Encyclopedia of Psychology and Law

misleading, suggesting details that are not part of the

narrative, and 5 are “true” questions, containing no

misleading information. These 5 true questions are

interspersed with the misleading or suggestive ques-

tions. Administration of the scales initially involves

presentation of the narrative to the interviewee; the test

administrator, or interviewer, reads out the narrative at

a steady pace. Following this, the interviewee is asked

to provide immediate free recall of the narrative. There

is then a 50-minute delay, followed by the intervie-

wee’s providing delayed recall of the narrative. The 20

questions about the narrative are then asked. When all

the questions have been answered, the interviewer

gives the interviewee negative feedback. Regardless of

level of accuracy, interviewees are told that they have

made some mistakes and that it will be necessary to

repeat the questions, and they are urged to try and be

more accurate. This negative feedback is to be deliv-

ered both clearly and firmly so as to convey an appro-

priate level of interrogative pressure to the interviewee.

Scoring

Immediate recall and delayed recall are scored accord-

ing to how many discrete pieces of information are

recalled correctly. Information is scored as correct if the

meaning is the same as the original item in the narra-

tive. Each correct item earns 1 point, with the maxi-

mum score being 40. There is also a score given for

Total Confabulation, which comprises a count of the

number of distortions and fabrications included when

recalling the narrative. A distortion represents a major

change to an existing piece of information from the nar-

rative, whereas a fabrication is the introduction of new

material. There are four suggestibility scores obtained

from the scales: Yield 1 is a measure of all leading

questions that are answered affirmatively in the first

round of questioning, with the range of possible scores

being 0 to 15; Yield 2 is the number of leading ques-

tions accepted following the administration of the neg-

ative feedback, and again the range of scores is 0 to 15;

Shift is a measure of any distinct changes in response to

all 20 questions in the second round of questioning,

with a range of 0 to 20; and Total Suggestibility is the

sum of Yield 1 and Shift, giving a range of 0 to 35.

Reliability and Validity

Factor analysis of the GSS 1 and GSS 2 questions shows

two clear factors, with items loading significantly on the

appropriate Yield or Shift factors. The scoring of Yield

and Shift are nondiscretionary in nature, and studies

assessing interscorer reliability confirm that it is very

high. Interscorer reliability for immediate and delayed

recall is also very high. Scoring of confabulation is

slightly less reliable, although correlations show that

this is still relatively high. Owing to the nature of the

scales, it is not possible to assess test-retest reliability

within each scale, as any memory of the narrative and

questions affects subsequent testing. However, compar-

ison of the scores of individuals who have completed

both the GSS 1 and the GSS 2 has shown high correla-

tions. The scales can therefore be said to have temporal

consistency.

Research

There has been extensive research using the scales to

test the hypotheses derived from the theoretical model

of interrogative suggestibility. The model postulates

that interrogative suggestibility is largely dependent

on individuals’ cognitive appraisal of the interrogative

situation. Research using the scales confirms that sug-

gestible responding is related to cognitive abilities.

For example, several studies have demonstrated that

GSS scores are negatively related to intelligence and

positively correlated with memory capacity. Studies

have also shown that increases in the perception of

psychological distance between the interviewer and

the interviewee are related to increases in scores on

the scales. Other research has demonstrated that there

are intra-individual differences, such as self-esteem

and self-monitoring, that also affect responses on the

scales. The research using the scales demonstrates that

interrogative suggestibility is a complex response

mediated by a range of cognitive and social psycho-

logical processes.

Stella A. Bain

See also Competency to Confess; False Confessions;

Forensic Assessment; Interrogation of Suspects; Postevent

Information and Eyewitness Memory; Test of Memory

Malingering (TOMM)

Further Readings

Bain, S. A., Baxter, J. S., & Ballantyne, K. (2007).

Self-monitoring style and levels of interrogative

suggestibility. Personality & Individual Differences,

42,623–630.

Gudjonsson Suggestibility Scales——— 341

G-Cutler (Encyc)-45463.qxd 11/18/2007 12:43 PM Page 341