The scientists involved in the ICC report, Dr Kimberley A Wade and Dr Ula Cartwright-Finch, tested the existing research in a commercial setting using an online study. The experiment confirmed that witness memory is liable to error. This was especially true when the participants were given roles as directors of one of the companies involved in the case study, or when participants were given biased internal memos on the facts of the case. The participants exposed to these biases were prone to present the facts differently when asked to recall them, and in ways that better supported their own company's case. A control group, which had not been presented with biased information and had not been given roles as company directors, recalled the facts differently, without the bias.
Contemporaneous recording of witness testimony
Recording events as they unfold will lessen the impact of memory contamination. This is the idea behind the first recommendation in the report: to establish procedures for teams to keep contemporaneous written or oral notes of issues being discussed at the time the relevant events unfold, especially in a potentially or actually contentious situation.
This measure should be considered carefully by in-house legal teams. For example, in the context of a construction project, measures could be put in place from the outset in order to record the context and detail of important events: discussions held during the negotiation of a contract amendment; agreement between the parties as to allocation of responsibility; or even setting out a chronology of facts in relation to a particular event which is likely to lead to a dispute – for example, a landslide on a road project.
From experience, gathering a witness' recollection of events can be particularly difficult. In a mega-construction arbitration, the efforts required by our team in order to compile the evidence related to a series of landslides on a linear project turned out to be challenging and costly. When witnesses were available, we often gathered conflicting accounts of the facts, while other potential witnesses had left the company and refused to participate in interviews. Ultimately, there were gaps in the story despite conducting dozens of interviews.
To avoid this, site teams could put in place measures to record important events via dated memoranda, signed by relevant team members. These memoranda could even be shared with the employer regularly, perhaps by appending them to progress reports, which would increase credibility of these documents while providing an opportunity for the employer to challenge the information provided.
We often highlight the importance of notices and of record-keeping. The contemporaneous recording of witness testimony would be a welcome addition to this practice and would allow a more efficient dispute resolution process, potentially by allowing a tribunal to rely on these records as written evidence in support for witness statements or even as an alternative to witness statements altogether – thereby reducing or even eliminating memory contamination.