Encyclopedia of Psychology and Law

the similarity of faces on multiple dimensions using

Euclidean distances. Results derived from PCA have

been shown to relate to lineup identification perfor-

mance and to measures of lineup fairness.

Special Considerations

in Selecting Fillers

In employing the perpetrator-description-matched

strategy to select fillers, a number of issues may arise.

One difficulty is that the witness may provide an inad-

equate number of details regarding the perpetrator’s

appearance for selecting fillers for the lineup. Selecting

foils when there is little information regarding the cul-

prit’s appearance may result in a lineup in which the

members are highly dissimilar in relation to one

another and to the suspect. As a consequence, the per-

petrator-description-matched strategy might in some

cases increase the rate at which innocent suspects are

identified. Additionally, the degree to which the

lineup members are similar can affect the eyewitness’s

decision standard. If the foils in the lineup are rela-

tively low in their similarity to the culprit, then wit-

nesses may be less cautious in identifying a face compared

to when there is a higher degree of similarity across

lineup members. Finally, research comparing the

suspect-matched and perpetrator-description-matched

strategies has produced findings that are mixed,

thereby leading some researchers to maintain that it is

premature to recommend one strategy over the other

at this time.

Nevertheless, additional procedural safeguards

have been suggested for selecting fillers when using

the perpetrator-description-matched strategy. For

example, if the witness does not mention a character-

istic, such as facial hair, the police might assume the

perpetrator has the default value of the characteristic,

and assume that the perpetrator was clean shaven.

Though placing the clean-shaven suspect among foils

with facial hair would not violate the witness’s

description, doing so would draw attention toward the

suspect and perhaps bias the witness to choose the

suspect only because the suspect’s appearance is dif-

ferent from the others. Therefore, to avoid this possi-

bility, if the suspect is clean shaven, then he should be

placed among clean-shaven foils. Additionally, when

the suspect does not fit the description given by the

eyewitness on some level, it has been recommended

that investigators select foils based on their match to

the appearance of the suspect on the features in which

the description of the culprit does not match the sus-

pect. Along these same lines, if there are multiple wit-

nesses involved in a case, a recommendation that has

been made is to create a separate lineup for each eye-

witness. In laboratory studies in which researchers

feel that it is not feasible to create lineups for each

participant, the foils can be selected based on their

match to the typical or modal description given by

research participants viewing the suspect.

Finally, care should be taken to ensure that the

arrangement of the lineup is uniform. The pho-

tographs themselves should be similar and presented

to witnesses in a standardized fashion. For example, if

one photograph is slightly tilted away, or the focal

length is greater, or the facial expression differs from

the others, responses may be biased toward that pho-

tograph simply because it stands apart from the others

in its presentation. Additionally, uniformity across

members with respect to clothing should also be

achieved. Moreover, if the eyewitness describes the

perpetrator as wearing a particular type of clothing,

the suspect should not be the only one wearing that

type of clothing in the lineup.

Assessing Lineup Fairness

Researchers use the mock-witness procedure to deter-

mine lineup fairness, or the adequacy of lineup fillers.

In the mock-witness procedure, participants who have

not seen the perpetrator are given a description of the

perpetrator along with a lineup. They are asked to pick

the member of the lineup that most closely resembles

the perpetrator’s description. If the lineup is fair, then

mock witnesses should not select the suspect at a rate

significantly above chance.

Mock-witness choices can also be used to deter-

mine whether the fillers that have been selected for the

lineup bias choices toward or away from the suspect.

Bias measures include functional size and defendant

bias. Mock-witness choices are also used to determine

lineup size, which refers to the extent to which identi-

fication responses are distributed evenly across lineup

members. Measures designed to examine lineup size

include effective size, number of acceptable foils, and

Tredoux’s E.

Lineup bias and size can vary depending on how

the members of the lineup are arranged. For instance,

if the suspect is placed between two foils that are low

in similarity relative to the suspect, the suspect may

“pop out” and hence be chosen significantly more

456 ———Lineup Filler Selection

L-Cutler (Encyc)-45463.qxd 11/18/2007 12:43 PM Page 456