Context and bias are important concepts when discussing any type of evidence evaluation. They apply to any and all forms of forensic evidence so some awareness of them is important.

First, evidence always exists within some context, more commonly referred to as a framework (or background), that is specific to the matter at hand. That is just an unavoidable fact which must be acknowledged. Even a purported absence of information is a type of ‘framework’ — it is just an artificial and unrealistic one.¹

Framework information can be relevant or irrelevant and, at the same time, either biasing or not biasing to the evaluation process. To clarify the terminology, we can use the Report of the Expert Working Group for Human Factors in Handwriting Examination which discusses all of this at length.[note]Forensic Handwriting Examination and Human Factors: Improving the Practice Through a Systems Approach. (NSITIR 8282). U.S. Department of Commerce, National Institute of Standards and Technology. https://doi.org/10.6028/NIST.IR.8282.[/note] That report provides the following definitions pertinent to this discussion:

• Bias: A systematic pattern of deviation.
• Cognitive Bias: A systematic pattern of deviation in human judgment.
• Relevant Information: Information that is pertinent and applicable to the subject, material, or question being considered. The consideration may be broad (i.e., discipline level) or specific (i.e., task level).

Stoel et al provide guidelines that can be used to determine when action should be taken to manage contextual information, based on the presence/absence of a bias effect and relevance/irrelevance of the information being provided.²

The relationships of interest are shown in the following table:

The above indicates that task-relevant information should be used in the evaluation process. That makes sense since the presence or absence of that information is likely to change the outcome. If the information is not biasing (bottom left cell in the table), then it can be safely provided to the examiner and used. If the information is biasing (or potentially biasing), then it should be used but carefully, and in accordance with some type of Context Information Management (CIM) System. Obviously, if information is irrelevant then the examiner should not be given the information, particularly if it is also biasing (or potentially biasing).³

So, what is a CIM, and how does it work? The approach to this issue will differ for a public laboratory vs. a private examiner. In a public laboratory with plenty of staff there may be an examiner who reviews a case to vet the material and control task-irrelevant information before handing it over to another examiner who handles the actual examination. There may even be a dedicated unit for case receipt and review, one that covers all sections in the lab. A private examiner, on the other hand, cannot do that and can only try to avoid problems by forewarning clients of the potential risk to their case. At the same time, mechanisms like blind peer review can be used in either situation.

Second, as the above indicates, framework information is not something to be avoided at all costs. It often provides information about the perpetrator, the scenario, or the specific alternative proposition that could be key to formulating propositions or using/considering an appropriate population under the main or alternative hypothesis. The real issue is having access to the right information, at the right time. That can be difficult to achieve in some situations.

When working with a private examiner the client is best off waiting to be asked for more information. Initially, provide only the most basic information and do not provide details to the examiner, until asked. That way the examiner has a chance to guide the discussion with a view to minimizing any potential issues while getting the necessary, relevant information.

1. This is a common result when inappropriate/excessive vetting is done, often with the laudable interest of avoiding bias.
2. See also: Stoel, R. D., Berger, C., Kerhoff, W., Mattijssen, E. J. A. T., & Dror, I. E. (2014). Minimizing Contextual Bias in Forensic Casework. In K. J. Strom & M. J. Hickman (Eds.), Forensic Science and the Administration of Justice: Critical Issues and Directions. Sage Publications.
3. The ‘trick’ in all of this is that some information has the potential to be biasing without actually such an effect. What is the cost-benefit to removing/controlling such information in the ‘off chance’ that it might affect things, particularly id that chance is very low? A blanket policy or approach may be very counter-productive.