ATP 2-33.4, page 30
Intelligence Analysis
Army Techniques Publication: Intelligence Analysis
2014 public edition (Archive.org)
Searchable page text (OCR / PDF)
Chapter 2
Personal Bias
2-41. Personal bias is the tendency to base assessments on personal beliefs. This can cause the rejection of
valid arguments that conflict with these beliefs. A racially or religiously prejudiced person may reject
arguments because of the source. A person with strong political views may discount every argument from.
another political group.
2-42. There are several types of personal bias. Three common biases exhibited by analysts are—
e Confirmation Bias. This bias causes analysts to undervalue or ignore evidence contradicting an
early judgment and value evidence that tends to confirm already held assessments.
e Assimilation Bias. This bias involves the modification and elaboration of new information to fit
prior conceptions of hypotheses. The bias is toward confirming a preconceived answer.
© Anchoring Bias. This bias involves the use, often unwitting, of arbitrary values in
decisionmaking, including the use of conclusions developed by others.
Cognitive Bias
2-43. The intelligence analyst evaluates information from a variety of sources. The degree of reliability,
completeness, and consistency varies from source to source and even from report to report. This variance
often creates doubt about the reliability of some sources. Cognitive biases that affect the analyst are
discussed below:
@ Vividness. Clear and concise or vivid information has a greater impact on analytical thinking
than abstract and vague information. A clear piece of information is held in higher regard than a
vague piece of information that may be more accurate. Analysts must consider that an enemy
may use deception to portray vivid facts, situations, and capabilities that they want the friendly
intelligence effort to believe.
© Absence of evidence. Lack of information is the analyst’s most common problem, especially in
the tactical environment. Analysts must do their best with limited information and avoid holding
back intelligence because it is inconclusive. To avoid this bias, the analyst should—
= Realize that information will be missing.
= Identify areas where information is lacking and consider alternative conclusions.
= Adapt or adjust judgments as more information becomes available.
= Consider whether a lack of information is normal in those areas or whether the absence of
information itself is an indicator.
e@ Oversensitivity to consistency. Consistent evidence is a major factor for confidence in the
analyst’s judgment. Information may be consistent because it is appropriate, or it may be consistent
because it is redundant, is from a small or biased sample, or is the result of the enemy’s deception
efforts. When making judgments based on consistent evidence, the analyst must—
= Be receptive to information that comes in from other sources regardless of whether it
supports the hypothesis or not.
= Be alert for circular reporting, which is intelligence already obtained by the unit that is then
reformatted by other units and intelligence organizations, modified slightly, and
disseminated back to the unit. This is a common problem; particularly in digital units, where
large volumes of information is being processed. It helps to know, to the degree possible,
the original source for all intelligence to ensure that a circular report is not used as evidence
to confirm an intelligence estimate or conclusion.
e Persistence on impressions. When evidence is received, there is a tendency to think of
connections that explain the evidence. Impressions are based on these connections. Although the
evidence eventually may be discredited, the connection remains and so do the impressions.
e@ Dependency on memory. The ability to recall past events influences judgment concerning
future events. Since memory is more readily available, it is easy to rely on memory instead of
seeking new information to support analysis.
© Acceptance of new intelligence. Often new intelligence is viewed subjectively; either valued as
having more value or less value than current intelligence.
2-14 ATP 2-33.4 18 August 2014
Chapter 2
2-14
ATP 2-33.4
18 August 2014
Personal Bias
2-41. Personal bias is the tendency to base assessments on personal beliefs. This can cause the rejection of
valid arguments that conflict with these beliefs. A racially or religiously prejudiced person may reject
arguments because of the source. A person with strong political views may discount every argument from
another political group.
2-42. There are several types of personal bias. Three common biases exhibited by analysts are—
Confirmation Bias. This bias causes analysts to undervalue or ignore evidence contradicting an
early judgment and value evidence that tends to confirm already held assessments.
Assimilation Bias. This bias involves the modification and elaboration of new information to fit
prior conceptions of hypotheses. The bias is toward confirming a preconceived answer.
Anchoring Bias. This bias involves the use, often unwitting, of arbitrary values in
decisionmaking, including the use of conclusions developed by others.
Cognitive Bias
2-43. The intelligence analyst evaluates information from a variety of sources. The degree of reliability,
completeness, and consistency varies from source to source and even from report to report. This variance
often creates doubt about the reliability of some sources. Cognitive biases that affect the analyst are
discussed below:
Vividness. Clear and concise or vivid information has a greater impact on analytical thinking
than abstract and vague information. A clear piece of information is held in higher regard than a
vague piece of information that may be more accurate. Analysts must consider that an enemy
may use deception to portray vivid facts, situations, and capabilities that they want the friendly
intelligence effort to believe.
Absence of evidence. Lack of information is the analyst’s most common problem, especially in
the tactical environment. Analysts must do their best with limited information and avoid holding
back intelligence because it is inconclusive. To avoid this bias, the analyst should—
Realize that information will be missing.
Identify areas where information is lacking and consider alternative conclusions.
Adapt or adjust judgments as more information becomes available.
Consider whether a lack of information is normal in those areas or whether the absence of
information itself is an indicator.
Oversensitivity to consistency. Consistent evidence is a major factor for confidence in the
analyst’s judgment. Information may be consistent because it is appropriate, or it may be consistent
because it is redundant, is from a small or biased sample, or is the result of the enemy’s deception
efforts. When making judgments based on consistent evidence, the analyst must—
Be receptive to information that comes in from other sources regardless of whether it
supports the hypothesis or not.
Be alert for circular reporting, which is intelligence already obtained by the unit that is then
reformatted by other units and intelligence organizations, modified slightly, and
disseminated back to the unit. This is a common problem; particularly in digital units, where
large volumes of information is being processed. It helps to know, to the degree possible,
the original source for all intelligence to ensure that a circular report is not used as evidence
to confirm an intelligence estimate or conclusion.
Persistence on impressions. When evidence is received, there is a tendency to think of
connections that explain the evidence. Impressions are based on these connections. Although the
evidence eventually may be discredited, the connection remains and so do the impressions.
Dependency on memory. The ability to recall past events influences judgment concerning
future events. Since memory is more readily available, it is easy to rely on memory instead of
seeking new information to support analysis.
Acceptance of new intelligence. Often new intelligence is viewed subjectively; either valued as
having more value or less value than current intelligence.