Episodic memory - 'Do you remember/the twenty-first night of September?'

Witness memory is an issue that keeps being scrutinised by the courts, such as in the recent case of SATA Internacional - Azores Airline SA v Hi Fly Ltd [2024] EWHC 2762 (Comm) (31 October 2024). In the first part our series on witness memory, we looked at common influences on memory. In this part we dive deeper into that subject and look at the importance of episodic memory when working with witnesses.

Before reading on, please watch this short video: 

What is episodic memory?

The science of memory is a vast subject (see for example link to ‘Legal aspects of memory: a summary of scientific evidence issued by the Psychology and Law Sections of the British Academy’). However, put briefly, human memory is not a unitary faculty; it comprises an alliance of systems. Three of those have a central importance in our context: working memory (a short-term memory system), semantic memory and episodic memory (both long-term memory systems). Episodic memory and semantic memory represent the distinction between remembering and knowing. Episodic memory is what allows us to remember events and to distinguish one from other similar events. To me (Andrew), it has a cinematic or pictorial quality (albeit, often grainy and flickering).

Why is episodic memory important?

The episodic memory system is of central importance when thinking about witness evidence, which is often used in litigation in establishing relevant facts, such as things said or done. It is also important to think about this because episodic memory is the most fallible of the memory systems. As Lord Justice Browne said in a 1982 article:

“I think that in civil cases (unlike criminal cases) the witnesses are seldom lying deliberately. But I am very sceptical about the reliability of oral evidence. Observation and memory are fallible and the human capacity for honestly believing something which bears no relation to what really happened is unlimited.”

Before we talk more about that, think back to the video you watched before you started reading. How many basketball passes did you see?

Imagine a dispute that turns on what was said at a particular meeting. Witness evidence could be crucial. It might be the case that the witness knew the meeting would be important and focussed on committing details to memory. Sort of like you counting basketball passes. However, did you see the gorilla walking through the middle of the group? Even attentive individuals often fail to commit key matters to memory.

In the seminal podcast series Serial, Sarah Koenig boiled down the guilt of Adnan Syed to a question of his whereabouts in a short period that was unaccounted for on 13 January 1999. If guilty, he would, presumably remember where he was (although he might lie about it). However, if innocent, how likely would it be that he could, following his arrest on 28 February 1999, distinguish that afternoon some 6-7 weeks earlier from any other mundane after-school period and recall what he was doing in a specific twenty minutes?

If we apply that line of thought to our civil claim example, what if the pivotal meeting was just one of many weekly meetings attended by a similar group over a long-term project? The witness might simply not remember and the potential for conflation and using semantic memory to fill in the gaps would be huge.

Episodic memory is also sufficiently malleable that patently false memories can be created. In a classic experiment at the University of Washington, a group of individuals were manipulated into recalling detailed memories of meeting Bugs Bunny at Disneyland. Of course, Bugs Bunny (a Warner Bros. character) would never be allowed on Disney property.

These issues can all be exacerbated in the context of litigation, where “memories” are often examined repeatedly and witnesses could have a significant personal stake in a particular version of events being true.

A few ideas on tackling the issue

Interview witnesses early. Delays in securing statements can lead to further deterioration in memory and more opportunities for the original memory to be contaminated or changed. Any trial witness statement should be prepared avoiding any practice that might alter or influence the recollection of the witness other than refreshing the witness’ memory. Science suggests that a witness’s first spontaneous account of an event is likely to be the most accurate, but also incomplete. However, while that indicates the best course, the very nature of the litigation process hampers early proofing. The witness evidence phase, which is often costly for the client (and therefore might be something the client does not wish to expedite), comes relatively late procedurally. It might also come after a key witness has provided inputs that inform pleadings, which, in turn, might mean the client’s case therefore relies on a particular, less fully explored version of events.

Interview planning is critical but while lists of pre-prepared questions can provide structure to an interview, they can also limit the depth and flexibility needed to obtain a full and accurate witness account. This is further complicated where question quality is questionable!, i.e. leading, misleading or complex for example. A more open-ended exploratory interview technique tends to yield more comprehensive and reliable information.

Hone interviewing techniques. Lawyers need to expand upon the information to gain a detailed account, but in doing so they risk distortion through inappropriate questioning. Advanced interviewing techniques are underutilised in civil litigation. Christie Knowles founder of CKITT has years of experience and works with Dispute Resolution lawyers at Burges Salmon. Christie is a strong advocate for adopting the PEACE model (see table below), a structured approach to investigative interviewing, designed to gather accurate and reliable information from suspects, witnesses or victims while promoting ethical and fair treatment. The PEACE model emphasises a non-coercive, ethical approach to interviewing, focused on obtaining truthful and useful information without resorting to manipulation or undue pressure. It is widely used in many law enforcement and investigative agencies and other jurisdictions influenced by best practices in interviewing.

We will be looking in more detail at questioning in our next blog.

This article was written by Andrew Walls and Stacie Bourton in the Burges Salmon Dispute Resolution team, and Christie Knowles of CKITT.