Accurate eyewitness identification requires that a witness to a crime correctly sense, perceive, and remember objects and events that occurred and recall them later. The veracity of the witness’ identification thus depends on the limits of sensation, perception, and memory. Recent scientific studies have yielded great advances in our understanding of how vision and memory work. This chapter provides a brief overview of current knowledge, identifies areas in which vision and memory are imperfect, and describes implications for the accuracy of eyewitness identification. These implications, in turn, have guided much of the applied research on this topic (see Chapter 5) and provide a general framework for the recommendations made herein (see Chapter 6).
VISION AND MEMORY IN CONTEXT
This chapter begins by offering a concrete example to place the body of basic scientific research on vision and memory in context so as to better communicate its relevance to eyewitness identification. In the sections that follow the example, the different functional steps of the sequence (highlighted in italics) are dissected in some detail, with special reference to its limitations and the ways in which it may fail to deliver accurate eyewitness identification.
While returning home late, you hear a muffled scream from around the street corner. Seconds later, you come face-to-face with a man turning the corner and moving swiftly past you. Instantaneously, properties of the
scene are conveyed to you through patterns of light cast on the backs of your eyes and sensed by photoreceptors in your retina. Only a fraction of the information sensed is selected for further processing; in this case you focus your attention on certain features of the man’s face. Those features are integrated and interpreted to yield a coherent percept of the man. As you round the corner, you perceive, through an identical process, the victim slumped lifelessly against a wall. You quickly grasp the meaning of these perceptual experiences, and they immediately elicit both cognitive and visceral components (e.g., increased heart rate) of fear and anxiety. Your percepts are initially encoded in short-term working memory, where content is limited and labile. Your elevated level of arousal may cause interference and some loss of content, but with time and recognition of the importance of the experience, your percepts are consolidated into long-term memory. Long-term memories are maintained in storage but subject to ongoing updates and modifications resulting from new experiences and perhaps distortions caused by sustained levels of stress.
At a later date, you are asked to look at a police lineup that includes a suspect apprehended near the crime scene. Visual features of the men in the lineup are sensed, selectively attended, and perceived, using the same visual processes engaged on the night in question. Some of these features—the high brow and sharp cheekbones of one man in the lineup—elicit retrieval of memories of your visual experiences on the night of the crime. The simultaneously perceived and retrieved experiences are implicitly compared, leading to a cycle of greater visual scrutiny of the man in front of you and retrieval of additional details of the original percept. The context of the lineup procedure, the sight of the man, and the retrieved memories trigger latent emotions and anxiety, which may interfere with your comparison of percept and memory. Eventually, the comparison reaches your internal criterion for identification: You decide, with an implicit level of certainty, that your current visual percept and the percept from the night of the crime were caused by the same external source (the man now in front of you), and you assert that you have identified the person you witnessed at the crime scene.
Functional Processes of Vision
To understand the contributions and limitations of vision to eyewitness identification, it is useful to consider the workings of three functional stages of visual processing—sensation, attention, and perception—bearing in mind that they comprise highly interdependent elements of a continuous operation. Sensation is the initial process of detecting light and extracting basic image features. Sensations themselves are evanescent, and only a small
fraction of what is sensed is actually perceived. Attention is the process by which information sensed by the visual system is selected for further processing. Perception is the process by which attended visual information is integrated, linked to environmental cause, made coherent, and categorized through the assignment of meaning, utility, value, and emotional valence. In addition, memories and emotions resulting from prior experiences with the world can influence all stages of visual processing and thus define a thread that weaves throughout the following discussions.
All of the functional processes of vision are beset by noise, which affects the quality and types of information accessible from the visual environment, and bears heavily on the validity of eyewitness identification. Before considering the processes of sensation, attention, and perception in greater detail, consideration is given to the concept of noise in visual processing and to ways of interpreting its impact on visual experience.
The Fundamental Role of Noise
Vision is usefully understood as the process of detecting informative signals about the external world and using those signals to recognize objects, make decisions, and guide behavior. As with any signal detection, there are occasionally factors that lead to uncertainty on the part of the observer about whether a particular signal is present. These factors are generically termed noise, following the definition used in electronic signal transmission, in which noise refers to random or irrelevant elements that interfere with detection of coherent and informative signals. In vision, noise comes from a variety of sources, some associated with the structure of the visual environment (e.g., occluding surfaces, glare, shadows), some inherent to the optical and neuronal processes involved (e.g., scattering of light in the eye), some reflecting sensory content not relevant to the observer’s goals (e.g., a distracting sign or a loud sound), and some originating with incorrect expectations derived from memory. Consider, for example, the seemingly simple problem of detecting a green light while waiting at a traffic signal. In this case, your ability to “see” the green light may be compromised by glare or dust on your windshield, by poor visual acuity, by your eyes having been aimed instead at the driver of the adjacent car, by the presence of other (irrelevant) colored lights in your field of view (e.g., a traffic signal at a different intersection or the lights of a nearby restaurant), by a cell phone conversation, or by the news on the car radio. The significance of this view for eyewitness identification is profound, as it helps us to realize that the accuracy of information about the environment—the face
of a criminal, for example—gained through vision is necessarily, and often sharply, limited by noise.1
The fact that vision is noise-limited suggests a familiar statistical framework—signal detection theory—for assessing and understanding the effects of noise on visual perception and recognition ability.2 Signal detection theory has long been successfully applied to analogous problems in electronic signal reception.3 To illustrate these principles as applied to sensory processing, consider the problem of detecting a vibrating cell phone in your pocket. Anyone who has operated a cell phone in vibrate mode will be familiar with two types of signal detection errors: (1) the occasional sense that the phone is vibrating in your pocket, only to discover that it is not, and, conversely, (2) the phone call that is sometimes missed because you attribute the vibration to some other cause. Signal, in this example, is a subtle tactile stimulus resulting from an incoming phone call. Noise, in this example, is all of the other things in your environment that may also lead to subtle tactile stimulation, such as vibration of your car seat, a shift of keys in your pocket, or the touch of another person.
Signal detection theory posits that there are three main factors that determine whether a signal will be detected: (1) the distribution of stimuli (e.g., the variety of stimulus magnitudes) that reflect noise only, (2) the distribution of stimuli that reflect signal, and (3) the observer’s criterion for “deciding” that a specific stimulus resulted from noise sources or signal. An important factor for the fidelity of signal detection is the degree to which noise and signal distributions overlap with one another. In the case of the vibrating cell phone, if the distributions of tactile stimuli resulting from noise and signal overlap, as is often the case, then there will always be some cases in which you believe the phone is vibrating when it is not (noise stimuli attributed to signal source), and there will be some cases in which the phone is vibrating and you miss the call (signal stimuli attributed to noise source).
The third factor that influences signal detection in the presence of noise is the observer’s decision criterion, which is simply the value (e.g., stimulus amplitude) above which a stimulus is attributed to signal, and below which a stimulus is attributed to noise. In the same sense that your car radio is programmed to “decide” (and allow you to hear) when informative patterns of electromagnetic radiation (signal) are sufficiently different from random fluctuations (noise), an observer adopts a criterion for deciding whether a
1W. S. Geisler, “Sequential Ideal-Observer Analysis of Visual Discriminations,” Psychological Review 96(2): 267–314 (1989).
2D. M. Green and J. A. Swets, Signal Detection Theory and Psychophysics (New York: Wiley, 1966).
3W. W. Peterson, T. G. Birdsall, and W. C. Fox, “The Theory of Signal Detectability,” Proceedings of the IRE Professional Group on Information Theory 4(4): 171–212 (1954).
stimulus is caused by a signal or is simply a manifestation of noise. This criterion reflects the level of precision acceptable for the observer’s needs, given uncertainty about whether a given stimulus reflects a real signal.
In practice, the criterion4 used is determined by a host of factors unique to the circumstances, including psychological and social demands and behavioral goals. These factors collectively determine the relative “costs” of incorrect attributions of signal as noise (“misses”) and of noise as signal (“false alarms”).
If an individual places high value on not missing a phone call, then she or he will adopt a very liberal criterion, in which all stimuli reflecting real incoming calls (signal) are successfully detected, but many noise stimuli (e.g., shifting keys in a pocket) are erroneously (and frustratingly) believed to be incoming calls. By contrast, if an individual places little value on detecting incoming phone calls, she or he will adopt a conservative criterion, in which many calls are missed and noise stimuli rarely elicit an effort to answer the phone, which may be of value to the individual who wishes to avoid distraction.
The example of the signal detection logic used for the vibrating cell phone applies similarly to all aspects of visual perceptual experience, including the conditions of witnessing criminal events. The uncertainty about visual events caused by manifold sources of noise will inevitably lead to inaccurate visual perceptual experiences, which result from conditions in which an observer fails to detect a critically informative stimulus as “real” (attributing the stimulus instead to a source of noise) or confidently perceives a noise stimulus to have originated from an informative source. The latter instance is problematic because it increases the likelihood that observers will unwittingly “construct,” on the basis of expectations derived from memory and situational context, perceptual experiences to account for noise erroneously interpreted as signal.
What follows from this consideration of uncertainty and decision criteria for visual perception is that the actual impact of factors that limit the amount of visual information available to an eyewitness (factors considered in more detail below) will depend on the criterion adopted. The criterion may reflect the values and prejudices of the eyewitness, his or her motivational and emotional state, and a variety of behavioral goals. In principle, the observer’s criterion can be altered by instruction or incentives, but it is important to note that the criterion held by an observer witnessing a crime scene cannot be anticipated, nor can it be altered after the fact. It is an “estimator variable,” which simply needs to be recognized and understood when evaluating eyewitness reports. By contrast, the decision criterion held
4The criterion is sometimes referred to as bias.
In the following discussions of sensation, attention, and perception, the various means and conditions under which many different types of noise introduce uncertainty in visual signal detection (and thus fundamentally limit the accuracy of eyewitness identification) are addressed.
When an observer views an object of any sort (such as a person) or events involving the object (a criminal act), patterns of light reflected from the environment are focused by the lens at the front of the eye and projected onto the back surface of the eye (the retina) to form the retinal image. Light in the image is initially “sensed” by the activation of photoreceptors, and early stages of sensory processing function to detect spatial and temporal contrast along a number of dimensions, including intensity and wavelength of light.6 These contrast measurements are integrated by subsequent processing stages in the brain to yield representations of basic image features, or primitives, such as oriented image contours.7
Several sources of noise, or factors that limit the ratio of signal to noise, can restrict the visual information accessible to these early sensory processes. Some factors are inherent to the visual system and largely uncontrollable (e.g., the scattering of light by the fluid and tissues of the eye) and can be exacerbated by common observer-specific visual deficits (e.g., myopia, poor contrast sensitivity, or color blindness). Others factors are dependent on viewing conditions (e.g., the effects of viewing time and level of illumination).8 Both of these types of factors predictably influence the quantity of information—the visual signal strength—that a viewer gains from a visual scene, and thus the degree to which the perceptual experi-
5L. Mickes, H. D. Flowe, and J. T. Wixted, “Receiver Operating Characteristic Analysis of Eyewitness Memory: Comparing the Diagnostic Accuracy of Simultaneous and Sequential Lineups,” Journal of Experimental Psychology: Applied 18(4): 361–376 (2012).
6M. Meister and M. Tessier-Lavigne, “Low-level Visual Processing: The Retina,” in Principles of Neuroscience, 5th Edition, ed. E. Kandel, J. H. Schwartz, T. M. Jessell, S. A. Siegelbaum, and A. J. Hudspeth (New York: McGraw-Hill Professional, 2012), 577–601.
7C. D. Gilbert, “Intermediate-level Visual Processing and Visual Primitives,” in Principles of Neuroscience, 5th Edition, ed. E. Kandel, J. H. Schwartz, T. M. Jessell, S. A. Siegelbaum, and A. J. Hudspeth (New York: McGraw-Hill Professional, 2012), 602–620.
8D. G. Pelli, “Uncertainty Explains Many Aspects of Visual Contrast Detection and Discrimination” Journal of the Optical Society of America A2(9): 1508–32 (1985). D. G. Pelli, “The Quantum Efficiency of Vision,” in Vision: Coding and Efficiency, ed. C. Blakemore (Cambridge: Cambridge University Press, 1990), 3–24. G. Sperling, “The Information Available in Brief Visual Presentations,” Psychological Monographs: General and Applied 74(11, Whole No. 498): 1–29 (1960).
ence can accurately reflect the properties of the external world.9 At the extreme, short viewing times and low levels of illumination simply reduce the number of correlated photons reaching the retina to the point where they scarcely exceed photon noise, and uncertainty is very high.10 At slightly longer viewing times and greater illumination levels, signal-to-noise levels improve, but there may remain marked limits on visual sensitivity. Visual acuity, for example, which is a measure of the ability to resolve the fine spatial details of a visual pattern, is known to decline significantly with decreases in illumination.11
Signal-to-noise loss can depend on the direction of the observer’s gaze. Visual acuity is highest at the observer’s center of gaze. The center is the part of your visual system that is used for fine sensing, such as reading or scrutinizing faces in a social context. Acuity drops off markedly with angular distance from this center, such that the quality and quantity of information sensed a mere 10 degrees from center are far less than what is available at the center of gaze.12
Under unrestricted viewing conditions, the movements of the eyes largely overcome the effects of gaze direction. However, under the viewing conditions associated with a typical crime, this source of noise may place severe limitations on the ability of the observer to sense key pieces of information that are not present at the center of gaze. To appreciate the impact of these limitations, consider that patients with macular degeneration are effectively blinded in the region of the visual field possessing highest acuity, and must rely instead on the much-reduced quality of visual information gained from the peripheral visual field. To compensate for this clinical loss, images and text must be greatly magnified to enable comprehension—an option that is clearly not available to an eyewitness.
Light falling on all parts of the retina is available to be sensed—and must be sensed for it to be available for further processing—but only a
9G. Sperling, “A Signal-to-Noise Theory of the Effects of Luminance on Picture Memory: Comment on Loftus,” Journal of Experimental Psychology: General 115(2): 189–192 (1986).
10S. Hecht, S. Schlaer, and M. H. Pirenne, “Energy, Quanta, and Vision,” Journal of General Physiology 25(6): 819–840 (1942).
11P. W. Cobb, “The Influence of Illumination of the Eye on Visual Acuity,” American Journal of Physiology 29: 76–99 (1911). S. Hecht, “A Quantitative Basis for the Relation Between Visual Acuity and Illumination,” Proceedings of the National Academy of Sciences 13: 569–574 (1927). S. Shlaer, “The Relation Between Visual Acuity and Illumination,” Journal of General Physiology 21 (2): 165–188 (1937).
12H. Strasburger, I. Rentschler, and M. Jüttner, “Peripheral Vision and Pattern Recognition: A Review,” Journal of Vision 11(5):13, 1–82 (2011).
small fraction of the information sensed reaches awareness or is used by the observer for recognition, action, or storage in memory. This limited access to visual sensory information is a product of selective attention.13 Attention is an active process that can be directed by external factors—visual attributes with high salience, such as a bright light or an unfamiliar object—or by internal control.14 If you are searching for a coffee cup, for example, you may explicitly direct your attention to the table where it was last seen. Attention can be directed to different types of image content, including specific locations in space,15 specific image features (such as a specific color),16 or to specific objects (such as the coffee cup).17
Attended image content is transiently enhanced to increase the fidelity of visual experience.18 Attention interacts with sensory processing, for example, by selectively enhancing contrast19 and potentially overcoming low signal-to-noise levels resulting from limited viewing time or illumination.20 The effects of attention on contrast enhancement can be potentiated further when attention is commanded by emotionally laden stimuli.21 Image con-
13W. James, Principles of Psychology (New York: Henry Holt, 1890); H. Pashler, J. Johnston, and E. Ruthruff, “Attention and Performance,” Annual Review of Psychology 52: 629–651 (2001).
14M. I. Posner, “Orienting of Attention,” Quarterly Journal of Experimental Psychology 32: 3–25 (1980).
16A. F. Rossi and M. A. Paradiso, “Feature-specific Effects of Selective Visual Attention,” Vision Research 35(5): 621–634 (1995).
17J. Duncan, “Selective Attention and the Organization of Visual Information,” Journal of Experimental Psychology: General 113(4): 501–517 (1984).
18H. Pashler, J. Johnston, and E. Ruthruff, “Attention and Performance,” Annual Review of Psychology 52: 629–651 (2001).
19M. Carrasco et al., “Attention Alters Appearance” Nature Neuroscience 7: 308–313 (2004).
20M. I. Posner, C. R. Snyder, and B. J. Davidson, “Attention and the Detection of Signals,” Journal of Experimental Psychology 109(2): 160–174 (1980). M. Carrasco and B. McElree, “Covert Attention Accelerates the Rate of Visual Information Processing,” Proceedings of the National Academies of Science 98(9): 5363–5367 (2001). Y. Yeshurun and M. Carrasco, “Attention Improves or Impairs Visual Performance by Enhancing Spatial Resolution,” Nature 396, 72–75 (1998). M. Carrasco et al., “Covert Attention Increases Spatial Resolution with or without Masks: Support for Signal Enhancement,” Journal of Vision 2(6): 467–79 (2002). E. Blaser et al., “Measuring the Amplification of Attention,” Proceedings of the National Academies of Science 96(20): 11681–11686 (1999). K. Anton-Erxleben and M. Carrasco, “Attentional Enhancement of Spatial Resolution: Linking Behavioural and Neurophysiological Evidence,” Nature Reviews Neuroscience 14(3):188–200 (2013). J. W. Couperus and G. R. Mangun, “Signal Enhancement and Suppression During Visual-Spatial Selective Attention,” Brain Research 1359:155–177 (2010).
21E. A. Phelps, S. Ling, and M. Carrasco, “Emotion Facilitates Perception and Potentiates: The Perceptual Benefits of Attention,” Psychological Science 17(4): 292 (2006).
tent not falling within the focus of attention is processed with less fidelity.22 In some cases, unattended content is effectively invisible: It does not reach awareness, it is not perceived, and it is not available for use in guiding decisions or actions, or for storage in memory.23
Different pieces of visual information compete for selection,24 as their attributes of physical salience, location in space, novelty, and relevance to the observer’s needs and behavioral goals are always changing.25 The outcome of the competition is highly susceptible to noise (in this instance, noise is defined as uncontrolled factors that bias the focus of attention and create uncertainty about the content of a visual scene), because the informational content of the visual image vastly exceeds what can be attended at any point in time. The implications of such noise for eyewitness identification are profound. An observer must “select” what to attend to, often within a short window of time, without advance warning, in the presence of many novel objects and events, and under such confounding influences as anxiety and fear.
The signal detection framework is readily adaptable to the problem of noise in visual attention and provides some insights into the limits of attentional selection in the presence of noise.26 In essence, this signal detection approach quantifies the extent to which multiple items competing with one another for attention affect attentional enhancement for any one of the items.27 Reductions in efficiency are common under such noise conditions. Indeed, sensitivity to unattended items can be markedly reduced under conditions of high “perceptual load,” in which there are many objects si-
22Posner, Snyder, and Davidson, “Attention and the Detection of Signals.” Y. Yeshurun and M Carrasco, “Attention Improves or Impairs Visual Performance by Enhancing Spatial Resolution,” Nature 396: 72–75 (November 1998).
23A. Mack and I. Rock, Inattentional Blindness (Cambridge, MA: MIT Press, 1998).
24R. Desimone and J. Duncan, “Neural Mechanism of Selective Visual Attention,” Annual Review of Neuroscience 18: 193–222 (March 1995).
25J. M. Wolfe and T. S. Horowitz, “What Attributes Guide the Deployment of Visual Attention and How Do They Do It?” Nature Reviews Neuroscience 5: 495–501 (June 2004). H. E. Egeth and S.Yantis, “Visual Attention: Control, Representation, and Time Course,” Annual Review of Psychology 48(1): 269–297 (February 1997). M. I. Posner, “Orienting in Attention,” Quarterly Journal of Experimental Psychology 32(1): 3–25 (1980). A. Treisman and G. Gelade, “A Feature Integration Theory of Attention,” Cognitive Psychology 12(1):97–136 (January 1980). L. Itti and C. Koch, “A Saliency-based Search Mechanism for Overt and Covert Shifts of Visual Attention,” Vision Research 40(10–12): 1489–1506 (June 2000).
26G. Sperling and M. J. Melchner, “The Attention Operating Characteristic: Examples from Visual Search,” Science 202(4365): 315–318 (October 1978). G. Sperling and B. A. Dosher, “Strategy and Optimization in Human Information Processing,” in Handbook of Perception and Human Performance, ed. K. Boff, L. Kaufman, and J. Thomas (New York: Wiley, 1986).
multaneously competing for attention.28 The spacing of items in the visual field also impacts visual sensitivity.29 When objects are closely spaced, their discriminability is reduced. One explanation offered for this “crowding effect” is that the spacing of visual items is smaller than the resolution of visual attention.30 The visual phenomenon of crowding suggests that a crime committed in a visually complex scene, such as a sporting event, could easily place limits on the ability of a witness to accurately perceive the facial features of a perpetrator.
A related consequence of attentional noise is that competing interests can readily hijack the attentional focus. The technique of misdirection—one of the original mainstays of performance magic—directs attention to uninformative image content and exploits the invisibility of unattended features.31 The well-studied inattentional blindness effect is another example of this phenomenon, in which attention that is pre-directed to one behaviorally significant property of a visual scene precludes awareness of other features that also may be important.32 (For a dramatic demonstration of this effect, produced by Simons and Chabris,33 see http://tinyurl.com/inattentional-blindness.)
Inattentional blindness effects translate well to real-world interactions between people. An individual can be surprisingly unaware of surreptitious changes to the physical appearance of another person while engaged in conversation.34 One demonstration of this phenomenon involved two strangers (experimenter and pedestrian) in a brief face-to-face conversation on a sidewalk. At some point in the conversation an opaque door was carried between the two individuals, and another person with different appearance, clothing, and voice quickly replaced the experimenter. More than half of
28N. Lavie, “Perceptual Load as a Necessary Condition for Selective Attention,” Journal of Experimental Psychology: Human Perception and Performance 21(3): 451–468 (June 1995). J. W. Couperus, “Perceptual Load Influences Selective Attention Across Development,” Developmental Psychology 47(5):1431–1439 (September 2011).
29D. M. Levi, “Crowding—An Essential Bottleneck for Object Recognition: A Mini-review,” Vision Research 48: 635–654 (2008).
30J. Intriligator and P. Cavanagh, “The Spatial Resolution of Visual Attention,” Cognitive Psychology 43: 171–216 (2001).
31G. Kuhn et al., “Misdirection in Magic: Implications for the Relationship Between Eye Gaze and Attention,” Visual Cognition 16(2–3): 391–405 (2008). S. L. Macknik, S. Martinez-Conde, and S. Blakeslee, Sleights of Mind: What the Neuroscience of Magic Reveals About Our Everyday Deceptions (New York: Henry Holt and Co., 2010).
32A. Mack and I. Rock, Inattentional Blindness (Cambridge, MA: MIT Press, 1998). U. Neisser and R. Becklen, “Selective Looking: Attending to Visually Specified Events,” Cognitive Psychology 7(4): 480–494 (October 1975). D. Simons, “Attentional Capture and Inattentional Blindness,” Trends in Cognitive Sciences 4(4): 147–155 (April 2000).
33D. J. Simons and C. F. Chabris, “Gorillas in Our Midst: Sustained Inattentional Blindness for Dynamic Events,” Perception 28: 1059–1074 (1999).
34D. J. Simons and D. T. Levin, “Failure to Detect Changes to People During a Real-World Interaction,” Psychonomic Bulletin and Review 5(4): 644–649 (1998).
the participants (pedestrians) failed to notice that their conversation partner had changed. This finding suggests that naturally occurring events that briefly divert attention have the potential to markedly impair the accuracy of eyewitness identifications.
Attentional hijacking is particularly characteristic of stimuli that elicit strong emotional responses, such as fear and arousal.35 Visual stimuli that trigger fear responses act as powerful external cues that command attention.36 While this potentiates sensitivity to those stimuli, at the considerable expense of sensitivity to others, it is often the case that the attended emotional stimuli are not the ones with relevant informational content.37The so-called weapon focus is a real-world case in point for eyewitness identification, in which attention is compellingly drawn to emotionally laden stimuli, such as a gun or a knife, at the expense of acquiring greater visual information about the face of the perpetrator (see also discussion of weapon focus in Chapter 5).38 (One might argue that this is an adaptation that benefits immediate action or engagement with a threatening stimulus, but is surely detrimental to one’s efforts to bear witness.)
Visual perception is the conscious functional result of efforts to identify the environmental causes of the pattern of light cast onto the back of the eye.39 Perception does not reflect the sensory world passively, as camera film detects patterns of light. On the contrary, visual perception is constructive
35C. H. Hansen and R. D. Hansen, “Finding the Face in the Crowd: An Anger Superiority Effect,” Journal of Personality and Social Psychology 54: 917–924 (1988). E. Fox et al., “Facial Expressions of Emotion: Are Angry Faces Detected More Efficiently?” Cognition and Emotion 14(1): 61–92 (2000). R. Compton, “The Interface Between Emotion and Attention: A Review of Evidence from Psychology and Neuroscience,” Behavioral and Cognitive Neuroscience Reviews 2(2): 115–129 (2003). R. L. Bannerman, E. V. Temminck, and A. Sahraie, “Emotional Stimuli Capture Spatial Attention But Do Not Modulate Spatial Memory,” Vision Research 65: 12–20 (15 July 2012).
36J. A. Easterbrook, “The Effects of Emotion on Cue Utilization and the Organization of Behavior,” Psychological Review 66(3): 183–201 (1959).
37E. Ferneyhough et al., “Anxiety Modulates the Effects of Emotion and Attention on Early Vision,” Cognition and Emotion 27(1): 166–176 (2013). G. Pourtois and P. Vuilleumier, “Dynamics of Emotional Effects on Spatial Attention in the Human Visual Cortex,” Progress in Brain Research 156: 67–91 (2006).
38T. Kramer, R. Buckhout, and P. Eugenio, “Weapon Focus, Arousal, and Eyewitness Memory: Attention Must Be Paid,” Law and Human Behavior 14(2): 167–184 (1990). R. S. Truelove, “Do Weapons Automatically Capture Attention,” Applied Cognitive Psychology 20(7): 871–893 (2006). E. F. Loftus, G. R. Loftus, and J. Messo, “Some Facts About ‘Weapon Focus’,” Law and Human Behavior 11(1): 55–62 (1987).
39W. James, Principles of Psychology (New York: Henry Holt, 1890). S. Harnad, ed., Categorical Perception: The Groundwork of Cognition (New York: Cambridge University Press, 1987). T. D. Albright, “Perceiving,” Daedalus (in press).
and entails (1) integrating and segmenting attended attributes of the visual image into objects, (2) complementing and interpreting the product with expectations derived from memory of prior experiences with the world, and (3) assigning meaning and emotional valence by reference to prior knowledge of function and value.40 All of these perceptual processes are affected by noise. Because the things perceived are the things we place into memory, perceptual noise can dramatically limit the accuracy of eyewitness identification.
The process of feature integration and interpretation may be distorted by images of an object unique to a specific angle of view.41 The retinal pattern generated by a face viewed directly from the front differs considerably—with changes in aspect ratio and relative placement of facial features—from that generated by a face viewed from an oblique side angle. Viewing a face from an angle above or below center (as might be the case if the criminal were standing over you, or below you on the stairs) also yields retinal distortions of facial features. In this case, the distortions prominently mimick facial gestures of smiling versus frowning, and perhaps cause incorrect inferences about the emotional state of the person observed and his or her intentions and motivations. (This distortion is the basis for the Japanese Noh Theatre mask effect, in which a rigid mask tilted forward leads to the appearance of a smile and backward leads to the appearance of a frown—an effect you can simulate by simply looking into the mirror and tilting your face up or down.)42
Viewing conditions can also affect the perception of face, gender, and age.43 Investigators found that faces that were physically identical—and particularly those bordering on androgyny—were perceived as unambiguously male or female depending on where they appeared in the observer’s visual field. The spatial patterning of these effects was distinctive and stable for each observer. Perceptual distortions of this sort are a source of noise that may have important implications for the accuracy of eyewitness identification.
Perceptual distortions also may be introduced through memory recall.
40C. D. Gilbert, “The Constructive Nature of Visual Processing,” in Principles of Neuroscience, 5th Edition, ed. E. Kandel, J. H. Schwartz, T. M. Jessell, S. A. Siegelbaum, and A. J. Hudspeth (New York: McGraw-Hill Professional, 2012). T. D. Albright, “On the Perception of Probable Things: Neural Substrates of Associative Memory, Imagery, and Perception,” Neuron 74 (2): 227–245 (2012).
41W. G. Hayward and P. Williams, “Viewpoint Dependence and Object Discriminability,” Psychological Science 11(1): 7–12 (2000).
42M. J. Lyons et al., “The Noh Mask Effect: Vertical Viewpoint Dependence on Facial Expression Perception,” Proceedings of the Royal Society B: Biological Sciences 267(1459): 2239–2245 (2000).
43A. Afraz, M. Vaziri-Pashkam, and P. Cavanagh, “Spatial Heterogeneity in the Perception of Face and Form Attributes,” Current Biology 20(23): 2112–2116 (2010).
The way an observer experiences a visual scene—the setting, the people, and the actions associated with a crime —is commonly influenced as much by expectations from prior experience with the world as it is by the precise patterns of light cast upon the retina. There are good reasons why this is true. As noted above, the sensory input (the pattern of light received) is often noisy, incomplete, and ambiguous, and memories of what is likely to be out there, given the context, are called on to fill in the blanks, reconcile ambiguities, and leave clear and coherent percepts.44 This perceptual completion is probabilistic.45 It is an hypothesis, and the accuracy naturally depends on the degree to which the observer’s expectations match the noisy sensory data.
What is implied is that the same mechanism that grants the certainty of perceptual experience in the face of noise and ambiguity is also capable of implicitly fabricating content that does not correspond to external reality and yet is experienced with no less certainty. Performance magic relies on this constructive nature of perceptual experience, and that nature is also the foundation for many visual illusions and forms of visual art.46 In a classic experiment that drives home the point, Bruner and Postman looked at the ability of observers to recognize “trick” playing cards.47 The trick cards were created by altering the color of a given suit (e.g., a red seven of spades). Observers were shown a series of cards with brief presentations. Some cards were trick, and the remainder normal. With astonishing frequency, observers reported that the trick cards were normal. When questioned, observers defended their reports, even after being allowed to scrutinize the trick cards, thus demonstrating that learned properties of the world are capable of sharply altering our experience and, moreover, reinforcing our convictions about what we have seen, even in the face of countermanding sensory evidence. In view of this inherent dependence of perception on prior experiences and context—and, importantly, the fact that the viewer is commonly none the wiser when perception differs from
44Albright, “On the Perception of Probable Things.”
45D. C. Knill and W. Richards, Perception as Bayesian Inference, ed. D. C. Knill and W. Richards (Cambridge: Cambridge University Press, 1996). D. Kersten “High-level Vision as Statistical Inference,” in The New Cognitive Neurosciences, 2nd Edition, ed. M. S. Gazzaniga (Cambridge: MIT Press, 1999), 353–363. D. Kersten, P. Mamassian, and A. Yuille, “Object Perception as Bayesian Inference,” Annual Review of Psychology 55: 271–304 (February 2004).
46E. H. Gombrich, Art and Illusion. A Study in the Psychology of Pictorial Representation (London: Phaidon 1960). T. D. Albright, “The Veiled Christ of Cappella Sansevero: On Art, Vision and Reality,” Leonardo 46(1): 19–23 (2013). Macknik, Martinez-Conde, and Blakeslee, Sleights of Mind: What the Neuroscience of Magic Reveals About Our Everyday Deceptions (New York: Henry Holt and Co., 2010).
47J. S. Bruner and L. Postman, “On the Perception of Incongruity: A Paradigm,” Journal of Personality 18(2): 206–223 (1949).
the “ground truth” of the external world—it appears that accurate eyewitness identification may be difficult to achieve.
Additional noise (in this case defined as uncertainty resulting from loss of perceptual resolution) may result from the fact that visual perception is categorical.48 Although the objects of our experience vary broadly along multiple sensory dimensions, we lump them into categories based upon prior associations, many of which stem from common functions, physical properties, meanings, or emotional valence. Apples in a basket or the many typographic fonts for the letter “A” are visually distinct, yet we readily perceive them as categorically identical. For most behavioral and cognitive goals, perceptual processing is greatly simplified by treating all members of a category as the same, despite their differences. It rarely matters, for example, whether the apple we choose is dappled on one side or irregular in shape, nor does the font used bear greatly on our ability to read. One of the functional corollaries of categorical perception is that observers are far better at discriminating between objects from different categories than objects from the same category.49 Evidence indicates that the structure of object memory is also categorical, suggesting that perceived objects are encoded in memory as a category type, often without specific detail.50
Perceptual categorization naturally applies to faces.51 We readily categorize faces by distinctions along the obvious dimensions of gender, age,
48W. James, Principles of Psychology (New York: Henry Holt, 1980). S. Harnad, ed., Categorical Perception: The Groundwork of Cognition (New York: Cambridge University Press, 1987).
49R. Goldstone, “Influences of Categorization on Perceptual Discrimination,” Journal of Experimental Psychology General 123(2): 178–200 (1994). R. Goldstone, Y. Lippa, and R. M. Shiffrin, “Altering Object Representations Through Category Learning,” Cognition 78(1): 27–43 (2001).
50E. Tulving, “Episodic and Semantic Memory,” in Organization of Memory, ed. E. Tulving and W. Donaldson (New York: Academic Press, 1972), 381–403. L. K. Tyler et al., “Processing Objects at Different Levels of Specificity,” Journal of Cognitive Neuroscience 16(3): 351–362 (2004). M. J. Farah and J. L. McClelland, “A Computational Model of Semantic Memory Impairment: Modality Specificity and Emergent Category Specificity,” Journal of Experimental Psychology: General 120 (4): 339–357 (1991). C. Gerlach et al., “Categorization and Category Effects in Normal Object Recognition: A PET Study,” Neuropsychologia 38(13): 1693–1703 (2000). G. W. Humphreys and E. M. Forde, “Hierarchies, Similarity, and Interactivity in Object Recognition: ‘Category-Specific’ Neuropsychological Deficits,” Behavioral and Brain Sciences 24(3): 453–476 (2001).
51J. M. Beale and F. C. Keil, “Categorical Effects in the Perception of Faces,” Cognition 57(3): 217–239 (1995). D. T. Levin, “Classifying Faces by Race: The Structure of Face Categories,” Journal of Experimental Psychology: Learning, Memory, and Cognition 22(6):1364–1382 (1996). D. T. Levin and J. Beale, “Categorical Perception Occurs in Newly Learned Faces, Cross-Race Faces, and Inverted Faces,” Perception and Psychophysics 62: 386–401 (2000). M. A. Webster et al., “Adaptation to Natural Facial Categories,” Nature 428(6982): 557–561 (2004). Y. Lee et al., “Broadly Tuned Face Representation in Older Adults Assessed by Categorical Perception,”Journal of Experimental Psychology: Human Perception and Performance 40(3): 1060–1071 (2014).
and race, but we also draw distinctions along dimensions such as skin tone, hair color and style, presence and type of facial hair, such subtler factors as shape of cheeks and jaw, and subjective qualities such as attractiveness. The practical consequence of this for eyewitness identification is that the precision of a perceptual experience may be reduced within any of these categories, particularly because we typically witness criminal events for such a brief period of time. The ensuing memory of the experience will likely reflect that reduced precision, and the memory retrieved may regress to a category prototype or to other exemplars of the perceived category.52 The witness may categorically perceive a square jawed man with a moustache, but the fine details needed for individuation of a suspect are neither perceived nor encoded in memory. For example, although you may have seen the iconic Marlboro Man countless times on billboards and in magazines, it is unlikely that you could distinguish him in a lineup from other square jawed mustachioed men.
Functional Processes of Memory
Conscious visual perceptual experiences, rendered by the processes described in the previous section on vision, are commonly stored as declarative memories, meaning that they can be consciously accessed and expressed as knowledge about the world (as distinct from procedural memories, such as motor skills).53 Declarative memories are of two types, semantic and episodic, reflecting a distinction between memories of meanings, facts, and concepts versus memories of events (such as those witnessed during a crime).54 Declarative memories are conceptualized as involving three core processes—encoding, storage, and retrieval—which refer to the placement of items in memory, their maintenance therein, and subsequent access to the stored information.55
Like vision, memory is also beset by noise. Encoding, storage, and remembering are not passive, static processes that record, retain, and divulge
52J. Huttenlocher, L. V. Hedges, and J. L. Vevea, “Why Do Categories Affect Stimulus Judgement?” Journal of Experimental Psychology: General 129(2): 220–241 (2000). R. Goldstone, Y. Lippa, and R. M. Shiffrin, “Altering Object Representations Through Category Learning,” Cognition 78(1): 27–43 (2001).
53W. James, Principles of Psychology (New York: Henry Holt, 1890). B. Milner, Physiologie de l’hippocampe, ed. P. Passouant (Paris: Centre National de la Recherche Scientifique, 1962), 257–272. L. R. Squire and J. Wixted, “The Cognitive Neuroscience of Human Memory since H.M.,” Annual Review of Neuroscience 34: 259–288 (2011).
54Tulving, “Episodic and Semantic Memory.”
55E. Tulving, “Organization of Memory: Quo vadis?” in The Cognitive Neurosciences, ed. M. S. Gazzaniga (Cambridge, MA: MIT Press, 1995), 839–847.
their contents in an informational vacuum, unaffected by outside influences. The contents cannot be treated as a veridical permanent record, like photographs stored in a safe. On the contrary, the fidelity of our memories for real events may be compromised by many factors at all stages of processing, from encoding through storage, to the final stages of retrieval. Without awareness, we regularly encode events in a biased manner and subsequently forget, reconstruct, update, and distort the things we believe to be true.56
The following sections discuss memory encoding, storage, and retrieval, with emphasis on the limits of these processes as they pertain to eyewitness identification. Emotions can strongly influence these processes of memory; some specific actions are highlighted. The phenomenon of “recognition memory” is also discussed. This refers to the specific type of memory retrieval in which a stimulus (e.g., a face) is used to probe memory, and the rememberer (e.g., an eyewitness) must decide whether the strength of the elicited memory evidence is sufficient to declare that the stimulus was previously encountered or is novel. Recognition memory underlies eyewitness identification, as the witness must make a recognition decision.
Memory encoding refers to the process whereby perceived objects and events are initially placed into storage. The encoding process involves two stages, which are commonly distinguished by the quantity of information stored, the duration of storage, and the susceptibility to interference.57 Short-term or working memory is the conscious content of recent perceptual experiences or information recently recalled from long-term storage. Information that remains at the focus of attention persists in and forms the contents of short-term memory. This form of memory is of limited duration
56J. T. Wixted, “The Psychology and Neuroscience of Forgetting,” Annual Review of Psychology 55: 235–269 (2004). E. Tulving and D. M. Thomson, “Encoding Specificity and Retrieval Processes in Episodic Memory,” Psychological Review 80(5): 352–373 (1973). Y. Dudai, “Reconsolidation: The Advantage of Being Refocused,” Current Opinion in Neurobiology 16(2): 174–178 (2006). E. F. Loftus, “Planting Misinformation in the Human Mind: A 30-Year Investigation of the Malleability of Memory,” Learning and Memory 12(4): 361–366 (2005). R. A. Bjork, “Interference and Memory,” in Encyclopedia of Learning and Memory, ed. L. R. Squire (New York: Macmillan, 1992), 283–288. J. A. McGeoch, “Forgetting and the Law of Disuse,” Psychological Review 39(4): 352–370 (1932). J. G. Jenkins and K. M. Dallenbach, “Obliviscence during Sleep and Waking,” The American Journal of Psychology 35(4): 605–612 (1924). B. J. Underwood and L. Postman, “Extra-Experimental Sources of Interference in Forgetting,” Psychological Review 67 (2): 73–95 (1960).
57R. C. Atkinson and R. M. Shiffrin, “Human Memory: A Proposed System and its Control Processes,” in The Psychology of Learning and Motivation (Volume 2), ed. K. W. Spence and J. T. Spence (New York: Academic Press,1968), 89–195. W. James, Principles of Psychology (New York: Henry Holt, 1890). A. Baddeley, “Working Memory: Looking Back and Looking Forward,” Nature Reviews Neuroscience 4(10): 829–839 (2003). A. Baddley, Working Memory (New York: Oxford University Press, 1986).
and capacity58 and labile, decaying quickly with time and easily disrupted by other perceptual or cognitive processes.59 Through cellular and molecular events that play out over time, the contents of short-term memories may be encoded and consolidated into long-term memory,60 which is more enduring (albeit evolving with ongoing experience), and of greater capacity.
The structure of an individual’s full library of long-term declarative memories can be thought of as a collection of associations between items of specific semantic (e.g., the fact that that person X is a 34-year-old female) or episodic content (e.g., the fact that person X was at location Y on the night of the witnessed crime).61 As the individual gains new experiences, long-term declarative memories may be updated by adding new content to the existing library or by forming new associations between existing content.62
Memories are particularly labile during the encoding process. The contents of short-term memory are limited and highly subject to interference by subsequent sensory, cognitive, emotional, or behavioral events; the contents can also be biased by prior knowledge, expectations, or beliefs, resulting in a distorted representation of experience. Short-term memories of events that happened early in a witnessed proceeding may simply be forgotten with the passage of time or badly compromised by attention directed to subsequent emotional events or cognitive and behavioral demands (e.g., anxiety, fear, the need to escape). In such cases, the compromised information may never be consolidated fully into long-term storage or that storage may contain distorted content.63 At the same time, the quality of encoding of stimuli that are attended is commonly enhanced by highly emotional content.64
58G. A. Miller, “The Magical Number Seven,” The Psychological Review 63(2): 81–97 (1956).
59J. Jonides et al., “The Mind and Brain of Short-Term Memory,” Annual Review of Psychology 59: 193–224 (2008).
60E. Kandel and L. Squire, Memory: From Mind to Molecules (New York: Scientific American Library, 2008).
61J. R. Anderson, The Architecture of Cognition (Cambridge: Harvard University Press, 1983). J. R. Anderson and C. Lebiere, The Atomic Components of Thought (Mahwah: Lawrence Erlbaum Associates, 1998).
62M. P. Walker et al., “Dissociable Stages of Human Memory Consolidation and Reconsolidation,” Nature 425: 616 (2003).
63J. L. McGaugh, “Memory—a Century of Consolidation,” Science 287(5451): 248–251 (2000). J. L. McGaugh and B. Roozendaal, “Role of Adrenal Stress Hormones in Forming Lasting Memories in the Brain,” Current Opinion in Neurobiology 12(2): 205–210 (2002).
64K. N. Ochsner, “Are Affective Events Richly Recollected or Simply Familiar? The Experience and Process of Recognizing Feelings Past,” Journal of Experimental Psychology: General 129 (2): 242–261 (2000). D. Talmi, et al., “Immediate Memory Consequences of the Effect of Emotion on Attention to Pictures,” Learning and Memory 15(2008): 172–182. E. A. Kensinger and D. L. Schacter, “Neural Processes Supporting Young and Older Adults’ Emotional Memories,” Journal of Cognitive Neuroscience 7 (2008): 1–13. E. A. Phelps. “Emotion and Cognition: Insights from Studies of the Human Amygdala,” Annual Review of Psychology 57: 27–53 (2006).
Memory storage refers to the long-term retention of information after encoding. The stability of stored information is continuously challenged and subject to modification. We forget, qualify, or distort existing memories as we acquire new perceptual experiences and encode new content and associations into memory.65
Forgetting can be partially mitigated, and memories stabilized, by habits of retrieval (or reactivation) and reconsolidation, which happen whenever we tell the story of our experiences.66 Reactivation is not perfect. With each implicit retrieval or explicit telling of a story, we may unconsciously smooth over inconsistencies or modify content based on our prior beliefs, the accounts of others, or through the lens of new information. We may add embellishments that reflect opinions, emotions, or prejudices67 rather than observed facts; or we may simply omit disturbing content and pass over fine details.68
A second threat to the stability of long-term memories is, ironically, our life-long ability to learn new things. Because memory mechanisms are inherently plastic throughout life, content stored for the long term is surprisingly labile in the face of new information. Our memories are thus an ever-evolving account of our experiences. A memory that reflects witnessing person X at location Y on a particular evening might be readily and notably updated by subsequent learning that location Y is the home of a business associate of person X. Our memories of the witnessed actions of person
65J. T. Wixted, “The Psychology and Neuroscience of Forgetting,” Annual Review of Psychology 55: 235–269 (2004). Tulving and Thomson, “Encoding Specificity and Retrieval Processes.” Y. Dudai, “Reconsolidation: The Advantage of Being Refocused,” Current Opinion in Neurobiology 16(2): 174–178 (2006). E. F. Loftus, “Planting Misinformation in the Human Mind: A 30-Year Investigation of the Malleability of Memory,” Learning and Memory 12(4): 361–366 (2005). R. A. Bjork, “Interference and Memory,” in Encyclopedia of Learning and Memory, ed. L. R. Squire (New York: Macmillan, 1992), 283–288. J. A. McGeoch, “Forgetting and the Law of Disuse,” Psychological Review 39(4): 352–370 (1932). J. G. Jenkins and K. M. Dallenbach, “Obliviscence During Sleep and Waking,” The American Journal of Psychology 35 (1924): 605–612. B. J. Underwood and L. Postman, “Extra-Experimental Sources of Interference in Forgetting,” Psychological Review 67(2): 73–95 (1960). E. F. Loftus, “The Malleability of Human Memory,” American Scientist 67(3): 312–320 (1979). D. J. Yi et al., “When a Thought Equals a Look: Refreshing Enhances Perceptual Memory,” Journal of Cognitive Neuroscience 20(8): 1371–1380 (2008).
66C. M. Alberini, Memory Reconsolidation (Waltham: Academic Press, 2013).
67D. L. Schacter, Psychology, Second Edition (New York: Worth Publishers, 2011), 253–254. E. F. Loftus and H. G. Hoffman, “Misinformation and Memory, the Creation of New Memories,” Journal of Experimental Psychology 118(1): 100–104 (1989). G. Mazzoni and A. Memon, “Imagination Can Create False Autobiographical Memories,” Psychological Science 14(2): 186–188 (2003).
68F. C. Bartlett, Remembering: A Study in Experimental and Social Psychology (London: Cambridge University Press, 1932).
X may be qualified by new knowledge of his or her life history. Moreover, because new content can be added and the source of that content forgotten, we may attribute our updated memories to the originally witnessed events—in some cases substantially changing what we believe we have seen.69 It is thus not surprising that newly incorporated information need not be true to fact. Research on false memories shows that it is possible to plant fabricated content in memory, which leads us to recall things we never experienced.70
The emotional content of stored memories is a factor that appears to promote long-term retention; memories of highly arousing emotional stimuli, such as those associated with a witnessed crime, tend to be more enduring than memories of non-arousing stimuli.71 Highly salient, unexpected, or arousing events—such as the Kennedy assassination or the Space Shuttle disaster—are commonly more strongly stored in memory, and their later retrieval is often associated with the subjective experience
69D. S. Lindsay and M. K. Johnson, “Recognition Memory and Source Monitoring,” Bulletin of the Psychonomic Society 29(3): 203–205 (1991). D. L. Schacter and C. S. Dodson, “Misattribution, False Recognition and the Sins of Memory,” Philosophical Transactions of the Royal Society: Biological Sciences 356(1413): 1385–1393 (2001). L. A. Henkel, N. Franklin, and M. K. Johnson, “Cross-Modal Source Monitoring Confusions Between Perceived and Imagined Events,” Journal of Experimental Psychology: Learning, Memory, and Cognition 26(2): 321–335 (2000). D. L. Schacter, ed., Memory Distortion: How Minds, Brains, and Societies Reconstruct the Past (Cambridge, MA: Harvard University Press, 1995). K. J. Mitchell and M. K. Johnson, “Source Monitoring: Attributing Mental Experiences,” in The Oxford Handbook of Memory, ed. E. Tulving and F. I. M. Craik (New York: Oxford University Press, 2000), 179–195. H. L. Roediger III and K. B. McDermott, “Creating False Memories: Remembering Words Not Presented in Lists,” Journal of Experimental Psychology: Learning, Memory, and Cognition 21(4): 803–814 (1985).
70Loftus, “Planting Misinformation in the Human Mind.” E. F. Loftus and J. E. Pickrell, “The Formation of False Memories,” Psychiatric Annals 25(12): 720–725 (1995). M. K. Johnson and C. L. Raye, “False Memories and Confabulation,” Trends in Cognitive Sciences 2(4): 137–145 (1998).
71L. J. Kleinsmith and S. Kaplan, “Paired-Associate Learning as a Function of Arousal and Interpolated Interval” Journal of Experimental Psychology 65(2): 190–193 (1963). M. W. Eysenck, “Arousal, Learning, and Memory,” Psychological Bulletin 83(3): 389–404 (1976). F. Heuer and D. Reisberg, “Vivid Memories of Emotional Events: The Accuracy of Remembered Minutiae,” Memory and Cognition 18(5): 496–450 (1990). T. Sharot and E. A. Phelps, “How Arousal Modulates Memory: Disentangling the Effects of Attention and Retention,” Cognitive, Affective, and Behavioral Neuroscience 4(3): 294–306 (2004). E. A. Kensinger, R. J. Garoff-Eaton, and D. L. Schacter, “Memory for Specific Visual Details Can Be Enhanced by Negative Arousing Content,” Journal of Memory and Language 54(1): 99–112 (2006). E. Kensinger, “Remembering Emotional Experiences: The Contribution of Valence and Arousal,” Reviews in the Neurosciences 15(4): 241–251 (2004).
of high vividness and a sense of reliving72 (although not necessarily with greater accuracy, as detailed below). The stronger encoding and storage of emotional memories results from the engagement of a specialized system of stress hormones (glucocorticoids) which is triggered by arousing content and has potentiating effects on the neuronal processes underlying memory consolidation and storage.73 Despite the vividness and the sense of reliving that characterizes retrieval of emotional memories, there are many indications that such memories are just as prone to errors.74 This may reflect, in part, memory enhancements, of the sort described above, which accompany frequent re-consolidation or re-telling of the story of the emotional experience, and often include details (some true to fact, some not) learned after the experience.75 Although emotional memories are often inaccurate in detail, one important corollary of their vividness is that they are frequently
72G. Wolters and J. J. Goudsmit, “Flashbulb and Event Memory of September 11, 2001: Consistency, Confidence and Age Effect,” Psychological Report 96: 605–619 (2005). E. A. Kensinger, A. C. Krendl, and S. Corkin, “Memories of an Emotional and a Nonemotional Event: Effects of Aging and Delay Interval,” Experimental Aging Research 32: 23–45 (2006). U. Neisser and N. Harsch, “Phantom Flashbulbs: False Recollections of Hearing the News about Challenger,” in Affect and Accuracy in Recall: Studies of “Flashbulb” Memories, ed. E. Winograd and U. Neisser (New York: Cambridge University Press, 1992): 9–31. K. S. LaBar and E. A. Phelps, “Arousal-Mediated Memory Consolidation: Role of the Medial Temporal Lobe in Humans,” Psychological Science 9(6): 490–493 (1998).
73J. L. McGaugh, “Memory: A Century of Consolidation,” Science 287(5451): 248–251 (2000). J. L. McGaugh and B. Roozendaal, “Role of Adrenal Stress Hormones in Forming Lasting Memories in the Brain,” Current Opinion in Neurobiology 12(2): 205–210 (2002).
74E. A. Kensinger, “Remembering the Details: Effects of Emotion,” Emotion Review 1(2): 99–113 (2009). T. Sharot, M. R. Delgado, and E. A. Phelps, “How Emotion Enhances the Feeling of Remembering,” Nature Neuroscience 7(12): 1376–1380 (2004). H. Schmolck, E. A. Buffalo, and L. R. Squire, “Memory Distortions Develop over Time: Recollections of the O. J. Simpson Trial Verdict after 15 And 32 Months,” Psychological Science 11 (1): 39–45 (2000). S. R. Schmidt, “Autobiographical Memories for the September 11th Attacks: Reconstructive Errors and Emotional Impairment of Memory,” Memory and Cognition 32(3): 443–454 (2004). T. W. Buchanan and R. Adolphs, “The Role of the Human Amygdala in Emotional Modulation of Long-Term Declarative Memory,” in Emotional Cognition: From Brain to Behavior, ed. S. Moore and M. Oaksford (Amsterdam: John Benjamins Publishing, 2002), 9–34.
75E. Soleti et al., “Does Talking About Emotions Influence Eyewitness Memory? The Role of Emotional vs. Factual Retelling on Memory Accuracy,” Europe’s Journal of Psychology 8(4): 632–640 (2012).
The enduring plasticity of stored memories is a serious concern for the validity of eyewitness identification. A witness’ inevitable interactions with law enforcement and legal counsel, not to mention communications from journalists, family, and friends, have the potential to significantly modify the witness’ memory of faces encountered and of other event details at the scene of the crime.78 Thus, the fidelity of retrieved events—and the accuracy of identification—is likely to be greater when retrieval occurs closer to the time of the witnessed events. The conclusion above has important implications for law enforcement and the legal process and calls into question the validity of in-court identifications and their appropriateness as statements of fact.
Memory retrieval refers to the process by which stored information is accessed and brought into consciousness, where it can be used to make decisions and guide actions. Retrieval of long-term declarative memories is often triggered through association with an external stimulus (i.e., a retrieval cue).79 For example, the slight stubble on a lineup participant’s face may be enough to elicit retrieval of a suspect’s entire face. These same retrieval processes can also be engaged internally—a verbally triggered stream of thought related to a witnessed crime may readily bring to mind visual features of the perpetrator. A corollary of this association-based phenomenon is that memory retrieval is often context dependent; a memory may be more
76U. Rimmele et al., “Emotion Enhances the Subjective Feeling of Remembering, Despite Lower Accuracy for Contextual Details,” Emotion 11(3): 553–562 (2011). Kensinger, “Remembering the Detail.” Neisser and Harsh, Affect and Accuracy in Recall. E. A. Phelps and T. Sharot, “How (and Why) Emotion Enhances the Subjective Sense of Recollection,” Current Directions in Psychological Science 17(2): 147–152 (2008).
77K. A. Houston et al., “The Emotional Eyewitness: The Effects of Emotion on Specific Aspects of Eyewitness Recall and Recognition Performance,” Emotion 13(1): 118–128 (2013). R. B. Edelstein et al., “Emotion and Eyewitness Memory,” in Memory and Emotion, ed. D. Reisberg and P. Hertel (New York: Oxford University Press, 2004): 308–346. S-A. Christian-son, “Emotional Stress and Eyewitness Memory: A Critical Review,” Psychological Bulletin 112(2): 284–309 (1992).
78M. S. Zaragoza and S. M. Lane, “Sources of Misattribution and Suggestibility of Eyewitness Memory,” Journal of Experimental Psychology: Learning, Memory, and Cognition 20 (4): 934–945 (1994). W. C. Thompson, K. A. Clarke-Stewart, and S. J. Lepore, “What Did the Janitor Do? Suggestive Interviewing and the Accuracy of Children’s Accounts,” Law and Human Behaviour 21(4): 405–426 (1997). D. S. Lindsay and M. K. Johnson, “The Eyewitness Suggestibility Effect and Memory for Source,” Memory and Cognition 17(3): 349–358 (1989).
79E. Tulving and Z. Pearlstone, “Availability Versus Accessibility of Information in Memory for Words,” Journal of Verbal Learning and Verbal Behaviour 5: 381–391 (1966).
readily retrieved if the observer is in physical surroundings that are the same as or similar to those in which the original experiences took place (because the surroundings provide additional cues to trigger memory retrieval).80
Memory retrieval is heavily affected by various sources of noise. Similarities of meaning or appearance between retrieval cues and items in memory can easily lead to retrieval of the wrong item, producing a false memory.81 This is particularly a problem given the categorical nature of memory.82 The rugged mustachioed man in the lineup may lead to retrieval of the familiar categorical prototype—the Marlboro Man—rather than the specific person perceived at the scene of the crime, which in turn could interfere with or lead to errors in recognition (i.e., identification). Another type of memory retrieval failure is caused by “intrusion errors,” in which information known to be commonly associated with events of a general type becomes incorporated into the retrieved content of a specific memory (and subsequently incorporated into the reconsolidated memory). For example, because guns are often associated with robbery, an observer may readily and unwittingly incorporate a gun into the retrieved version of his or her memory of a witnessed robbery.
Intrusion errors are one manifestation of a larger retrieval problem in which there is loss of information about the source of a memory. In cases of “source memory failure,” we effectively forget how we know things (forget when and where we learned the content of our memories). What this means practically is that we may attribute later acquisition of information to earlier experiences. An eyewitness might learn from the police or some other source that a potential suspect has a moustache and then attribute
80D. Godden and A. Baddeley, “Context Dependent Memory in Two Natural Environments,” British Journal of Psychology 66(3): 325–331 (1975). S. M. Smith and E. Vela, “Environmental Context-Dependent Eyewitness Recognition,” Applied Cognitive Psychology 6: 125–139 (1992). S. M. Smith and E. Vela, “Environmental Context-Dependent Memory: A Review and Meta-Analysis,” Psychonomic Bulletin Review 8 (2): 203–220 (2001). Tulving and Thomson, “Encoding Specificity and Retrieval Processes.”
81J. R. Anderson, “A Spreading Activation Theory of Memory,” Journal of Verbal Learning and Verbal Behavior 22(3): 261–295 (1983). A. M. Collins and E. F. Loftus, “A Spreading-Activation Theory of Semantic Processing,” Psychological Review 82(6):407–428 (1975). H. L. Roediger III, D. A. Balota, and J. M. Watson, “Spreading Activation and Arousal of False Memories,” in The Nature of Remembering: Essays in Honor of Robert G. Crowder, ed. H. L. Roediger III, J. Nairne, I. Neath, and A. Surprenant (Washington, DC: American Psychological Association, 2001): 95–115. C. J. Brainerd and V. F. Reyna, The Science of False Memory (New York: Oxford University Press, 2005).
82Tulving, “Episodic and Semantic Memory.” M. J. Farah and J. L. McClelland, “A Computational Model of Semantic Memory Impairment: Modality Specificity and Emergent Category Specificity,” Journal of Experimental Psychology: General 120(4): 339–357 (1991). G. W. Humphreys and E. M. Forde, “Hierarchies, Similarity, and Interactivity in Object Recognition: ‘Category-Specific’ Neuropsychological Deficits,” Behavioral and Brain Sciences 24(3): 453–476 (2001).
that knowledge to the witnessed events, which may, in turn, have disastrous consequences for the ability of the eyewitness to accurately report what she or he has seen.
As for the processes of memory encoding and storage, the emotional content of memory also affects memory retrieval. As noted above, memory retrieval is commonly context dependent. A related and well-documented phenomenon that bears on emotional memories is state dependent memory, in which retrieval accuracy is best if the individual’s cognitive state at the time of retrieval matches cognitive state at the time of encoding.83 When memories have an emotional component, retrieval may be best when the individual is induced to a corresponding emotional state (mood dependent memory),84 which is accomplished by verbally or physically placing him or her in the same context, and may offer a valuable investigative tool for probing eyewitness accounts.85
Recognition memory is a specific type of declarative memory retrieval in which a sensory stimulus (a “cue” stimulus) elicits a memory of the stimulus stored following a prior encounter and often the sequence of events involving the stimulus, the spatial context in which the stimulus was experienced, and the presence of other objects, people, or thoughts that had appeared with the stimulus during the event.86 Recognition memory decisions are based on the retrieved memory evidence, which can be triggered by the stimulus and can also emerge from an active search of items
83D. W. Goodwin et al., “Alcohol and Recall: State-Dependent Effects in Man,” Science 163(3873): 1358–1360 (1969). Tulving and Thomson, “Encoding Specificity and Retrieval Processes.” Psychological Review 80(5): 352–373 (1973). E. Girden and E. Culler, “Conditioned Responses in Curarized Striate Muscle in Dogs,” Journal of Comparative Psychology 23(2): 261–274 (1937). D. A. Overton, “State-Dependent or ‘Dissociated’ Learning Produced with Pentobarbital,” Journal of Comparative and Physiological Psychology 57(1): 3–12 (1964).
84P. M. Kenealy, “Mood State-Dependent Retrieval: The Effects of Induced Mood on Memory Reconsidered,” The Quarterly Journal of Experimental Psychology Section A: Human Experimental Psychology 50(2): 290–317 (1997). P. A. Lewis and H. D. Critchley, “Mood-Dependent Memory,” Trends in Cognitive Sciences 7(10): 431–433 (2003). G. H. Bower, “Mood and Memory,” American Psychologist 36(2): 129–148 (1981). F. I. M. Craik and R. S. Lockhart, “Levels of Processing: A Framework for Memory Research,” Journal of Verbal Learning and Verbal Behavior 11(6):671–684 (1972). Kensinger, “Remembering the Detail.” K. A. Leight and H. C. Ellis “Emotional Mood States, Strategies, and State-Dependency in Memory,” Journal of Verbal Learning and Verbal Behavior 20(3): 251–266 (1981).
85S. M. Smith and E. Vela, “Environmental Context-Dependent Eyewitness Recognition,” Applied Cognitive Psychology 6: 125–139 (1992).
86G. Mandler, “Recognizing: The Judgment of Previous Occurrence,” Psychological Review 87(3): 252–271 (1980).
in memory. One factor affecting the strength of the evidence retrieved is the similarity between the cue stimulus and the stimulus or stimuli that was/were previously encountered during the event. An observer engaged in this process holds an implicit criterion for the strength of evidence required to reach a positive decision. In the case of eyewitness identification, this process is routinely elicited by viewing faces in a lineup. When the evidence retrieved is insufficient to reach a decision, this can lead to a cycle of ever-greater scrutiny of the cue stimulus and efforts to recollect additional details of the original event. Ultimately a decision must be made about whether the retrieved evidence is sufficient to declare that the stimulus was previously experienced (or previously experienced in the particular event of interest) or whether the stimulus is novel (or not from the event of interest). If a recognition event occurs—that is, if the memory search triggered by one of the faces in the lineup leads to a strong enough subjective experience that the face is familiar and/or the recollection of sufficient event details—then the witness may declare that they recognize the face as having been previously encountered. Recognition memory decisions can thus be thought of as the final stage in the process of eyewitness identification.
Because it is a form of memory retrieval, recognition memory is susceptible to all of the factors summarized above that are known to interfere with retrieval. Recognition memory differs from other forms of retrieval (such as recalling a phone number or a cake recipe), however, in that a comparison must be made between the retrieved evidence and a decision threshold. That is, as noted above, recognition judgments require a decision criterion, an understanding of which presents a unique set of challenges for eyewitness identification (and recognition memory, generally). In particular, an observer’s report of recognition (or, in a lineup setting, of identification) is influenced not simply by the strength or quality of the recalled memory evidence. The report of recognition (identification of a lineup member) is also influenced by the level of evidence that the observer finds acceptable to reach such a decision, i.e., by his or her decision criterion, or bias. An observer who holds a liberal criterion will likely recognize many true targets (i.e., the guilty), but will frequently err by reporting recognition of many false targets (i.e., innocents). Conversely, an observer who holds a conservative criterion will avoid the problem of erroneous recognition (identification), but will fail to identify some true targets. Estimating (or controlling) the observer’s decision criterion is thus a critical step in efforts to judge the validity of an identification (see also Chapter 5).
Recognition memory for faces differs greatly between familiar and unfamiliar faces.87 Because we often identify familiar individuals with ease,
87P. J. B. Hancock, V. Bruce, and A. M. Burton, “Recognition of Unfamiliar Faces,” Trends in Cognitive Sciences 4(9): 330–337 (2000).
we tend to think we are generally very good at face recognition. However, we are not as good with unfamiliar faces.88 All of the sources of noise that influence perception and memory contribute to these difficulties, and they are exacerbated by the attempts by criminals to conceal their identity (even a change in hairstyle and clothing can have a major effect on recognition).
The ability to recognize unfamiliar faces differs widely across individuals. At one extreme are those people, referred to as “super recognizers,” who rarely forget a face.89 At the other end of the spectrum are “face-blind people (prosopagnosics),” who have great difficulty recognizing even highly familiar faces.90 Current estimates of the fraction of the general population afflicted by prosopagnosia are as high as ~2 percent.91 The ability of an eyewitness to identify a suspect may thus differ greatly from individual to individual simply as a consequence of general variations in face recognition ability.
The shortcomings of eyewitness identification present a societal problem that has profound implications for our systems of law and justice. Ultimately, a solution to this problem must be informed by a thorough understanding of human vision and memory. The processes of vision and memory, which are fundamental to human experience, have been frequent targets of scientific investigation since the 19th century. The past few decades have seen an explosion of additional research that has led to important insights into how vision and memory work, what we see and remember best, and what causes these processes to fail. The committee has reviewed much of this research, as it pertains to eyewitness identification, and has identified restrictions on what can be seen under specific environmental and behavioral conditions (e.g., as poor illumination, limited viewing duration, viewing angle), factors that impede the ability to attend to critically informative features of a visual scene (e.g., the deleterious effect of an attention-grabbing element, such as a weapon, on the ability to correctly perceive the features of the assailant’s face), distortions of perceptual experience derived from expectations, and ways in which emotion and stress enhance or suppress specific perceptual experiences. Memory is often far
88V. Bruce, “Changing Faces: Visual and Non-Visual Coding Processes in Face Recognition,” British Journal of Psychology 73: 105–116 (1982).
89R. Russell, B. Duchaine, and K. Nakayama, “Super-Recognisers: People with Extraordinary Face Recognition Ability,” Psychonomic Bulletin and Review 16(2): 252–272 (2009).
90T. Susilo and B. Ducahine, “Advances in Developmental Prosopagnosia Research,” Current Opinion in Neurobiology 2(3):423–429 (2013).
91I. Kennerknecht et al., “First Report of Prevalence of Nonsyndromic Hereditary Prosopagnosia (HPA),” American Journal of Medical Genetics Part A 140(15): 1617–1622 (2006).
from a faithful record of what was perceived through the sense of sight: its contents can be forgotten or contaminated at multiple stages, it can be biased by the very practices designed to elicit recall, and it is heavily swayed by emotional states associated with witnessed events and their recall. From this analysis, the committee must conclude that there are insurmountable limits on vision and memory imposed by our biological nature and the properties of the world we inhabit. With this knowledge, it is possible to more fully appreciate the value and risks associated with eyewitness reports and accordingly advise those who collect, handle, defend, consider, and adjudicate such reports.