been shown experimentally that witnesses who
receive an appearance-change instruction do make
more lineup identifications, this did not result in an
increased number of correctidentifications. Instead,
the appearance-change instruction was shown to
increase the number of incorrect identifications of
fillers (i.e., lineup members who are known to be
innocent) without increasing the number of correct
identifications of the criminal. Although it is uncertain
whether these findings will be replicated by future
studies, they do nonetheless challenge the basic
assumption underlying the use of the appearance-
change instruction. Such an increase in false identifi-
cations without a concomitant increase in correct
identifications means that lineup identifications made
following an appearance-change instruction were, as a
whole, less accurate than identifications made without
an appearance-change instruction. Additionally, the
appearance-change instruction has been shown to
increase the length of time it takes witnesses to make
an identification and to decrease the confidence with
which witnesses report making an identification.
Although it is as yet not known why the appear-
ance-change instruction increased false identifications
without a concomitant increase in correct identifica-
tions, two hypotheses have been advanced. Both are
predicated on the assumption that a lineup identifica-
tion occurs when the similarity of a lineup member to
the witness’s memory of the criminal surpasses a min-
imum level.
The first hypothesis is that the instruction may sim-
ply lower a witness’s decision criterion (the minimum
level of similarity needed to result in an identifica-
tion). Witnesses who are given an appearance-change
instruction might conclude that due to possible appear-
ance change they should not expect a high degree of
similarity between the criminal in the lineup and their
memory of the criminal. However, because even inno-
cent lineup members may bear some moderate resem-
blance to the criminal, if a witness’s decision criterion
is low enough, even these innocent people may be
falsely identified.
The second hypothesis that explains the effects of
the appearance-change instruction is that the instruc-
tion may lead witnesses to mentally alter various
facial features of the lineup members. For example,
witnesses may imagine what the lineup members
would look like with different facial hair, different
hairstyles, or a chubbier face. If witnesses mentally
alter the various lineup members’ appearance in an
effort to match the lineup members to their memory ofthe criminal, then even innocent lineup members may
come to resemble the actual criminal. Thus, the
appearance-change instruction would make it even
more likely that the similarity between an innocent
lineup member and the criminal surpasses the wit-
ness’s decision criterion, thereby leading to a false
identification.
Whether the effects described here are replicable
and whether they generalize across variations in the
wording of the appearance-change instruction, across
different witnessed events, and across various other
lineup manipulations remain open empirical ques-
tions. A greater understanding of the effects of the
appearance-change instruction, and the explanation of
those effects, awaits further research.Steve D. CharmanSee also Estimator and System Variables in Eyewitness
Identification; Eyewitness Identification: Effect of
Disguises and Appearance Changes; Eyewitness Memory;
Instructions to the Witness; WITNESS ModelFurther Readings
Charman, S. D., & Wells, G. L. (2007). Eyewitness lineups:
Is the appearance-change instruction a good idea? Law
and Human Behavior, 31,3–22.
Douglass, A. B., & Gerety, M. (2005). Eyewitness accuracy
as a function of lineup instruction and expectation.
Unpublished manuscript.
Technical Working Group for Eyewitness Evidence. (1999).
Eyewitness evidence: A guide for law enforcement.
Washington, DC: U.S. Department of Justice, Office of
Justice Programs.AUTOMATISM
Automatismis an excuse defense against criminal lia-
bility for defendants who committed a presumptively
criminal act in a state of unconsciousness, semicon-
sciousness, or unawareness. Medically, the term
automatism refers to motor behavior that is automatic,
undirected, and not consciously controlled. The use of
automatism as a legal defense is relatively rare and is
typically claimed in cases where the defendant’s con-
scious awareness is compromised by epilepsy, brain
injury, somnambulism (sleepwalking), or trauma. The
automatism defense is recognized as a viable defense
in U.S. and British courts, but the definitions and34 ———AutomatismA-Cutler (Encyc)-45463.qxd 11/18/2007 12:41 PM Page 34