Eyewitness Identification: Effect of Disguises

People who wear a disguise are attempting to conceal their appearance or change how they look. Culprits may wear any of a number of possible disguises for the commission of a crime. For example, a bank robber may wear a ski mask, or dark sunglasses and a knit cap. Changes in facial characteristics may result not only from a deliberate attempt to change one’s physical appearance while committing a crime but also because, with the passage of time, a culprit naturally ages and thus may look different from when the crime took place. Research has examined the influence of several disguises and appearance alterations such as hairstyle and facial hair changes, removal or addition of eyeglasses, and the wearing of a cap. Overall, disguise and changes in appearance make accurate recognition significantly more difficult. This decrease in recognition can be dramatic depending on the degree of change. The greater the change, the greater the decrease in accuracy for witnesses, both adults and children, trying to make an identification. The hair, hairline, and upper portion of the face, if obscured, are particularly ineffective for later accurate identification. Both the simultaneous and sequential lineup procedures have been tested in laboratory settings to determine their efficacy when a culprit’s appearance has changed (e.g., hairstyle, facial hair). For child and adult witnesses, both lineup procedures produced comparable and lowered accurate identification rates when an appearance change occurred compared with the case when there was no change.

Remembering Faces

How do we remember a face? Do we remember the features of a face or do we remember the whole face as a gestalt? Some debate has occurred over this issue, with a number of questions remaining unanswered. It may be that both types of encoding occur or that one strategy is more relied on depending on the developmental stage of the observer. For example, it has been suggested that adolescents and adults are more likely to use a gestalt or holistic approach to remembering faces, taking the whole face in, whereas younger children may be more likely to rely on a featural strategy, focusing on individual features.

Academic Writing, Editing, Proofreading, And Problem Solving Services

Get 10% OFF with 24START discount code

Change of Appearance: Facial Characteristics

Regardless of the process that we use to remember a face, it becomes much more difficult to do this when facial characteristics change. Moreover, a change in one feature may make the whole face appear different. Consider the case when someone changes hairstyles or hair color or if a male shaves off his beard or grows facial hair. Changes in any of these features make it more difficult to correctly recognize that person.

The influence of three facial changes on recognition/identification accuracy has been examined across a number of studies: changes in hairstyle, facial hair, and the addition or removal of glasses. To study the influence of these changes, often participants are presented with several photographs featuring different “targets.” Following some delay, participants are presented with another set of photographs, some of which are never-before-seen faces, some are of the targets as they appeared in the initial set of photographs, and others are of the targets but with some changes in appearance—for example, the target may not be wearing glasses in the first set of photos but could be wearing glasses in the new set. When an alteration or change is made, there is a significant decrease in accurate identification. Moreover, when changes to facial features are combined, the difficulty with identification can increase. Most often, changes to facial features results in an inability to correctly recognize the person seen previously.

The natural aging process can also make accurate identification more difficult. For example, a 2-year difference in time can reduce recognizability, in particular if there is a large discrepancy between the two appearances, such as when facial hair is grown. In one study examining the aging process, participants initially viewed photographs of high school students whom they would later have to recognize in a set of photos taken 2 years later. Participants had difficulty in correctly recognizing photos if there was a large discrepancy between the high school photo and the photo taken of the same person 2 years later.


Culprits may attempt to evade identification by wearing a variety of disguises that conceal either part or most of their face during the commission of the crime. They have been known to wear ski masks revealing just the eyes and mouth, hats and sunglasses, stockings over their head, and other sorts of masks. Studies that have examined the influence of disguise often have participants watch a videotaped mock crime in which half the participants see the culprit wearing a disguise, such as a knit cap that obscures the hair and hairline. The remaining participants see the culprit without the disguise. Research indicates that participants are almost twice as likely to provide an accurate identification of the culprit when there is no disguise than when a disguise is donned. Moreover, the facial composites produced by participants who see a culprit with a cap show much greater variability than composites from those participants who saw a culprit without a cap.

Researchers have attempted to determine which features are more essential for later recognition/ identification. The upper portions of the face, including the hair, hairline, and eyes, seem more critical for later accurate recognition than the lower portions. Hair in particular is a feature that many people focus on and later use as a cue for recognition. Unfortunately, hair changes are very easy to achieve and can be used to avoid eyewitness identification. Other fairly easy changes that can prove problematic for eyewitnesses include removing or adding eyeglasses and growing or shaving off of facial hair.

The degree to which a change in appearance or disguise is successful in evading later recognition/ identification is determined by the extent of the change. For example, framed eyeglasses will likely have a weaker effect than tinted sunglasses; removing a partial beard will likely be less concealing than removing a full beard. It has been suggested that eye-witnesses’ ability for accurate identification declines because of cue mismatch; that is, a witness’s memory trace of the target/culprit does not match the person they are currently examining. This incongruent memory trace may lead a witness to identify an innocent person, or they may not identify the culprit.


Although it is generally accepted that adults are capable of encoding faces holistically, less is known about children’s encoding abilities. Some researchers believe that children encode faces featurally until approximately age 10 and then switch to a more holistic encoding strategy. Most research on children’s facial recognition abilities has suggested that younger children are more likely to pay attention to (and thus encode) specific features of the face. In fact, some studies find that younger children, 6 to 7 years old, are better at identifying individual features of a face (especially the eyes) than older children, aged 9 to 10.

Certain types of changes may be more challenging for children than for adults, especially if younger children are relying more on a featural encoding strategy than a holistic one. When children below 10 years are providing descriptions of strangers seen for a brief time (e.g., 2 minutes), often only two or three descriptors are reported. (It is important to note that although children may have more descriptors to report, they may not have the language skills or verbal ability to describe the features to report them.) Descriptors provided by children often pertain to hair characteristics. If hair changed from the time of encoding to recognition, children may have difficulty in correctly identifying the stranger’s face.

Research that has examined the influence of change in appearance on children’s recognition abilities finds that children can be misled when paraphernalia is used. For example, a number of facial recognition studies that initially show children photos of targets wearing hats and glasses find that children are likely to misidentify others provided they are wearing the same paraphernalia worn by the targets. Moreover, if the target removes the paraphernalia, children are unlikely to identify the person as someone who was previously seen.

Identification Procedures

It has been recommended that lineup members, other than the suspect, be selected by matching the descriptors that eyewitnesses provide in their description of the culprit. For example, if the witness describes the culprit as having short, dark hair, medium build, and a fair complexion, all lineup members should fit this description. The exception to this recommendation occurs when the suspect does not match one or more descriptors provided by the witness. In such a case, the other lineup members should match the suspect on the particular features reported by the eyewitness. The remaining features in the eyewitness’s description should match all the lineup members. This strategy allows for some variation among lineup members but also tries to ensure that the suspect does not stand out. Having the suspect stand out may lead to wrongful identification.

If a mask or another disguise is used, it may be possible for the police to construct a lineup for the mask or disguise. Similar to a person lineup, a lineup for a mask or disguise would allow the witness to view the suspected item, such as sunglasses, along with other distractors (e.g., other pairs of sunglasses). Witnesses can attempt to identify the sunglasses worn by the culprit during the crime, for example. Alternatively, the suspect and other lineup members may be requested to wear the disguise or mask for the lineup identification.

The police may choose from a number of lineup procedures when conducting an identification. In the simultaneous lineup procedure, the witness looks at the lineup members all at once. In a sequential lineup, witnesses look at lineup members one at a time. With the latter procedure, witnesses are required to make an identification decision for each lineup member without being able to look at other members. More specifically, witnesses are not able to move forward or backward in the sequence.

Both the simultaneous and sequential lineup procedures have been tested in laboratory studies when a culprit has changed appearance following the commission of a crime. Overall, when a culprit changes appearance (i.e., change in hairstyle, removal of a beard), the correct identification rate (i.e., picking out the guilty person when that person is in the lineup) decreases significantly compared with when there is no change in the culprit’s appearance. This result has been found for adults as well as children.

Also, in laboratory studies, simultaneous and sequential lineup procedures were compared in terms of a witness’s ability to correctly reject a lineup (i.e., saying the guilty person is not present in a lineup that does not contain him or her) when the suspect did not match the culprit’s description. Correct rejection rates have been found not to differ significantly across these two lineup procedures when there is an appearance mismatch. The evidence to date does not support the use of a sequential lineup when a suspected change of appearance or disguise has been used. Although the simultaneous procedure may be recommended over the sequential, no “ideal” procedure for lineup identification can be touted when there is a suspected change in appearance or when a disguise has been used to commit the crime.


  1. Memon, A., & Gabbert, F. (2003). Unraveling the effects of sequential presentation in culprit-present lineups. Applied Cognitive Psychology, 17, 703-714.
  2. Pozzulo, J. D., & Balfour, J. (2006). The impact of change in appearance on children’s eyewitness identification accuracy: Comparing simultaneous and elimination lineup procedures. Legal and Criminological Psychology, 11, 25-34.
  3. Pozzulo, J. D., & Marciniak, S. (2006). Comparing identification procedures when the culprit has changed appearance. Psychology, Crime, and Law, 12, 429-438.

Return to the overview of Eyewitness Memory in Forensic Psychology.