of a structured nine-step process comprised of confrontation and minimization strategies. The former entails forceful accusations, the presentation of evidence (either real or manufactured), and interruptions whenever denials are attempted by the suspect.  Minimization involves the sympathetic presentation of moral justifications or explanations for the crime, often accompanied by the implication that a confession will result in leniency.  The guilt-presumptive nature of this exercise creates a slippery slope for innocent suspects because it may set in motion a sequence of reciprocal observations and reactions between the suspect and the interrogator that serve to confirm the interrogator’s belief in the suspect’s guilt.  Increased distress on the part of the suspect may be interpreted as resistance, thereby motivating the interrogator to redouble his or her efforts to extract a confession.

 From the suspect’s perspective, isolation, fatigue, and fear may produce a compliant (but false) confession from a person who merely wants to extricate himself from an aversive situation and/or who succumbs to implied threats of dire consequences or implicit promises of clemency.  The suspect is stressed, anxious, scared, confused, without social support, and presented with falsely constrained alternatives.  In one recent Ontario case, the cognitively impaired defendant (ultimately acquitted) in a child sexual assault case was asked during his 2-hour interrogation:  “Are you another Paul Bernardo or did you do this out of love?”  The suspect may make a regrettable but understandable decision about the perceived costs and benefits of confessing.  Are they confessing to something they did, or are they signing on to the spin that the interrogator has persuaded them is in their best interests to endorse?  Notwithstanding the authors’ assertions regarding the imperviousness of innocent suspects to these tactics, there is no psychological reason to assume that they would be immune to them, and the numerous DNA-exonerated defendants who had confessed shows that they were clearly susceptible to these practices.  

            Sometimes a suspect produces an ‘internalized false confession’.  The suspect actually comes to believe in his or her own culpability.  Illusory memories of criminal actions can be cultivated by the interrogator’s use of pseudo-technical explanations of how the crime could have occurred without the culprit’s conscious awareness.  After three long interrogation sessions during which incriminating (but false) evidence was presented to him, Michael Crowe actually conceded that he had killed his sister and somehow blocked the event from his memory[15].

            Mock jury studies demonstrate that confessions are very influential .  Even when mock jurors were made fully aware of the high-pressure tactics that produced nonvoluntary statements, confessions nevertheless boosted the conviction rate significantly[16].   How adept are the police at detecting false confessions?  Kassin[17] and his colleagues recruited prison inmates who were instructed to give: (1) a full confession to their own crime, and (2) a manufactured confession of a crime committed by another inmate.  College students and police investigators judged the truth or falsity of the videotaped confessions.  The overall accuracy rate did not exceed chance, but the police were more confident in their judgments. 

Tunnel Vision and the Attribution Error

These fundamental misconceptions contribute to the much decried “tunnel vision” that is frequently invoked to account for a wrongful conviction.  The problem however, is not a consequence of investigative overreach on the part of occasional zealots blinded by their passionate desire for a conviction.  The problem is more ubiquitous and pernicious.   Tunnel vision is systemic and inherent in the investigative process.  The psychological makeup or idiosyncratic motives of individual investigators are irrelevant if the very methods of investigation routinely subject innocent suspects to psychological pressures that are known to be powerfully influential.   In the scientific community it is recognized that “objectivity” is not a state of mind.  True impartiality requires more than just good intentions.  Objectivity is enshrined in the methods, design features, control groups, etc. that characterize scientific rigor.  Researchers appreciate that they need to protect themselves from their own biases.  In the domain of police interrogations, the necessary controls are not in place and strong biases remain unacknowledged.  ‘Often in error, never in doubt’ is a fair characterization.  Consequently, false confessions are inevitable, and they will continue to occur at a rate that makes a mockery of Blackstone’s ratio.   

             The last half century of research in social psychology has delivered a clear message that social circumstances (“the situation of the moment”, as it were) play an enormously powerful role in the regulation of people’s actions and decisions.  The widespread tendency to attribute causality to internal (dispositional) forces, even in the presence of obvious situational determinants, has been called the fundamental attribution error [18].  In 1951 Solomon Asch[19] demonstrated that when individuals were confronted with group consensus that defied common sense, they would publicly express agreement with the group.  Stanley Milgram[20] showed that ordinary people would willingly deliver what they believed to be lethal amounts of electric shock to an innocent stranger, if instructed to do so by someone in authority.  In the Stanford Prison study[21], college students assigned roles as guards in a simulated prison situation behaved aggressively (and sometimes sadistically) after just a few hours in the mock prison.  The recognition of the power of the situation prompted Hannah Arendt’s much quoted phrase “the banality of evil”[22].  Wrongdoing emerges less from internal malevolence and more from social policies or situational forces that encourage or condone wrong acts.

Modern jurisprudence has some catching up to do because the widespread individualism at the core of the law’s behavioral assumptions is at odds with contemporary psychological findings.  The criminal justice system “… remains mired in an antiquated model of behavior that consistently compromises the fairness of legal doctrines and undermines the ability of legal institutions to deliver justice to those most in need of it”.[23]   When appellate reviews search retrospectively for evidence of the defendant’s “overborne will” they would do better to scrutinize the procedures and assumptions that govern the psychological architecture of the interrogation room.

 Conclusion

            Modern interrogation practices are based on fundamentally mistaken assumptions:

(1) Investigators believe that they can accurately identify a deceitful suspect when the evidence shows unequivocally that they cannot.

(2) Investigators believe that their interrogation practices cannot elicit false confessions from innocent suspects, when there is every reason to believe that they can and do. 

            The right to silence is supposed to rectify the disadvantage that a detained suspect is faced with when confronted with the powers at the disposal of the state.  As I have noted elsewhere[24], the caution is clumsily worded, difficult to understand, and routinely ignored even when it is asserted.  This means that the social chemistry of the interrogation room is psychologically disconcerting from the outset.  The suspect is informed that he need not say anything.  If he opts to remain silent, the same agent who moments earlier informed him that he could remain silent proceeds to ask a litany of questions, and the questions persist, no matter how often the right to silence is invoked.  On balance, it does not appear to be providing much of a safeguard.  The Singh decision has done nothing to change this state of affairs, and given what we know about current interrogation practices, there is much that the suspect needs protection from.