GUIDELINES FOR THE CONDUCT AND INTERPRETATION OF EYEWITNESS IDENTIFICATION TESTS

• Jurisdiction: Australia

Eyewitness identification tests ...................................................................... [65.700]

Line-up size ................................................................................................... [65.710]

Selection of line-up members ....................................................................... [65.720]

Line-up administrator .................................................................................... [65.730]

Line-up instructions ....................................................................................... [65.740]

The number of identification attempts .......................................................... [65.745]

Presentation format: Live line-up versus photoarray .................................... [65.750]

Presentation format: Simultaneous versus sequential ................................. [65.760]

The record of the witness's identification response ..................................... [65.770]

Confidence and speed of the identification response .................................. [65.780]

Influential events following the identification test.......................................... [65.790]

Summary ....................................................................................................... [65.800]

[65.700] Eyewitness identification tests

In the final section of this chapter, we discuss processes related to eyewitnesses' involvement in suspect identification tests. An identification test may take one of several forms, such as a (live or video) line-up comprising the suspect and a number of other people, an array of photos including the suspect (which we will also refer to as a line-up in this section), or a showup in which just a single suspect is presented to the witness. The witness's behaviour in any of these tests can have a significant impact on the ensuing investigation and the outcome of any court proceedings, irrespective of whether the evidence is actually admitted at the time of the trial. For example, a positive identification is likely to focus the investigation on the suspect and increase the likelihood of a prosecution being initiated (Cutler and Penrod (1995); Semmler, Brewer and Douglass (2011)). A negative decision (ie, the witness rejects the line-up, which in field settings may happen in as many as 50 per cent of line ups: Pike, Brace and Kynan (2002)) may satisfy the police that their suspect is not the culprit, thereby changing the course of the investigation, or perhaps leading the police to conclude that the witness is unreliable.

It is important to emphasise one very important general consideration from the outset. If line-ups (live or photo) are used by law enforcement officers to test what is really nothing more than a hunch about the identity of the culprit, culprit-absent line-ups and, in turn, mistaken identifications of innocent suspects are likely outcomes. Consequently, researchers have emphasised that it is important not to use line-ups to test hunches about the culprit's identity (cf Wise, Safer and Maro (2011)). Rather, line-ups should only be conducted when the investigating officers have established reasonable grounds for believing that the suspect is the culprit (Wells, Yang and Smalarz (2015)).

The likelihood of mistaken line-up decisions will also be shaped by interactions between a range of other variables that we discuss in the following sections. Some of the factors reported in the earlier sections of this chapter that affect the outcome of the interview process (eg, the timing of the test) also apply to identification test outcomes. In the current section, we review those factors that are particular to the administration of an identification test. These include the composition of the line-up (the size of the line-up and the way in which line-up members are selected); the presentation of the line-up (the line-up administrator, the instructions provided to the witness, a single identification attempt) and the format of the line-up (live line-up versus photoarray, simultaneous versus sequential presentation); the recording and interpretation of the witness's identification response; characteristics of the identification response, such as confidence and speed of responding; and influential events following the identification test.

[65.710] Line-up size

There are two dimensions of line-up size: nominal and functional size. Nominal size refers to the actual number of line-up members, whereas functional size indicates the number of members who are plausible fillers for the suspect. Nominal line-up size varies across different jurisdictions around the world, but typically ranges from six to 10 members. Some evidence suggests that, as long as the line-up contains at least three plausible fillers as well as the suspect, nominal size has little impact on the likelihood of witnesses choosing or on rates of correct and false identifications (Nosworthy and Lindsay (1990)). However, larger line-ups obviously reduce the likelihood that a witness who guesses when confronted with a culprit-absent line-up will select the innocent suspect.

One special case that warrants mention is a nominal size of one, typically referred to as a showup. North American field studies report that between 30 per cent and 77 per cent of identification tests involve the presentation of a single suspect to the witness (Dysart and Lindsay (2007)). A review of studies contrasting the outcomes of showups and line-ups (Steblay, Dysart, Fulero and Lindsay (2003)) revealed that witnesses are less likely to make positive identifications from showups (ie, they are less likely to say, "Yes, that's him") than from line-ups and that both procedures produced similar rates of correct identification. However, while the likelihood of a witness choosing from a showup is lower than that for a line-up, false identifications of innocent suspects are more likely. This is because any positive identification of an innocent suspect from a showup constitutes a false identification, whereas positive identifications from unbiased culprit-absent line-ups tend to be spread across the suspect and (known to be innocent) fillers. Note that a special problem for showups is what has been referred to as clothing bias. If an innocent suspect in the showup is wearing some distinctive item of clothing that is similar to what the culprit wore, a mistaken identification of the suspect is more likely (Dysart and Lindsay (2007)).

Considerable research attention has been devoted to who should go in the line-up with the suspect. One issue is whether the line-up should contain just one suspect or multiple suspects. Given that witnesses often choose line-up fillers who are known to be innocent (Steblay, Dysart, Fulero and Lindsay (2001); Wells (1993); Wells, Memon and Penrod (2006); Wells and Turtle (1986)), multiple-suspect line-ups increase the likelihood that an innocent suspect will be identified and (possibly) prosecuted. In contrast, a single-suspect line-up allows a clear distinction to be drawn between identification of the suspect and a known-innocent filler.

A second issue is the functional size of the line-up. Functional size refers to the number of line-up members who represent plausible fillers (see Wells (1993) for a detailed discussion of functional size). Imagine that a witness, when describing the culprit, clearly recalled the culprit having bright blue eyes. If a suspect (with bright blue eyes) then appears in a line-up with a number of fillers, all of whom might look very similar to the suspect but do not have blue eyes, the functional size of the line-up is one. While low functional size will increase the chances of a correct identification when the suspect is the culprit, it also increases the likelihood of a false identification when the...