The Psychology of Person Identification

Brian Clifford, Ray Bull

• Publisher: Routledge
• Year: 2017
• ISBN:
  9781315533513
• Edition: 1
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

Originally published in 1978, the laws and procedures governing person identification parades, photofit pictures and the forms of questions asked to obtain a description, had been increasingly called into question. The problem had been highlighted by several well publicised court cases, and considered by the Devlin Committee. This book reviews the status of psychological knowledge at the time concerning the many aspects of person identification and scientifically evaluates the methods and procedures used.

Contrary to the popular belief that identification is a simple affair, the authors use the theory and method of psychology to reveal the sources of the difficulties involved in recognising a once-seen person. Estimates of just how good a witness can be are drawn from laboratory studies using face photographs, from mock crime incidents, and from actual criminal cases, for reliability varies markedly in each of these three situations. Both an individual and a social perspective is taken of the eye-witnesses, and research into perception and memory, together with individual differences in such things as cognitive style, personality, suggestibility, age, sex, and ability to form both eidetic and memory images, are examined. The social aspects of stereotypes, the presence of other witnesses and the desire to be a 'good witness' are all discussed at length. Finally an extended examination of the possibility of voice parades and changes in identification procedures, together with man-machine interaction techniques, is undertaken.

1 Introduction

One of the main reasons for this book being written was the desire on the part of many organizations, groups and individuals to be cognizant of the factors which influence our identification and recognition of other people. In May 1974 in response to general disquiet the Home Secretary appointed Lord Devlin to chair a committee to examine the law and procedures on identification. The former chief of the Metropolitan Police, Sir Robert Mark, is quoted as saying that, 'If the police, the lawyers and the courts did their jobs properly the most likely cause of wrongful conviction is mistaken identity', and Lord Gardiner that, 'Most wrong convictions were on the matter of identification' (Cole and Pringle, 1974). If such statements are valid then not only may an innocent person be found guilty, but the true culprit could still be at large. However, to decide whether these views have any validity is a very difficult problem indeed. There are, of course, one or two cases in which incorrect recognition has played a part, but to determine how widespread such a phenomenon is seems almost impossible. Nevertheless, some people claim that, 'There is certainly a persuasive argument that the human memory for faces is so fallible that evidence of identification is virtually worthless. It seems likely that most people are quite incapable of remembering a face' (Cole and Pringle, 1974). Upon examination it appears that these views are based on what amounts to anecdotal evidence. Now anecdotal evidence may eventually be proved correct, but until it is, it cannot serve to inform the debates on either identification evidence or on potential changes in judicial procedure. We might ask then to what extent methodologically rigorous studies have investigated our powers of person identification. This book is an attempt to present the psychological information that is presently available concerning this question.

Marshall (1969) believes that, 'The life of our courts, the trial process, is based upon the fiction that witnesses see and hear correctly and so testify; and if they do not testify with accuracy on direct examination, cross-examination will straighten them out.' This (latter) 'is another fiction.' The credibility of an eyewitness is a matter of fact for the jury to determine, but in the past, and to a very large extent still today, jurors and potential jurors do not regard eyewitness identification with scepticism. The concept of identification as it is used by the legal profession seems to imply that perception and memory, as it operates in person identification, is a simple single process, and thus if we can assume suitable conditions of viewing and no mental abnormality then failure to recall or recognize is simply a case of wilfulness or lack of probity. This simplistic view of person identification is one we will be very concerned to deny. Person identification in a criminal situation will be argued to be at least a three-stage process. The first stage involves witnessing the incident and thus involves perception and all its vagaries. The second stage involves either recounting the person's appearance to the police, trying to construct a photo-fit or attempting to aid a police artist in creating a 'likeness' of the seen criminal and all these procedures involve memory and its inherent fallibilities. The third stage sees memory and perception operating in tandem when the witness either tries to locate the criminal in a photo album of criminals, or when he tries to match current perceptions of line-up paraders to his or her stored image of the seen criminal. The important point to stress is that error can creep in at any and all of these three stages.

As such then we take for granted, and will not be concerned to laboriously document, Kubie's (1959) assertion that people quite often do not see or hear things which are presented clearly to their senses, but do 'see' and 'hear' things which are not there, and further that people frequently do not remember things which happened to them but do remember things which did not happen to them.

While we accept the fact of fallibility in eyewitness testimony we will try to explain why this phenomenon occurs, and how remediation may be inaugurated. The reason why we can accept the facts of fallibility so easily is that they have been replicated many times in the experimental literature. The psychology of testimony has a long history and in fact was one of the first areas of applied psychology, although at that time it tended to be descriptive and programmatic rather than explanatory.

The most important source of early literature on forensic and general psychology was the annual reviews provided by Whipple between 1910 and 1917. The opening shot was fired by Freud in 1906 when he argued that psychologists should be used to ascertain the truth or otherwise of a testimony. This was taken up by Munsterberg in 1908 in the USA in his book On the Witness Stand where he stressed the need for legal reform based upon scientific experimentation. This call was reiterated by a German criminologist Gross (1911) and together they served to spawn a great deal of appropriate work which drove home the point that sense data were fallible, recall was idiosyncratic and eyewitness testimony was inherently unreliable. The academic psychologists were also very much involved in this forensically stimulated research area (this is best seen in the work of Binet (1897) and Stern (1903-38)). Binet investigated visual memory for pictures and this work was of at least indirect relevance to identification evidence. He showed clearly that observer factors contribute more than stimulus factors. Stern built upon this visual work but went beyond it by evolving the 'event', 'reality' or 'simulated crime' experiment. 'His experiment, demonstrating unreliability in recall for unexpected events, has become the paradigm for subsequent experimental studies of eyewitness reports by both psychologists and lawyers' (Levine and Tapp, 1973). The 'event' experiment was re-employed by Kobler (1915) and by Marston (1924), and has been developed by Buckhout (e.g. 1974b) and Loftus (e.g. 1975) in America, and by the Psychology Departments of the North East London Polytechnic and Nottingham University in England.

The fruitful reciprocity between law and psychology is best seen in the work of Hutchins and Slesinger (1928), the first a lawyer and the second a psychologist. They reviewed legal rules of evidence in terms of psychological findings on perception, memory and emotion, ultimately arguing for a still greater effort in interdisciplinary experimentation and application. This clarion call was echoed by Gardner, a lawyer, in 1933, who himself outlined positive lines of advance for future research. Two other treatises on legal psychology ought to be mentioned; McCarty (1929) and Burtt (1931) systematically related the role of such psychological parameters as stress, emotion and attention to perception and memory as they operated in a legal setting and verified experimentally the unreliability of identification. The later books by Wall (1965) and Marshall (1966) were designed' for the legal profession and unfortunately were somewhat light on sustained experimental evidence. Marshall succeeds in showing, by his own research using a staged incident, that social classes differ in the content of their recalls, that police are somewhat different from civilians in the amount and type of evidence they offered as testimony, and that we must be aware of the selective processes of perception and memory and the 'inventive reconstructions' of witnesses, together with the internal gambling procedures of witnesses at line-ups.

Hopefully, this potted history of classical studies will serve to show both an intimate interaction between forensic and general psychology, the possibility of extrapolation from pure to applied research in memory and perception and the application and elucidation of insights already existing in psychology to the area of identification evidence. This book is intended to build on these earlier studies and to try and explain why these documented fallibilities in identification occur.

Devlin (1976) called for exploration, of a research nature, into 'establishing ways in which the insights of psychology could be brought to bear on the conduct of identification parades and the practice of the courts in all matters relating to evidence of identification' (p.73). This has been a fundamental orientation of the present book, and is reflected in most of the chapters but especially the first two, where the insights we look at are those gleaned from research into cognitive processes and research of a more psycho-social nature.

Following Neisser (1967), cognitive psychology can be characterized as those branches of psychological research which deal with complex perception, the organization of memory, language, thinking and, most generally, man's intellectual processes.

Like all science, psychology has employed a number of general and specific simplifying assumptions and procedures. One such general simplifying procedure was to treat perception, memory, thinking, problem-solving and language as separate and separable processes. The reasons for breaking down, or of dividing up, the human being have been couched in terms of 'heuristic usefulness', 'experimental expediency' or simply 'manageability'. However, such devices have an unsavoury habit of becoming first conceptual distinctions and then ontological realities. Thus in the end one can glibly talk about 'laws of memory' which are unqualified by such factors as the role of language, thinking, valuing, believing or feeling.

Over and above the incapacitating outcome of the above general simplifying assumptions other more specific ways of conducting research arose which prevented the possibility of the direct application of experimental findings to the judicial process. These more specific simplifying procedures were concerned with the choice of theoretical models with which to interpret the data, and in the conception of man from which these interpreting theories were derived.

The theoretical models of man with which experimental psychology operates are noteworthy both for their dissimilarity one from the other and for their common dissimilarity with the reality which exists when a person sees a criminal episode and tries to remember it (and the chief criminal figures) for later reconstruction for the police. The models which were tried and failed were man as a black box, man as a telephone switch-board and man as a computer. All these failed for a greater or lesser number of reasons, but all were inadequate uniformly because they failed to take into account man's thinking, feeling, believing totality.

This last point leads into the single most damaging feature of psychology's intellectual history. Our conception of man's nature has hampered development in both pure and applied research. Behaviourism was the heir to associationism and in turn left a legacy to memory research which stipulated that the memorizer was essentially a passive recipient rather than an active perceiver, conceiver and memorizer. He was regarded as passive in the sense of contributing nothing to the input in terms of pre-existing knowledge, perhaps because he was a black box, and he was seen as contributing nothing to the encoding, storage, and rearrangement of memories, perhaps because perception was judged to be like a camera and memory like a tape recorder or videotape.

To a large extent the influence of these theoretical biases in psychology's history can be seen in the methodological decisions it made. Ebbinghaus (1885) can be viewed as the founding father of experimental memory research, and he wished to study the 'pure processes of memory' uncontaminated by everyday habits of thought and experience. As a result of this wish, and influenced by the pre-theoretical assumptions described above, the vast literature in early memory research was composed almost entirely of memory for nonsense material, especially the consonant-vowel-consonant trigram (CVC). The power of implicit assumptions about man and the progress of science can be seen clearly here. While Ebbinghaus was aware of the possible interrelationship between cognitive sub-processes he none the less fell foul of the general simplifying assumption that they could be separated. His use of nonsense syllables was the monumental error which misguided future psychology for nearly three quarters of a century, and indirectly rendered law enforcement agencies poverty-stricken in terms of a scientific basis for the execution of their duties. In terms of person identification Ebbinghaus, and especially his tradition, was doubly crippling because it used material devoid of meaning, and he used only verbal material. It can legitimately be argued that Ebbinghaus could not do everything at once, and this would be conceded, but it cannot be accepted that hundreds upon hundreds of psychologists should have slavishly followed his methodology and used his material to the exclusion of all else, never asking themselves questions about generality and applicability. Jenkins (1974) captures the tenor of this argument when he says that Ebbinghaus committed a particularly disastrous error, and goes on to say that his method, his material and his own orientation combine to hide the otherwise overwhelming effect of knowledge which is the most powerful variable in memorization. If one studies any effect after first removing the chief source of variance one will get the most odd results. We will see this clearly in terms of meaning, and factors over and above meaning, when we compare identification of people in the laboratory with identification of people in more real-life situations, where it will be shown that performance drops markedly in the latter.

Fortunately for an applied psychology of memory the edifice of theoretical error and the mist of misbegotten facts are now beginning to be demolished and dispersed, and we can but hope that the 'cultural lag' between the practical help we can offer those who have to deal with psychological phenomena in practice and the institutions of psychological research will not be too protracted. It will be some time but the future looks decidedly brighter. The theoretical tide began to turn about the 1960s, although it was charted as early as 1932 with the, now realized, futuristic work of Bartlett. The watershed was the publication in 1967 of a book called Cognitive Psychology by Ulric Neisser. The cognitive approach to human functioning refers to the study of all the processes by which the sensory input is transformed, reduced, elaborated, stored, recovered and used. As such it recognizes that perception, language, thinking and problem-solving are intimately interconnected, and must be given full consideration in any one field of interest, such as person identification. All these processes must be considered in a full and adequate explanation of eyewitness recognition.

Thus, at a stroke cognitive psychology overcame the first general simplifying procedure of decoupling the memory system from all the other subsystems. It also stressed the active subject and the fallacy of a black box approach. Most cognitive psychologists pay at least lip service to the fact of emotion and man's connotative being, but very little research has manipulated these as possible important considerations. This, however, is basically a residual problem based on, and largely a function of, the fact that most research has been carried out in the laboratory. It will only be a matter of time before such factors are being routinely explored. We will see from experiments which use more real-life situations, from our own studies at the North East London Polytechnic, and from court cases, that many cognitive phenomena associated with person identification are incomprehensible unless one takes into account what the person is, what he is trying to do, and the way his beliefs, values and motivations act not only at the time of perception but also during the period of storage and especially at the time of recall or recognition.

The second set of insights which we explore in terms of person identification are those found within social psychology. While we are concerned to stress that perception and memory are selective and interpretative we also want to go beyond this by postulating that these systems are at base social systems. The mechanisms of perception and memory are indistinguishable and inseparable from man as a social being. Person identification is customarily carried out in a social context, thus an adequate analysis of the topic requires that the eyewitness be viewed from a social-psychological perspective. An understanding of perception, memory and decision-making in identification situations involves considering such things as stereotyping, prejudice, implicit personality theorizing, and conformity to others, together with an appreciation of personality needs, social values, expectancies and evaluative apprehensiveness. Some of these 'insights' are discussed at length in this book, others more tangentially.

A few words should perhaps be said about the type of literature we will be discussing and analysing. Throughout the book we will draw on evidence which is both directly and indirectly relevant to the specific question of person identification. This we believe to be legitimate because while stimulus material may differ (for example, words can be presented auditorially or visually, objects can be presented themselves or as pictures, and faces can be presented either 'live' or in the form of photographs), the basic processes of memory are the same, at least qualitatively and should at the very least offer speculative predictions as to what would happen in real-life situations of criminal episodes, and later recognition of the criminal. We will thus draw liberally upon 'laboratory' studies which look at memory for faces presented as photographs and pointedly compare estimates of man's ability to recognize a seen face from these studies with estimates of the same ability obtained under 'mock crime' situations where actual people are seen perpetrating a crime and are then re-presented for identification. We will quickly see that these two estimates differ markedty. Thus, throughout we will be concerned to indicate, describe and draw inferences from extant research which we believe is relevant and transferable to, but not directly derived from, actual real-life person recognition or identification. This may limit the validity of our conclusions and render our judgments susceptible to refinement and modification in the light of future, more applied, research undertakings. A question which exercised us greatly was whether we were writing this book too soon but we are of the opinion that the current status of psychological theory and factual knowledge is sufficiently developed to warrant this volume. Not only are rich insights and relevant researches now available but some guidelines for future research are badly needed. Over and above the state of psychological research there is also the need of the judiciary to have at their fingertips an up-to-date and comprehensive review of current findings in person identification if the letter and spirit of the Devlin Report is to be executed. Chapter 2 thus looks at the new approach to memory and perception, and then goes on to argue that while apparently quantitatively superior, visual memory is at base no different from verbal memory and both are concerned with extracting meaning from the various inputs. We next compare visual memory for facial and non-facial material and argue that, once again, we need postulate no separate processes for the two systems. As a precursor to Chapter 4 we summarize the evidence which shows excellent memory of faces presented as photographs, but immediately counsel caution in using these estimates to argue that person identification is not a problem by presenting evidence from 'simulated crime' studies which give very much lower estimates of man's ability to remember and recognize a seen face. These latter studies strongly suggest that unexpectedness, brevity of exposure and stress are three of the many factors which are necessary considerations in understanding and evaluating eyewitness testimony, over and above the inherent limitations of man's perceptual and memorial ability.

Chapter 3 examines the possibility that witnesses may have stereotypic expectancies and notions about the kinds of physical appearance they expect criminals to have. It is argued that such stereotypes may play a role in person identification especially when confabulation occurs because memory is inefficient. It will be shown that contrary to the belief of many pundits a member of the general public may well have powerful expectancies about the appearance of criminals. If you glimpsed a mugger briefly would you on an identification parade be inclined to pick out someone who was young and scruffy or someone more mature dressed in a smart suit?

Chapter 4 looks at the chief source of identification - identification by means of face. In this chapter we look at recognition of seen faces (from photographic presentation and real-life crime episodes) and at recall, as measured by photo-fit construction. In so doing we outline what is currently known about facial scanning patterns, what features are most salient for later recognition, whether single features, features in combination or attributes are most important for identification, and then relate these to the actual practice of photo-fit construction as expressed by a recent Home Office survey. Once again we compare real-life recognition with recognition levels found in the laboratory and indicate the differences, and hopefully, their causes. We conclude the chapter by looking at man-machine recognition devices.

Chapter 5 looks at person identification by means other than by face, especially voice. A crime seldom takes place in silence. Usually the criminal issues instructions or demands some sort of obedience. Just how good memory is for a person's voice, how much of a speech sample we must hear, whether memory for a voice shows the same decrease over t...