TEST CONSTRUCTIONS

INTRODUCTION

The central element of any Psychophysiological Detection of Deception (PDD)

examination is the examinee's physiological responses to a set of questions.
Questions are the fundamental tools of a PDD examiner. Improper construction
of questions may lead to confusion of the subject that would affect the result of the
polygraph test.

In this lesson, you will learn the guidelines in construction test questions in
polygraph examination. Moreover, proper formulation of the different test questions,
definitionS, and the difference between each type of questions will be tackled.
Understanding how to formulate these questions is essential for conducting accurate
and effective tests. Additionally, it covers the significance of sacrifice relevant questions,
symptomatic questions, and supplementary test questions in the examination process.

Learning outcomes:

At the end of this lesson, the students are expected to:

• Define relevant questions and explain their importance in lie detection

examinations.
• Construct relevant questions effectively, adhering to specific guidelines to
ensure clarity and reliability.
• Differentiate between primary and secondary relevant questions and
understand their respective roles in testing.
• Describe the purpose and construction of comparison questions, including
probable lie and directed lie questions.
• Understand the role of sacrifice relevant questions in the testing process
and how they contribute to accurate data collection.
• Explain the significance of symptomatic questions and their application in
lie detection examinations.
• Recognize the purpose of supplementary test questions, including peak of
tension and quilt complex tests, and understand their role in enhancing
examination accuracy.

I. RELEVANT QUESTIONS

a. Definition

A relevant question is a question asked during an examination that pertains

directly to the matter under investigation for which the examinee is being tested.

b. Constructing Relevant Questions

Relevant questions should:
• be written in a clear and concise style,
• not contain any unnecessary verbiage, as this tends to make the
questions more cumbersome and difficult to understand.
• constructed in a manner so their meaning is unmistakable.
• developed in a manner consistent with an examinee's intelligence
level. If the examinee does not understand a relevant question due
to complicated wording or an inadequate vocabulary cognitive
processing could occur possibly affecting the examination outcome.
• address only one issue, i.e., should not be compound questions.
•
The following are examples of poorly worded questions
adjacent are the correct constructions:
Did you steal any money or anything else from that store
yesterday? (Did you steal any of that money?)
Did you manipulate any part of that female's sex organs for
sexual gratification, while at her house? (Did you touch that
female's vagina?)
• The examiner should also avoid using legal jargon or other words
that could be perceived as emotion evoking. Words such as assault,
rape, sodomize, murder, molest, mutilate, butcher, and kill should not
be utilized in relevant questions, instead, should use an action verb
in the relevant question that describes how the crime was committed,
i.e., hit, strike, stab, shoot, cut, steal, etc.
• pose a dichotomy so they will have to answer "Yes" or "No. This helps
to avoid vague or extended answers and ensures that the
examinee's position on the issue is clear. If the examinee is allowed
to verbally reply in any other manner, this could generate artifacts or
other excessive noise in the physiological tracings, making test data
analysis difficult or impossible.
• not construct relevant questions that are worded in the form of an
accusation or appear to be judgmental. This might cause the
examinee to feel the examiner has already reached a conclusion
about his honesty or involvement in the crime. For obvious reasons,
the following types of relevant questions SHOULD NOT be asked
during the examination:
After you forged that check, did you cash it?
After John shot that bank teller, did he tell you where he hid
the weapon?
Did you use any of that cocaine you purchased from Smith?
• always use the phrase "any of" when the issue involves multiple
items or amounts of money. Because of the possibility of multiple
suspects sharing fruits of the crime, or money amounts being inflated
 to cover other possible criminal activity. This will help to prevent the
examinee from rationalizing his involvement in the crime because he
only stole/received a portion of the money. In this regard.
The following are examples of improper relevant questions
adjacent are the correct constructions:
Did you steal the $495.00 from that room? (Did you steal any
of that money?)
Did you burn those classified documents? (Did you burn any
of those classified documents?)
•
c. Prioritizing/Selecting Issues

There are many guidelines that influence how proper relevant questions are
selected and constructed. Two critical steps examiners must consider when
developing proper relevant test questions are target selection and appropriate
relevant question construction.
Target selection is an extremely critical element in developing relevant
questions. In complex investigative matters, the examiner must first determine the
number of issues or targets to be resolved. In specific issue testing, it is imperative
that the correct targets or issues be selected for testing. Additionally, in
investigations where there are multiple targets to be tested, it is critical that the
correct target is selected for the initial series. If necessary, additional targets or
issues can be resolved with additional testing.
During the target analysis selection phase, the examiner will generally
consider four elements in selecting the initial issue for testing. The four elements
are:
(1) determine the number of issues to be tested and then select the most
serious/intense issue;
(2) focus on the action that describes what the suspect did to commit the
act(s);
(3) focus on the probable motive of the suspect; and
(4) select the target that connects the suspect most closely to the crime.

For instance, a male examinee is suspected of stealing a female's purse

and then raping her. In terms of potential confinement for robbery and rape, the
more serious offense is rape. Therefore, the issue to be addressed in the primary
and secondary relevant questions for the initial series should focus on the rape. If
additional testing is necessary for the robbery issue, an additional series can be
conducted for that aspect of the investigation.
After the most serious offense has been determined in multiple issue
investigations, the examiner should then focus on the action that describes what
the suspect did to commit the offense. In performing this task, the examiner must
ensure that the most appropriate relevant questions have been developed for the
“case-at-hand.” Additionally, if the correct “action verb” is presented in the question,
it will help to prevent the suspect from rationalizing out of the act posed by that
relevant question. The following are examples of appropriate verbs that best
describes the act committed by the suspect:
Did you "smoke" any of that marihuana?
Did you "tear" that woman's blouse?
Did you "shoot" that man?
Did you “make” any of the writings on that check?
Did you "hit" that woman?
Did you “steal” any of that money?

d. General types of Relevant Questions

i. Primary Relevant Questions

A primary relevant question addresses the primary issue. The
purpose of the primary relevant question is to test for direct involvement,
i.e., the “did you” aspect of PDD testing. It uses an action verb to describe
the act that was committed. Unless dictated, sa primary relevant question
requires the examinee to answer “No”.
The following are examples of primary relevant questions:
Did you steal any of that money? No
Did you smoke any of that marihuana? No
Did you shoot that man? No
Did you stab that woman? No

When conducting examinations of informants, victims, or

witnesses the examiner may construct relevant questions that require
affirmative answers. In this type of examination, the issue becomes the
examinee’s truthfulness, not whether they committed a specific crime. The
examiner is simply attempting to verify information previously provided by
the examinee.

The following are examples of “Yes” answered relevant questions:

Did that man bring marijuana into the classroom? Yes
Did that man show you a substance he identified as
marijuana? Yes
Did that man offer to sell you marijuana? Yes

During the development of primary and secondary relevant

questions, the examiner should avoid asking questions that probe the
suspect's "perceptions" or "intent" while committing the crime. Sometimes
these conditions change over time or a suspect’s original intent may have
changed after the crime was initiated. This may allow a suspect who
committed the crime to rationalize out of the relevant question. For example,
 a suspect commits a burglary to steal property inside a residence. After
obtaining entry to the residence, he discovers a woman inside and decides
to rape her. His original intent was to steal property but changed to rape
after the crime was initiated.
The following are examples of questions that might probe an
examinee’s perception or intent and should be avoided:
Did you "purposely" touch that female's buttocks?
Did you "deliberately" set fire to that house?
Did you "intend" to hurt that child?

ii. Secondary relevant question

Test question that addresses a secondary element of the issue

addressed in the primary relevant question.
Typically, if the case facts allow, this category of relevant question
addresses four secondary elements of the primary issue:
(1) secondary involvement, such as help, plan, or participate;
Did you help anyone steal any of that money? No
(2) secondary element, such as seeing, hearing or knowing
who committed the primary act;
Did you see anyone steal any of that stereo
equipment? No
(3) knowledge of the nature or location of evidence; and
Do you know where any of those stolen weapons are
now? No
(4) physical acts that support the primary issue, such as
tearing, cutting or breaking.
Did you tear any of that woman’s clothing? No

This question does not address direct involvement as direct

involvement is reserved for the primary relevant question. The secondary
relevant question(s) shall be constructed to require the same answer as the
primary relevant question(s).

Types of secondary relevant questions:

o Evidence-Connecting. An evidence-connecting question is
designed to determine if the examinee was involved with any
of the evidence of the crime or is aware of the nature or
location of various items of evidence. ( Do you know where
any of that money is now?
o Guilty-Knowledge. A guilty-knowledge relevant question is
used to determine if the examinee has any knowledge of who
committed the incident under investigation. (Do you know for
sure who shot that man? ; Do you know who stole any of
that money?)
 o Secondary-Involvement. A secondary-involvement relevant
question tests for secondary involvement such as seeing or
hearing or focuses on physical acts that support the primary
offense. (Did you participate in taking any of that money)

II. COMPARISON QUESTIONS (PROBABLE LIE and DIRECTED LIE)

a. Probable lie

In most specific issue testing formats, a comparison question is designed

to be a probable lie. A probable lie is defined as the denial of a misdeed that a
person has more-than-likely engaged in or considered.
In constructing probable lie comparison questions, they should be similar in
nature, but unrelated to the specific crime or issue being tested. To ensure that the
comparison question is not potentially relevant, a comparison question must be
separated from the relevant issue by time or place. For example, in an issue
involving the theft of Php 10,000 in February 2024, the primary relevant and
comparison questions might be:
Relevant question: Did you steal any of that money?

Comparison question (time): Before 2024, did you ever steal anything?
(Time bar is before crime occurred)

Comparison question (place): Before arriving in Laoang, did you ever steal
anything? (Crime occurred after examinee arrived in Laoang)

In certain instances, comparison questions can also be separated from the

relevant issue by category. However, the examiner must exercise extreme caution
in using a “category” comparison question without a time or location event. This is
to ensure that the life experiences incorporated in the comparison question do not
fall within the relevant issue category, thus invalidating the entire examination. In
some instances, category comparison questions that appear to be separated from
the relevant area may actually be relevant. For instance, if an examinee is
suspected of stealing their roommate’s private property, a proper “category”
comparison question might be: “Did you ever steal any government property?”
However, an examiner would not want to ask this same examinee, “Did you ever
steal anything at all.” This comparison question would clearly incorporate the
roommate’s stolen property. As such, it would become a relevant question rather
than a “category” comparison question.
A probable lie comparison question should be broad in scope and time so
that it captures as many past life experiences of the examinee as possible.
Theoretically, as the scope and time of the comparison question increases, the
more significant that question becomes (signal value increases) for the examinee.
Each comparison question in a specific issue examination should address a single
issue, i.e., should not be a compound question.

Acceptable: Before, this year, did you ever steal anything?

Unacceptable: Before this year, did you ever steal anything from a loved
one or a place you worked (compound issue in same question)?

Typically, if a comparison question contains an “and” or an “or” in describing

the issue, it will consider an unacceptable compound question.

Imagine that at 2100, 11 Feb2024 Patrolman Jones was stopped for a

routine traffic inspection. At that time, a bag of marihuana was found in his car.
Jones denied any knowledge or involvement with the marihuana. If the following
comparison question was utilized, it would be improper, even though it is separated
from the relevant issue by time: “Before 2024, did you ever smoke any
marihuana?” If it incorporated any active-duty time prior to February 2024 this
comparison question would be improper. In this instance, using a “lie” comparison
question would be the proper type of comparison question to use, such as, “Before
2024, did you ever lie about anything important?”
As described previously, there are some instances where using the same
action verb in the comparison and relevant questions would be inappropriate. In
these instances, it is generally acceptable to utilize “lie” comparisons only.
Typically, the categories of relevant issues where lie comparison questions would
be more appropriate are: (1) Confirmatory testing, (2) Use of illegal drugs, and (3)
Child sexual abuse (examination of pedophiles). Finally, in most instances, it is
allowable to use one “lie comparison” in most examinations as long as that lie
comparison is related to the relevant issue. For example, in a theft issue, it is
generally acceptable to utilize one comparison question that is phrased: “Before
_____, (separated from relevant issue) did you ever lie about anything you ever
stole?”

When an examinee makes an admission to a comparison question, it must

be incorporated into the question so the examinee can still answer "No." Typically,
this is accomplished by adding a prefix at the beginning of the comparison
question. Some examples of possible comparison question prefixes are (not all
inclusive):

Other than what you told me, before 200___, did you…

Besides what we discussed, prior to arriving at_____, did you…

Other than that one time, before your ____ birthday, did you…

Types of Probable Questions:

• Exclusionary Comparison Question (ECQ). A probable-lie

question should be similar in nature but unrelated to the issue being
 tested. The question should be separated from the relevant issue by
time, place or category. The comparison question should use the
same action verb or similar in nature action verb as that of the
relevant issue. A comparison question should be broad in scope and
time so that it captures as many of the examinee's past life
experiences as possible. (Before this year, did you ever steal
anything from someone who trusted you?)
• Screening Comparison Question (SCQ). The SCQ question
should be similar in nature or motivation as the primary relevant issue
being tested and one to which the examinee will lie or be doubtful as
to the accuracy of the answer. The SCQ topic should be broader in
scope and of lesser import than the relevant issue. These questions
allow the examinee to make a clear dichotomy between the relevant
and SCQ as established by a thorough pretest explanation. The SCQ
is used only for CSP or fall scope polygraph issue examinations. The
SCQ may be used for both initial and breakdown series. (Have you
ever talked around classified information over-the telephone?)
b. Directed-Lie Comparison.
The DLC question is a specialized comparison question. A properly
constructed DLC question involves a minor transgression which should have some
personal significance to the examinee. Upon acknowledging having committed
such a transgression, the examinee is directed to lie when asked that question on
the test. The question is separated from the relevant issue by category. (Did you
ever commit a minor traffic violation?)

III. SACRIFICE RELEVANT QUESTION

A sacrifice relevant question is designed to serve as an introduction to the relevant

issue questions. During the data collection phase, whether innocent or guilty, some
examinees may physiologically respond to the first relevant question that is asked. As
such, in testing formats that have a sacrifice relevant question, it is not scored during
the test data analysis phase. Thus, it is "sacrificial." In most testing formats there is a
standard format for the sacrifice relevant question. It is worded so the examinee will
answer "Yes.” Most specific issue testing formats will provide the exact wording of the
sacrifice relevant question. Below are some generic examples:
Regarding that stolen vehicle, do you intend to answer each question
truthfully? Regarding your complaint about that man, do you intend to
answer each question truthfully?
Regarding whether or not you shot that man, do you intend to answer each
question truthfully?

IV. SYMPTOMATIC QUESTIONS

 Symptomatic questions are designed to test for an outside issue that may be more
significant (have more signal value) for an examinee than the comparison and relevant
issues. Symptomatic questions are not scored during the test data analysis phase of
examination. Symptomatic questions are utilized in the “Zone Comparison Test”
format. When used, the specific test will dictate their placement in the testing format.
Some examples of symptomatic questions are:
Do you believe I will only ask you the questions we reviewed?
Is there something else you are afraid I will ask you a question about?
V. IRRELEVANT QUESTIONS
An irrelevant question is the first question asked during examination. It is designed
to introduce the examinee to the beginning of the test questions and to allow the
orienting response to habituate before a scoreable question (comparison or relevant)
is asked. It is a neutral question that must be unrelated to the issue being tested.
Irrelevant questions are worded so the examinee always answers "Yes." They are not
scored during the data analysis phase. During the data collection phase of most
specific issue formats, an irrelevant question can also be inserted after an artifact
occurs or there is continued response from previous stimuli. This will allow the
examinee to return to a state of homeostasis before a scoreable question is asked.
Some examples of irrelevant questions are:

Are you now in Laoang?

Are the lights on in this room?

Are you now sitting down?

VI. SKY Question - These are three groups of question place as one by Backster and they
are intended to verify the previous charts and detect indirect participation or guilty
knowledge. The "S" stands for "suspect"; the "K" stands for "know"; and the "Y"
stands for "you". Example, "Do you SUSPECT anyone in particular in stealing
Florida's money?" "Do you KNOW who stole Florida's money?" and "Did YOU steal
Florida's money?"

VII. SUPPLEMENTARY TEST QUESTIONS

a. Peak of tension test

The subject may be given this test if he is not yet informed of the details of
the offense for which he is being interrogated by the investigation, or by other
persons or from other sources like the print media. This valid test is only made
when there is no widespread publicity about a crime where intimate details as to
the methods of commission or certain facts of the case is known from the victim
and investigator.

The questions formulated are similar in nature and construction, only one
of which is true and the perpetrator who would naturally be in possession of such
 unpublicized knowledge will usually exhibit a rise in the tracing up to that particular
question followed by a decline thereafter, caused by the relief of knowing that a
dreaded question dangerous to his well-being, is past.

Examples of Peak-of-Tension Test

• Do you know whether the stolen watch from Rommel is a Rolex?

(This is an introductory phrase plus padding question)
• Is it an Omega? (Padding)
• Is it a Seiko? (Padding)
• Is it Timex? (Relevant question)
• Is it Alba quartz? (Padding)
• Is it a Citizen? (Padding)

b. Quilt Complex test

This test is applied when the response to relevant and control
questions are similar in degree and in consistency and in a way that the
examiner cannot determine whether the subject is telling the truth or not.
The subject is asked questions aside from the irrelevant, relevant and
control questions, a new series of relevant questions dealing with a real
incident and that which the subject could not have committed.
If the subject does not respond to the added relevant questions, it
indicates that the subject was being deceptive as to the primary issue under
investigation. However, no conclusion can be drawn if the response to
added guilt complex is similar to the real issue questions.

Example: Did you steal that Ferrari toy collection?

On this case the stealing of Ferrari toy collection was not
actually committed because the alleged crime investigated was the
loss of the Ipod.

SUMMARY

Relevant questions in examinations directly relate to the matter being investigated

and must be constructed with clarity and conciseness, avoiding unnecessary language.
They should match the intelligence level of the examinee and steer clear of complex
wording to ensure understanding. Each question should address a single issue without
compound constructions and pose a dichotomy for "Yes" or "No" responses to avoid
ambiguity. Emotive language, legal jargon, and accusatory phrasing should be avoided
to maintain neutrality. Examiners prioritize issues by selecting the most serious and
focusing on the action describing what the suspect did to commit the crime. Primary
relevant questions test for direct involvement, typically requiring a "No" response and
avoiding probing perceptions or intent. Secondary relevant questions address supporting
elements of the primary issue, such as secondary involvement or knowledge of evidence,
also requiring a consistent response format.

Comparison questions serve a critical role in lie detection examinations, with two
main types: probable lie and directed lie. A probable lie question is designed to elicit a
deceptive response from the examinee by denying a misdeed they likely engaged in or
considered. These questions must be unrelated to the specific crime and are typically
separated by time, place, or category to ensure they are not potentially relevant. Directed
lie questions involve acknowledging a minor transgression and then lying about it during
testing. Both types aim to create a clear contrast between truthfulness and deception.
Sacrifice relevant questions introduce the examinee to the testing process, serving as an
initial neutral question to allow the orienting response to habituate before scoreable
questions. Symptomatic questions probe for outside issues with significant signal value
for the examinee but are not scored during analysis. Irrelevant questions also serve an
introductory role, eliciting a "Yes" response from the examinee and unrelated to the issue
being tested. The SKY question, consisting of three groups regarding suspicion,
knowledge, and personal involvement, verifies previous charts and detects indirect
participation or guilty knowledge. Finally, supplementary test questions like the peak of
tension and quilt complex tests provide additional insight into deception, often through
eliciting physiological responses to specific scenarios or unpublicized details of the crime.

ACTIVITY:
Students will be divided into groups. They will formulate different types of
questions based on hypothetical scenarios provided. They will present their works and
critique each other's questions, ensuring they meet the specified criteria for clarity,
simplicity, and adherence to guidelines.

SUGGESTED READINGS:
• Honts, C. R., Raskin, D. C., & Kircher, J. C. (2013). The Comparison Question
Test: Testing Polygraph Procedures and Measurement Techniques. In Handbook
of Polygraph Testing (pp. 85-112). Academic Press.
• Horvath, F., & Palmatier, J. J. (1995). The detection of deception: A review of the
literature. U.S. Department of Justice, Office of Justice Programs, National Institute
of Justice.
• Lykken, D. T. (1998). A Tremor in the Blood: Uses and Abuses of the Lie Detector.
Plenum Press.
• Matsumoto, D., & Hwang, H. S. (2012). Assessing deception. In Handbook of
Psychology (pp. 515-533). John Wiley & Sons, Inc.

REFERENCES:
• Polygraph Guide for Standards and Practices dated 25 October 2018
https://antipolygraph.org/documents/plea-guide-2018-10-25.pdf
• Federal Psychophysiological Detection of Deception Examiner Handbook dated 2
October 2006 https://antipolygraph.org/documents/federal-polygraph-handbook-
02-10-2006.pdf
• National Center for Credibility Assessment Test Question Construction Dated
March 2011. https://antipolygraph.org/documents/ncca-test-question-
construction.pdf