Memory Puzzle: What Every Investigator Should Know

By Brian D. Fitch, Ph.D.
Stock image of an investigator working on a desktop computer in an office.

Human beings process information with amazing efficiency and, in many cases, perform better than sophisticated machines on a variety of tasks, including problem solving and critical thinking.1 However, in spite of these extraordinary abilities, the human mind is plagued by a number of shortcomings. The natural tendencies of people to forget appointments, overlook important details, and distort the past demonstrate the inherent limitations of the human mind. Nevertheless, the human abilities to notice, remember, and recall play such pervasive roles that people often take them for granted until something goes awry.

A clear understanding of memory—how the brain processes, stores, and retrieves information—is vital to an investigator’s ability to separate fact from fiction. It is common to refer to memory as a filing cabinet full of individual folders, with each folder storing all the facts and details associated with a particular occurrence. As an event unfolds, a person’s senses record all facets of the incident, which then is stored in the brain for later retrieval. It is believed that this process works much like storing a photograph or a tape-recorded conversation in a sealed envelope. Regardless of when the envelope is reopened and the contents are removed, nothing will have changed. The images will be older, but no different, with the description of the event appearing precisely as it did when it first was deposited. The information stored in memory represents nothing less than an unvarnished account of the original event.


Despite the continued popularity of this theory, most experts believe that memory is far more complex and elusive. Memory does not transfer information from the outside world to the inner world “as is”; rather, it is a constructive and reconstructive process influenced heavily by attention, expectations, emotions, and other factors that shape how the brain processes, remembers, and recalls events.2

Subjects often lie to conceal their guilt; however, there are other instances when an inaccurate account results from systematic, well-documented memory biases or errors. The better an investigator understands the strengths and limitations of memory, the greater the chances of obtaining ground truth.  


Memory is not a single process. It involves a series of systems, each of which operates by its own set of principles. Perhaps, the best-articulated and well-researched memory paradigm is the three-stage model, which suggests that memory actually is the product of three different storage systems or stages—sensory, short-term, and long-term.3

Dr. Fitch
Dr. Fitch, a lieutenant with the Los Angeles County Sheriff’s Department, holds faculty positions with California State University in Long Beach and the Southwestern University School of Law in Los Angeles.


Sensory memory refers to the initial perceptual processing of incoming kinesthetic stimuli. These memories contain lingering impressions of information from the five senses. Research indicates that information in sensory memory is held only long enough for perceptual analysis to occur. Once completed, most of the information is discarded or quickly forgotten, much like the flash of a camera. During the brief time that information remains, the brain only pays attention to the most salient aspects. That information is matched to recognizable patterns and assigned meaning.4 Only the most significant features are transferred to the next stage for further processing.


The second stage of the model, short-term memory, is where conscious thought occurs—where thinking happens. Whenever a person, for instance, recalls an event, performs addition mentally, or reads a string of words, the information to execute these functions is held temporarily in short-term memory. Information in short-term memory remains longer than that held in sensory memory; however, it seems that much of the data is forgotten within 30 seconds unless some further processing occurs.5

One way to prevent short-term memory loss is to repeat information to keep it active, thereby preventing it from being displaced by new material. However, before the brain can retain material, it must transfer it to the final stage—long-term memory. Similar to the first stage of transfer—sensory memory to short-term memory—information that relates easily to existing memories has a better chance of being retained than novel data.  


Long-term memory is the storage of information for extended periods. Unlike the limited capacity of short-term memory, the amount of information stored in long-term memory is unlimited. Typically, when people talk about memory, they are referring to the material held in long-term storage.

Psychologists further distinguish between two types of long-term memory—procedural and
declarative.6 Procedural memory represents knowing how to perform certain activities, such as driving a car, using a computer, or throwing a baseball. Declarative memory refers to facts and knowledge (e.g., the capital of California, the location where a birthday was celebrated, or what a robbery suspect was wearing) that are consciously recalled. These are examples of explicit memory because the information is explicitly stored and retrieved.

The three-stage model of memory provides a better understanding of why victims, witnesses, and suspects often are unable to recall important information. A subject’s failure to notice, encode, or remember information can occur during any one of the three stages and for any reason. These shortcomings are exacerbated by the fact that a person’s memory is influenced heavily by attention, emotion, experience, and depth of processing.


To understand how memory works, it is important to explore the four fundamental principles of cognition: 1) the limitations of attention, 2) the roles of explicit and implicit processing, 3) the constructive nature of memory, and 4) the depth with which information is processed. Several interrelated factors, rather than any one isolated factor, influence how people process, remember, and recall events.


Attention—the mechanism we use to concentrate our effort on a stimulus or mental event—is similar to other cognitive processes in that it is limited. Because the brain continuously is bombarded with sensory stimuli, it must select what information it pays attention to and what it ignores. Whenever the brain chooses to attend to a particular stimulus, either consciously or unconsciously, it must ignore information from other sources.7 This is complicated further by the fact that individuals vary in the types of stimuli they find interesting or relevant and, therefore, worthy of attention and further processing. Thus, it is common for people experiencing the same event to perceive and remember it in different ways. It is nearly impossible for witnesses or victims to process simultaneously all the details of a crime—the perpetrator’s description, weapon used, what someone said, or other people who were present—without missing or misinterpreting important facts.


Some of the mental systems used to process, store, and recall information require conscious and deliberate effort; others operate automatically and involuntarily.8 Explicit processing—for instance, committing a list of words to memory, studying for an examination, memorizing a license plate number or a suspect’s clothing—involves conscious thought and awareness that a task is being performed. It is deliberate and willful—people intentionally choose what to pay attention to and what to ignore—and requires attention, focus, and extensive cognitive resources. A person’s ability to process information explicitly is limited.

In contrast, implicit processing occurs outside conscious awareness and typically requires little, if any, effort or willful attention—signing one’s name or recognizing a close friend. While implicit processing requires few cognitive resources, a person relying on it can miss important details that otherwise might have been observed and retained using explicit processing.   


Memory involves more than just the transfer of sensory information into long-term storage. Unlike information recorded and placed in a file cabinet, the encoding of information and experiences is a constructive process where knowledge is created and re-created based on prior learning and new information. Previous learning and the context within which it occurred influence the way the brain makes sense of material. Facts and beliefs act as filters that help the brain attend to stimuli that has meaning while it disregards material that does not.

Existing knowledge directly affects the way people perceive information, recognize patterns, and assign meaning.9 Rather than remembering what actually occurred, the brain remembers what it was doing at the time the information was encoded. Memories also are re-created based on new knowledge, with recent information often displacing older memories. This can be problematic for investigators, especially when people are unable to recall the source or reality of the new information.   


A person’s ability to recall information depends on how well the individual originally learned the material. Most stimuli receive only incidental attention and are processed to a shallow level of memory; however, other stimuli are given deliberate attention and elaboration, thus enhancing their transfer to long-term memory.10 Elaboration occurs when the brain draws relationships between data already stored in memory and the items being contemplated. The depth-of-processing framework suggests that deeper refining leads to better memory. Accordingly, information that is processed deliberately and consciously has a better chance of being encoded and remembered than data that is produced superficially.

A psychologist and memory researcher demonstrated how these shortcomings affect the recall of victims, witnesses, and suspects. As part of this study, participants watched a series of slides about a burglary. A key element in the crime was a screwdriver used by the suspect. The subjects read a written account of the same crime that differed slightly from the original version. It referred to a hammer instead of a screwdriver being used to commit the burglary. On a subsequent test of memory, 60 percent of the participants referred to a hammer, rather than a screwdriver, as being used in the crime.11 The subjects’ poor recall was exaggerated by the fact that they demonstrated as much confidence in their fabricated accounts of the crime as they did in their more accurate memories of the original event.


In an ideal world, victims and witnesses would evaluate, process, and remember events in objective, impartial, and error-free ways. The human mind possesses astonishing capabilities; however, it is also subject to systematic biases and errors. These include biases of confirmation, attention, confidence, order, and emotion

Confirmation Bias

Individuals can selectively gather information in ways that confirm and reinforce their existing beliefs, while ignoring data that contradicts or challenges those ideas. In one experiment, subjects were asked to rate the guilt or innocence of interviewees based on their responses to guilt-presumptive questions or to neutral questions. The researchers found that the mock answers provided confirmation for the interviewers’ initial expectations—whether they believed the interviewees were guilty or innocent.12 Those who were asked guilt-presumptive questions—even if they were innocent—were rated as more nervous and defensive and less believable than those who were asked neutral questions. This suggests that guilt-presumptive questions cause suspects to act deceptively—particularly if they are innocent—confirming the interrogator’s existing beliefs.

Attention Illusion

In a well-known experiment, students were asked to count the number of times a basketball was passed between players wearing white uniforms while ignoring players wearing black uniforms.13 Halfway through the one-minute video, a female student wearing a gorilla suit walked into the scene, spent about nine seconds thumping her chest, then exited. When questioned about what they saw, half the students failed to notice the gorilla.

This study has been repeated numerous times, under different conditions, with diverse audiences, in multiple countries, always with the same results. Researchers refer to this phenomenon of missing a highly salient object precisely where one is looking as the illusion of attention—a form of overconfidence that causes people to miss more of the world around them.

Confidence Levels

Victims and witnesses often report high levels of confidence in their memories despite overwhelming evidence to the contrary. One area where people report particularly high levels of confidence is termed “flashbulb” memories. Most people can produce vivid recollections of emotional events in their lives; however, studies have found no relationship between a person’s level of confidence and the accuracy of their memories. In one study 44 students were surveyed the day after the space shuttle Challenger disaster and again 3 years later to test the accuracy of these flashbulb memories.14 Less than 7 percent of second reports matched all the details of the original report. Fifty percent of second reports were incorrect in two-thirds of assertions, while 25 percent were wrong in every major detail.  

Information Order

A number of factors—type, vividness, saliency, and utility—affect how people remember and recall information. One aspect that influences the type of information individuals remember is the order of the material itself. The brain’s ability to process and interpret information is limited, so the order of events affects what people remember.15 All things being equal, individuals seem to pay closer attention to their first and last encounters with people, places, and things. Simply put, the laws of primacy and recency mean that people tend to remember their first (law of primacy) and their last (law of recency) impressions because information gathered at the start of an event influences cognitive set, while data obtained at the end of an encounter is the most fresh in the observer’s memory.

Emotional Impact

The brain evolved with the primary function of keeping its host alive. As a result, data with the potential to influence a person’s well-being or survival are provided primacy over other types of input. The autonomic nervous system constantly is on alert for anything that might impact safety. At the first sign of danger, the limbic system can inhibit the functioning of the prefrontal cortex—the area of the brain associated with reasoning, planning, and decision making.16 Strong emotions can amplify the ability to remember the affective aspects of an event. They also can shut down the conscious processing of information and make it more difficult to remember nonemotional details of an incident.  

Despite these well-documented biases, many people remain confident in their abilities to notice, process, and recall information and events. One area where this has proven especially problematic is the eyewitness identification of alleged criminal suspects. Jury members often view a witness’ level of confidence in the ability to recall events and to recognize an assailant in the courtroom as reliable indicators of a person’s memory. However, the growing number of eyewitness cases overturned by recent advances in DNA provides strong evidence of just how unreliable eyewitness testimony can be.17 While many witnesses provide reliable, accurate, and confident testimony, mistaken identification by confident eyewitnesses remains the main cause of wrongful convictions. In fact, it is estimated that eyewitness testimony is responsible for over 75 percent of wrongful convictions eventually overturned by DNA evidence.18


Memory is a constructive and reconstructive process influenced by numerous factors—context, emotional state, order of information, and types of questions asked. Overly suggestive questions can alter a person’s recollection, making it difficult to separate original memories from distorted reconstructions.19

The malleability of memory and the importance of properly worded, open-ended questions are illustrated in the changing account of President Kennedy’s assassination provided by the “Lady in Red” (Jean Hill), who was standing next to the president’s motorcade when he was shot. Less than an hour after the incident, when first questioned by a TV reporter who asked Ms. Hill if she saw the person who fired the fatal shot, she replied, “No…I didn’t see any person fire the weapon.” Ninety minutes later during a second interview, Hill told reporters that although she did not see anyone shooting, she heard “four to six shots.” Later that same day, while being questioned along with other eyewitnesses at the Dallas County Sheriff’s Office, Jean Hill told investigators that she noticed rifles being drawn and possibly fired by “men in plain clothes.” She also described a man running toward “the monument.” She said that she followed him for a short distance before returning to the scene of the shooting.

In March 1964, approximately four months after the shooting, FBI agents interviewed Hill, and she changed her story again. This time she said that she heard four to six shots that she believed came from just west of the Texas School Book Depository. She also described a man sporting a raincoat, wearing a hat, running toward the railroad tracks. She said she followed him, adding that a motorcyclist nearly hit her in the process. Finally, the following day when the Warren Commission debriefed Ms. Hill, she said that she was “…too stunned to move, so I didn't get down. I just stood there and gawked around.” This contradicted her earlier statements and two photographs of her sitting down just after the shooting. Toward the end of the Warren Commission inquiry, Ms. Hill admitted that her memories of the event may have beenb“…colored by what I heard.”20 The Hill case illustrates the constructive and reconstructive nature of memory and the importance of how interviewers ask their questions.


Interviewers should avoid leading or suggestive questions whenever possible. They should use open-ended questions that require detailed responses and that do not offer built-in “default” answers. Closed-ended questions restrict the amount of information provided, whereas open-ended questions offer no assumptions and place no restrictions on answers. Investigators may use follow-up and closed-ended questions after an interviewee has provided as many details as possible through open-ended questioning.

Investigators must remind themselves of the pervasive effects of cognitive biases and memory errors. The brain’s limited attention and processing capacities, the strong influence of emotions, and the order of events affect what people notice, interpret, and remember. This is exacerbated by the illusion of attention, confirmation bias, and overconfidence. By better understanding these shortcomings and remaining vigilant, investigators can improve the likelihood of obtaining ground truth.

For additional information, Dr. Fitch can be reached at


1 Deanna Kuhn, “A Developmental Model of Critical Thinking,” Educational Researcher 28, no. 2 (March 1999): 16-25.

2 Daniel L. Schacter, The Seven Sins of Memory: How the Mind Forgets and Remembers (New York, NY: Houghton Mifflin, 2001).

3 Richard C. Atkinson and Richard M. Shiffrin, “Human Memory: A Proposed System and its Control Processes,” in The Psychology of Learning and Motivation: Advances in Research and Theory, vol. 2, ed. K.W. Spence and J.T. Spence (New York, NY: Academic Press, 1968), 89-195. 

4 Roger H. Bruning, Gregory J. Schraw, Monica M. Norby, and Royce R. Ronning, Cognitive Psychology and Instruction, 4th ed. (Upper Saddle River, NJ: Prentice Hall, 2003).

5 Richard M. Shiffrin and Richard C. Atkinson, “Storage and Retrieval Processes in Long-Term Memory,” Psychological Review 76, no. 2 (March 1969): 179-193.

6 David G. Myers, Psychology, 9th ed. (New York, NY: Worth, 2009).

7 Mark H. Ashcraft, Cognition, 3rd ed. (Upper Saddle River, NJ: Prentice Hall, 2001).

8 Kevin B. MacDonald, “Effortful Control, Explicit Processing, and the Regulation of Human Evolved Predispositions,” Psychological Review 115, no. 4 (2008): 1012-1031.

9 Bruning, Schraw, Norby, and Ronning, Cognitive Psychology and Instruction, 6.

10 Fergus I. M. Craik and Robert S. Lockhart, “Levels of Processing: A Framework for Memory Research,” Journal of Verbal Learning and Verbal Behavior 11 (1972): 671-684.

11 Elizabeth F. Loftus and Hunter G. Hoffman, “Misinformation and Memory: The Creation of New Memories,” Journal of Experimental Psychology: General 118, no. 1 (March 1989): 100-104. 

12 Carole Hill, Amina Memon, and Peter McGeorge, “The Role of Confirmation Bias in Suspect Interviews: A Systematic Evaluation,” Legal and Criminological Psychology 13, no. 2 (September 2008): 357-371.

13 Christopher Chabris and Daniel Simons, The Invisible Gorilla: How Our Intuitions Deceive Us (New York, NY: Random House, 2009).

14 Ulric Neisser and Nicole Harsch, “Phantom Flashbulbs: False Recollections of Hearing the News About Challenger,” in Affect and Accuracy in Recall: Studies of “Flashbulb” Memories, ed. Eugene Winograd and Ulric Neisser (New York, NY: Cambridge University Press, 1992), 9-31.

15 Elliott Aronson, Timothy D. Wilson, and Robin M. Akert, Social Psychology, 2nd ed. (New York, NY: Longman Publishing Group, 1997).

16 Daniel Goleman, Emotional Intelligence: Why It Can Matter More Than IQ (New York, NY: Random House, 1995).

17 Steve McGonigle and Jennifer Emily, “18 Dallas County Cases Overturned by DNA Relied on Heavily Eyewitness Testimony,” Dallas Morning News, October 12, 2008, (accessed December 1, 2014).

18 Innocence Project, “Eyewitness Misidentification,” Eyewitness-Misidentification.php (accessed September 2, 2014).

19 Elizabeth F. Loftus, “Planting Misinformation in the Human Mind: A 30-Year Investigation of the Malleability of Memory,” Learning and Memory 12 (2005): 361-366.

20 Peter R. Whitmey, “Jean Hill – The Lady in Red,” (accessed September 2, 2014).