News & Resources

Publications & Media

“How the Science of Memory Can Be Used in Fact Witness Questioning”

DRI For the Defense Daniel Long (co-author)

In personal injury and wrongful death litigation, memory plays a central role, forming the foundation on which the “facts” needed to determine ultimate liability rely. Too often, the reliability of that memory is not explored or even questioned, due to widespread misunderstandings about how memory works. Contrary to popular opinion, memory is not made up of photographic events that are completely accurate when first experienced, remain completely intact while stored in the mind, and are exactly accurate when retrieved. The reality is that there are a host of factors that affect the reliability of a memory, including many factors common to litigation. Understanding how memory works is essential to being able to question witnesses effectively and assess the reliability of the witnesses’ testimony. The science of memory can also help litigators successfully navigate the often-dangerous job of demonstrating to a jury that a witness’s testimony is not reliable, by shifting the focus from the credibility of often sympathetic witnesses to the reliability of those witnesses’ memory. This article will address the science of memory and the factors that can affect the three stages of memory, provide some strategies for identifying those factors and accounting for the reality of memory in fact witness examinations, and explain how that reality can be used to set up the use of memory experts at trial. This article will not, however, address how to convince a judge to allow expert testimony on the reliability of a witness’s memory.

Memory Is Not as Reliable as You May Think
Often, we find ourselves looking to our memory to make decisions about future actions. We base our behaviors on previous experiences, what we have learned, and what we recall of past events. Similarly, in litigation, a witness’s account of what she or he saw, heard, or was told often are taken as uncontestable fact by lawyers, judges, and juries and then used to decide liability. Memories are treated as essentially “flash-bulb” photographs that never change after they are recorded, no matter how often the witness recounts the memory, despite the well-established and accepted body of scientific literature demonstrating that this is not true. Memories are not physical copies that are indelibly preserved; they rely on an active reconstructive process. Research on memory demonstrates that the ability to re-call memories of a past event accurately is a complex task that can be affected by a wide variety of factors. There are three stages of memory, each of which is susceptible to error: encoding (when a person perceives the to-be-remembered event); storage (the time between encoding and memory retrieval); and retrieval (when a person attempts to re-call the remembered event).

Encoding: The Act of Forming a Memory
When a person perceives an event, both external environmental and internal observer variables can influence the accuracy with which event information is encoded into memory. Environmental variables (e.g., viewing at a distance, relative positioning and line of sight, viewing for a brief period of time) can decrease memory accuracy. If environmental conditions are poor, less information about the event will be available for the creation of a memory to begin with, resulting in a memory that is incomplete or possibly distorted. Furthermore, internal factors, including insufficient direction of attention, the stress of a situation, or a person’s expectancies and awareness, may also impair memory formation and result in an incomplete or possibly distorted memory. D.L. Schacter, The Seven Sins of Memory: Insights from Psychology and Cognitive Neuroscience, 54(3) Am. Psych. 182–203 (1999); E.F. Loftus & T.E. Burns, Mental Shock Can Produce Retrograde Amnesia, 10 Memory & Cogn. 318–23 (1982); M.R. Leippe & G.L. Wells, Crime Seriousness as a Determinant of Accuracy in Eyewitness Identification, 63(3) J. Appl. Psych. 345–51 (1978). Anything that impairs the witness’s ability to perceive an event accurately is the first factor that can affect the reliability of a memory when it is recalled later.

Storage: Filing a Memory Away
The second stage of memory, storage, is also susceptible to error. First, as time passes between the to-be-remembered event and when a person tries to recall that event, details may be forgotten. A.D. Bad-deley, Working Memory, 321C R Acad. Sci III., 167–73 (1998). Information learned or suggested during memory storage, called post-event information, can erroneously become integrated with the memory of the original event. Extensive scientific research has demonstrated the powerful ability of post-event information to distort or alter a person’s original memory. E.g., E.F. Loftus, Eyewitness Testimony (1979); D.L. Schacter, Memory Distortion: History and Current Status, in Memory Distortion 1–43 (D.L. Schacter et al. ed., 1995). Even inferences that a person makes about how things must have occurred can bias and distort the original memory and can introduce memories for events that did not occur but are consistent with the outcome of an event. For example, researchers have demonstrated that subjects can experience an illusion of memory: they commonly will make a mistake, a “backward infer-ence,” falsely reporting that a sequence of pictures included an image that was never presented to the subject but which was logically inferable in the context. S.L. Hannigan & M. Tippens Reinitz, A Demonstration and Comparison of Two Types of Inference-Based Memory Errors, 27(4) J. Experi. Psych.: Learn. Mem. & Cogn. 931–40 (2001). Backward inference mistakes are also more likely to occur with the passage of time, often substituting correct episodic memories, which decay with time. Id. Once multiple sources of information become integrated, future retellings will strengthen the new details and make them indistinguishable from the original true details. In essence, an unreliable “new” memory replaces the original memory.

Retrieval: The Act of Accessing a Stored Memory
Errors in memory can be introduced at the time of retrieval, that is, when a per-son tries to recall the event. Leading questions, biased tests, inferences, schemas, imagination, and knowing the outcome of an event can all serve to distort and alter memory. For example, subtle word choices used in probing a memory can alter that memory in significant ways. E.F. Loftus & J.C. Palmer, Reconstruction of Automobile Destruction: An Example of the Interaction Between Language and Memory, 13 J. Ver-bal Learn. & Verbal Behav. 585–89 (1974).

This effect is particularly pronounced in recollecting emotionally negative events as in the types of accidents and injuries that are often the basis of litigation. S. Porter, L. Spencer, & A.R. Birt, Blinded by Emotion? Effect of the Emotionality of a Scene on Susceptibility to False Memories, 35(3) Can. J. Behav. Sci. 165–75 (2003). One scientific study found that when subjected to misleading questions, participants who were recollecting negative events were twice as likely to recall seeing major false details than participants who were recollecting neutral or positive events. S. Porter, L. Spencer, & A.R. Birt, supra. In the negative event group, 80 percent of participants incorporated major misleading information into their testimony. This vulnerability of memory to retrospective influences can be attributed to the reconstructive nature of remembering. One study on people’s recollection of autobiographical facts found that participants’ performance was as accurate for false memories deliberately inserted by the experimenter as it was for real memories. S.F. Larsen & M.A. Con-way, Reconstructing Dates of True and False Autobiographical Memories, 9(3) Eur. J. Cogn. Psych. 259–72 (1997). This seemingly paradoxical finding can be explained by participants’ reliance on incidental details and logical constraints to make correct judgments about their false memories, illustrating that their recollections resulted from reasoning rather than retrieving self-contained memories.

Even when detailed memories are formed, and even when they are retrieved with relatively little distortion, witnesses may be highly inaccurate about the chronological sequence of these memories, such as which of two remembered events occurred before the other. Studies in which participants were asked to record events—either in real time or from memory—from their own personal lives and subsequently arrange the events in chronological order have found very low performance accuracy on these ordering tasks. In one study, on average, participants placed only two events of six into the correct sequential order. C.D.B. Burt, S. Kemp, J.M. Grady, & M. Conway, Ordering Autobiographical Experiences, 8(5) Mem. 323–32 (2000). From these findings, the study’s researchers concluded that “[t]he general proposition that an event-ordering task might produce a reasonably accurate expression of the temporal nature of past experiences are not supported by the data,” and “errors of temporal memory are prevalent.” A thematic relationship between to-be-ordered events (e.g., events related to a particular holiday) did not improve performance. Furthermore, a comparison of performance within and across different time periods indicated that the temporal order of events belonging to the same epoch is remembered more poorly than that of events belonging to different epochs. This finding was supported by another scientific study discovering that performance for events closer in time was inferior to performance for events more spread apart in time. J. Skowronski, W.R. Walker, & A. Betz, Ordering Our World: An Examination of Time in Autobiographical Memory, 11(3) Mem. 247–60 (2003). In other words, the more closely events occur in time, the more difficult it is to remember their actual temporal order correctly.

This finding was replicated by yet another study, in which events close together in serial position were more likely to be confused with one another in a tem-poral-order judgment task. E. Altmann, Reconstructing the Serial Order of Events: A Case Study of September 11, 2001, 17 Appl. Cogn. Psych. 1067–80 (2003). The study, which examined the accuracy of eyewitness memory for ostensibly “unforgettable” events (i.e., the attacks of September 11, 2001), found that “even when people are able to reconstruct the ‘gist’ of a sequence,… they often get neighboring events out of order.” Between 59 and 63 per-cent made at least one error in recollecting the temporal order of the “unforgettable” events of 9/11. The researchers, summarizing their findings, offered a caution:

These high error rates may offer yet another reason to be skeptical of eyewitness accounts of critical event sequences.… serial-order reconstruction can distort in a way that could materially affect causal reasoning by reversing the order of two neighboring events—even events that the observer may consider “unforgettable.”

The confidence with which a person retrieves a memory may also convey an inaccurate impression. In typical, everyday life, when someone is more confident in the reliability of their memory, that memory tends to be more accurate. As a consequence, most people intuitively believe that high confidence is always associated with high accuracy. Contrary to common sense, however, psychological research has shown that under certain conditions, confidence and accuracy can become disassociated—e.g., confidence increases as accuracy decreases—and the person’s confidence does not reflect the accuracy of his or her memory. E.g., T.A. Busey, J. Tunniciff, G.R. Loftus, & E.F. Loftus Accounts of the Confidence-Accuracy Relation in Recognition Memory, 7 Psychomet. Bull. & Rev. 26–48 (2000); K.A. Deffenbacher, Eyewitness Accuracy and Confidence: Can We Infer Anything About Their Relation? 4(4) Law & Human Behav. 243–60 (1980); S. Penrod & B. Cutler, Witness Confidence and Witness Accuracy: Assessing Their Forensic Relation, 1 Psych. Pub. Policy & Law 817–45 (1995).

In another study examining memory for the events surrounding September 11, subjects on September 12 were asked to write down a description of where they were when they learned of the terrorist attacks and also to report a memory of a recent, everyday event. J.M. Talarico & D.C. Rubin, Confidence, Not Consistency, Characterizes Flashbulb Memories, 14(5), Psych. Sci. 455–61 (2003). Recall for both types of experiences became inconsistent as it was tested across one-, six-, and thirty two-week delays. Subjects introduced more inconsistent details and omitted more key details as time passed. Despite the inconsistency in their accounts for all types of memories, subjects were more confident in their memories for the September 11-related memories than everyday memories. This data suggests that all types of experience-based memories are susceptible to distortions over time, including emotionally charged memories, and emotional memories are more likely to be perceived as accurate. Thus, while people may genuinely believe the authenticity of their memory, these reports are often unreliable. When a person witnesses an event, only fragments of that event are encoded into memory. These fragments are then integrated with other information sources such as prior expectations, inferences about how things must have occurred, facts previously stored in memory, and facts introduced after the event occurred. In this manner, recalling an event from memory is not like playing a record of the original event. Instead, the memory is reconstructed from a host of information sources that can be highly inaccurate. Ultimately, determining what actually happened in the past is not as easy as just asking a witness what she or he remembers.

The Unreliability of Memory in Litigation
All of the ways that memory can deteriorate, change, and become distorted pose a challenge in litigation where fact witness testimony is a major, and sometimes only, source of the information used to determine liability. The general unreliability of memory is exacerbated by multiple factors found in litigation that can also influence memory, such as the stress of depositions, the financial incentives to the plaintiff, and the introduction of post-event information into the memory process through deposition preparation. Understanding how memory works before you question the first witness will help you better organize your examination, assess the reliability of the witness’s memory during the examination, and set up the use of memory experts at trial to help demonstrate how unreliable the witness’s memory was without having to attack the witness’s credibility.

Addressing the reliability of the memory can be difficult because there are so many potential factors involved, most of which are specific to each witness and case. Fortunately, many of the normal deposition strategies commonly used to test whether the witness has a basis to know what she or he testifies about will address many of these factors. To further help, we have attempted below to suggest some other strategies for exploring the reliability of the memory.

As we have mentioned already, the first step is to change the typical defense lawyer thought process of challenging the credibility of the witness to challenging the reliability of a witness’s memory. Thinking in terms of reliability helps focus the inquiry on whether any of the factors commonly known to affect the reliability of memory are involved with each witness. Focusing on the factors affecting the reliability of a memory also depersonalizes the questions in a way that makes it easier to ask probing questions, which, at times, delve into sensitive areas, without drawing an objection or making the witness unnecessarily defensive. And this focus makes it easier to establish the factual predicate that memory experts will need to assess the reliability of the memory and to convince a judge thatyour memory expert’s testimony can satisfy the Frye/Daubert standards to allow the testimony. Remember, the ultimate goal in cases where there is a real dispute about facts related to the accident or injury is to be able to address the reliability of the witness’s memory without having to risk alienating a jury by attacking the credibility of the witness. By focusing on reliability factors, you can present a compelling scientific and factual argument that what a witness claims to recall is simply not a reliable memory and that witness’s testimony should be not be considered by the jury in its determination of liability. But to make that argument at trial, you have to start in the fact witness depositions.

Instructing the Witness
The first thing done in any fact witness deposition is to instruct the witness about the ground rules of depositions. This is where you should first address memory issues. Depositions create a number of factors that can influence the reliability of a witness’s memory, including the normal stress that the witness experiences, the fear of appearing stupid or having a faulty memory, and the common belief that the lawyer expects the witness to know the answer to each question asked. Putting the witness at ease about what will happen and how the deposition will proceed can reduce the witness’s stress and minimize yet another factor that could impair memory.

Explaining the mechanics of the deposition as you always do is helpful. But going further, to explain to the witness that “I do not know” is often the correct answer and does not mean that the witness’s memory is faulty, is essential. It is also important to instruct the witness that many, if not most, of the questions that you ask are designed to confirm what the witness does not know. Another important instruction, particularly in cases where the injury was caused by a product the witness routinely used, is to explain the importance of not using general memories or guesses to answer questions about the specific product or specific incident. For example, if a witness is injured by a product that she or he uses regularly, when asked about the specific product or the specific incident that caused the injury, the witness may not want to admit that she or he does not really recall what happened at that time as a natural result of stresses of the incident; instead, the witness may either use a memory of how she or he generally used the product or infer a memory about past use to answer the specific questions. Similarly, explaining the difference between belief and knowledge can help the witness understand that important difference, and hopefully, it will increase the odds that the witness will not claim to know something because the witness believes it is true.

Although not part of the actual instructions, make sure to provide positive reinforcement to the witness throughout the deposition when the witness follows those instructions, such as testifying “I don’t know.” It is human nature to seek positive reinforcement, and by providing it, you can help encourage the witness to behave as you would prefer. Realistically, there is no guarantee that a witness will listen to or follow these instructions, but for a witness who is trying to tell the truth, and such witnesses do exist, these instructions can help you obtain more reliable information.

Encoding: Establishing the Factors that Can Affect the Reliability
Unfortunately, many of the fact witnesses who we depose witnessed traumatic events that happened unexpectedly, quickly, and sometimes, to loved ones. And those factors can undermine the reliability of the witness’s perception of the event and how accurately it is encoded. One of the most important points to remember with the encoding process is that the belief that traumatic events are imprinted more forcibly and accurately into our memory than other events is simply false. The reality is that traumatic circumstances can make memory much less reliable instead.

In these depositions, we always walk the witness through her or his recollection of the event that caused the injury. And while it is important to establish the details necessary to determine what actually happened, it is also important to discuss all the circumstances or factors surrounding the event that could have negatively affected the reliability of the memory formation. Some of the factors to establish include how suddenly the event happened; how it unexpected it was; how the witness was not watching the event to remember the details purposefully; any environmental factors that could have affected the ability to perceive the event accurately; the witness’s personal feelings of surprise, fear, worry, anxiety, etc.; the witness’s focus on the injured plaintiff and attempts to help; whether the witness immediately recorded the chain of events in some fashion after the event; and whether the witness recounted the event in a consistent manner shortly after the event and at other times since then. It can feel uncomfortable eliciting testimony about what are essentially damages-related factors, but that testimony will come out at trial, and it is essential to identify some of the main factors affecting the encoding process.

One thing to keep in mind about this phase of the inquiry is the importance of doing your homework before the deposition to identify all other sources of information about the event that can be used to question the witness better. We live in an increasingly surveillant society in which cameras are everywhere. Not only is video a much more objective record of events than human memory, jurors are now conditioned generally to give greater credence to video evidence than human perspective. To the extent that you can, find all possible video of the incident before you depose or otherwise question a witness. Being able to confront a witness with this type of evidence can either prevent a witness from testifying about a distorted memory, or at the very least, provide an objective basis to challenge the reliability of the memory.

Storage: Establishing What Might Have Affected a Memory
Most depositions generally do not explore memory storage. But doing so can be an important way to establish factors that could have changed the memory. Asking the witness how many times she or he was asked to recount the memory is one step. Pay particular attention to more recent retellings of the event, including the times that the witness talked to the plaintiff’s counsel and prepared for the deposition. Studies show that new information, such as that which would commonly be shared by a plaintiff’s lawyer during deposition preparation, can actually have a greater effect on memory than information introduced closer to the time the memory was formed. While it may be difficult to get a witness to discuss what she or he discussed with plaintiff’s counsel, you can ask if the witness was shown any photographs, videos, testimony, or other information that the witness had not previously seen. To the extent that you are able, obtain those videos, photographs, or other information so that your expert can assess its potential effect on the memory. If the witness’s memory perfectly matches how a certain video showed the event, particularly when the camera provides a point of view that the witness did not have at the time, this can demonstrate that the witness has incorporated information into the memory beyond what she or he actually perceived.

Another step is asking the witness if she or he talked with other witnesses about the event and what those witnesses saw. As with comparing testimony to a video, comparing the recollection of the witness and another person with whom the witness discussed the event can provide further evidence that the witness has incorporated information that she or he came by later. Another way to assess if the witness’s memory is reliable is to ask the witness if the witness sees her- or himself in in the memory as she or he recalls it. If the witness sees her- or himself in the memory, then that generally is a created memory.

Beyond exploring the factors associated with recounting and solidifying the memory, it is also important to explore factors personal to the witness that could affect the witness’s brain. Some of those factors would include any health issues that could affect memory, any medications that the witness takes, the witness’s age, and any psychological issues, either caused by the event about which she or he will testify or originating elsewhere. Exploring the witness’s memory about non-incidentrelated events, such as birth dates and other important milestones, is also important to establishing a memory baseline against which you can compare the memory of the event about which the witness will testify. The ultimate goal in this area of inquiry is once again to establish all the factors that could negatively affect the reliability of the witness’s memory or even create a new memory, based on outside factors.

Retrieval: Identifying Testimony Reliability
The final stage of memory is retrieval, or in our context, eliciting testimony from a witness. This is the part of memory assessment that you are likely the most familiar with, and you already use many strategies to address it. Strategies such as ruling out all the ways the witness could know a key fact before asking if the witness knows that fact, establishing all the areas of information that the witness does not have, and other similar common strategies are all important and effective.

Perhaps the most important point for this stage of the inquiry is simply understanding all the factors that can affect the reliability of the retrieval stage. One key point in this stage is to understand that the apparent confidence a witness has in a recollection bears no correlation to the actual accuracy of the recollection. Studies have shown that many witnesses are more confident in false memories than in true memories. And as discussed above, contrary to popular belief, memories of traumatic events, which often are the basis of the lawsuit, are generally less reliable than non-traumatic events.

By exploring all the factors that can affect the reliability of the memory’s formation and storage, you will have already addressed many of the factors that can negatively affect the retrieval stage. This next stage of the examination, the deposition, can also identify other areas of inquiry that you want to pursue through discovery, and possibly, other depositions. To the extent that the witness testifies about receiving post-event information that you have not already received that could have affected the memory, you may want to consider continuing the deposition or leaving it open so that you can obtain the newly identified information to question the witness further. Retrieval errors are inherent in litigation where each witness has been “prepared” for her or his deposition, so do not forget to address this last stage of the inquiry.

Post-Deposition Analysis
Analyzing the reliability of the witness’s memory does not stop once the deposition is completed. As mentioned above, you may identify new evidence that you wish to obtain, such as videos or photographs that the witness was shown, or identify new witnesses who you want to question because of their interaction with the justdeposed witness. It is important to take the time to go over your notes or the transcript to decide if this witness or this case could use the help of a memory expert. And if you already have retained a memory expert, it can be helpful to discuss with the expert what you learned from the deposition or depositions to identify other information that the expert will need to render an opinion in the case.

Finally, if you do go to trial and have to question the same witness again, explore any other post-event information that may have been introduced to the witness after the deposition to identify other potential memory factors.

Conclusion: Memory, Now You Know
While we all understand that deposing fact witnesses involves asking about their memories, too often lawyers do not understand just how unreliable memory can be. As a result, lawyers may accept at face value testimony that was probably highly unreliable. Understanding the factors that are important to assessing the reliability of the witness’s memory can allow more effective and detailed examination of the fact witness. Beyond the deposition, though, understanding how a memory expert can be used to discount the reliability of a witness’s memory can provide an additional, and often less risky, defense in factually disputed cases. Many, if not most, product liability cases hinge on the memories of a few witnesses who are often highly sympathetic and compelling. Understanding, at least generally, the well-established body of reliable scientific literature on memory, and then using it in every phase of witness examination, can only help you better defend your cases. Cases turn on facts, so it is essential that the jury understand what actually is a fact, what is an unreliable memory, and why that is. Now that you understand this, you can use your skill, training, and experience to handle your witnesses more advantageously.

We have updated our Privacy Policy. Please click here to view our new policy.