Expertise in Crime Scene Examination

Abstract

Objective: The strategies of novice and expert crime scene examiners were compared in searching crime scenes.

Background: Previous studies have demonstrated that experts frame a scene through reconstructing the likely actions of a criminal and use contextual cues to develop hypotheses that guide subsequent search for evidence.

Method: Novice (first-year undergraduate students of forensic sciences) and expert (experienced crime scene examiners) examined two “simulated” crime scenes. Performance was captured through a combination of concurrent verbal protocol and own-point recording, using head-mounted cameras.

Results: Although both groups paid attention to the likely modus operandi of the perpetrator (in terms of possible actions taken), the novices paid more attention to individual objects, whereas the experts paid more attention to objects with “evidential value.” Novices explore the scene in terms of the objects that it contains, whereas experts consider the evidence analysis that can be performed as a consequence of the examination.

Conclusion: The suggestion is that the novices are putting effort into detailing the scene in terms of its features, whereas the experts are putting effort into the likely actions that can be performed as a consequence of the examination.

Application: The findings have helped in developing the expertise of novice crime scene examiners and approaches to training of expertise within this population.

Keywords

crime scene examination expertise own-point recording verbal protocol analysis

Santilla, Korpela, and Häkkänen (2004) demonstrate how experienced investigators of car crimes are better able than novices to find links in a series of crimes. This implies an ability to see general patterns in the evidence rather than a focus on specific instances (McRobert, Williams, Ward, & Eccles, 2009; Raab & Johnson, 2007). A core aspect of the expertise of the crime scene examiner (CSE) lies in the ability to recognize places to recover materials that could yield evidence (Baber, 2010). (In the United Kingdom, the role of the CSE is distinct from that of the forensic scientist or other investigators who might attend a crime scene. Typically, the CSE will be a civilian attached to a police force with the primary goal of gathering evidence to pass on, via an exhibits officer, to other parties for subsequent analysis in a laboratory [usually by forensic scientists]. As such, the CSE will most likely attend “volume crime” scenes, focusing on crimes such as street robbery, burglary, theft [including shoplifting], vehicle crime, criminal damage [Association of Chief Police Officers, 2002].) Schraagen and Leijenhorst (2001) studied the behavior of forensic scientists searching for particulate material on clothing and found that the superior performance of experts (in terms of knowing where to target their search) arose from the experts mentally reconstructing the actions or modus operandi of the suspect in that crime scene. This superior search also resulted in more forensic evidence being recovered by the experts. These studies chime well with the general literature on expertise. The expert is, by definition, often better able than the novice to appreciate the key information in a problem space (Abernethy, Neal, & Koning, 1994; Charness, 1979; Chase & Simon, 1973a, 1973b; de Groot, 1965; Didierjean & Fernand, 2008). Thus, compared to novices, expert physicians select a greater number of relevant cues from a case history (Joseph & Patel, 1990) and tend to generate a smaller number of working hypotheses early in the history-taking process (Joseph & Patel, 1990; Kassirer & Gorry, 1978; Patel, Arocha, & Kaufman, 1994).

These studies, however, do not mean that the expert is not prone to bias or error. In a series of experiments into the interpretation of fingerprints, Dror and his colleagues have shown that it is possible to induce bias in analysts’ assessment of fingerprint matching through the provision of background information describing the nature of a crime (Charlton, Fraser-Mackenzie, & Dror, 2010; Dror & Charlton, 2006; Dror, Péron, Hind, & Charlton, 2005).

One can situate the concept of “CSE expertise” on two dimensions. The first concerns the constraints that a particular mode of working imposes of the individual, particularly (for crime scene examination) in terms of time pressure (Smith, Baber, Hunter, & Butler, 2008). It is hypothesized that the expert CSEs will seek to complete the task more quickly than (novice) students. The second dimension of CSE expertise concerns the ability to recognize evidential value. It is proposed that this recognition of evidential value is, in part, analogous to interpreting symptoms in diagnostic activity. However, recognition of evidential value also requires the “expert” to judge whether the recovered object can be useful to someone else. In this case, this will require appreciation of the methods applied by forensic and other scientists to the recovered materials. Thus, it is hypothesized that, in addition to describing the modus operandi to explain the likely places to search for evidence, the expert will also consider the potential evidential value of the material and its potential utility for subsequent analysis.

Method

The study is conducted in a “crime house,” that is, an environment for training crime scene examination. Crime houses allow video recording from ceiling-mounted closed-circuit television (CCTV) cameras, which can be used to support debriefing during training exercises. It was felt that these recordings would not provide sufficiently fine-grained detail for this study. Hence, head-mounted cameras were used to capture the participants’ forward view. Omodei, Wearing, and McLenna (1997) explored head-mounted video recording as a methodology for studying naturalistic decision making by firefighters, which they termed own-point recording.

Initially, we intended to use a wearable eye-tracking system (the ASL MobileEye system) to capture search activity. However, fluctuations in ambient lighting disrupted the infrared tracking, and so the study employed the video recording from the camera without using the eye tracking. This own-point recording was supplemented with concurrent verbal protocol. Using a similar methodology, Eccles, Walsh, and Ingledew (2006) asked orienteers, of varying abilities, to run courses while wearing head-mounted cameras with instructions to declare the focus of attention when it changed (e.g., to say “map” when looking at the map or “travel/ground” when running, or “looking for bridge” when searching for a specific object in the environment). These instructions helped with the coding of the video. The results indicated that although experienced orienteers ran the course faster, there were no differences in proportion of time spent attending to these different categories of visual cues. Although the Eccles et al. approach, of requiring participants to use speech to tag different activities, was appropriate for the dynamic context of that data collection, we chose a method that provided us with a more detailed report from participants.

Ericsson and Simon (1993) distinguish between verbalizing the thoughts that occur to a person when performing a task and the justifications that the person uses to explain his or her task and claim that “only information in focal attention can be verbalized” (p. 90). This distinguishes between verbalization of information as heeded (Level 1) and information as encoded (Level 2) or information as explained (Level 3). Thus, Level 3 verbal protocols are said to exhibit reactivity in that the verbalization changes the nature of the task being performed. Ericsson and Simon note that when participants provide Level 1 or Level 2 verbalization, their performance appears to be similar to conditions without verbalization, but when asked to produce Level 3, performance can change as a result of interference between performing the primary task and producing the verbal report.

A challenge with the use of the think-aloud verbal report method, particularly in dynamic, real-world environments, is to provide sufficient task instruction and feedback to maximize Level 1 or Level 2 verbalization by participants (Fox, Ericsson, & Best, 2011). Despite our best efforts in this regard, the constraints inherent in the present experiment mean that it is not possible to guarantee that the protocols are entirely free from Level 3 verbalization. To take a section from the example provided later in the article, “I can also see something’s been covered over with … erm … plastic and I take it this is the footwear mark that was mentioned in the brief” is interpreted as information as heeded because it indicates the content of the verbal protocol describing the items in the scene. “What I am looking at now is how would I actually get into this window, what would I hold onto” is interpreted as information as encoded because it indicates the focus of attention on the search of scene, that is, in terms of the modus operandi that the perpetrator might have followed. As such, we chose this method because the value added from the information obtained, even if approaching Level 3 verbalization, provided us with important insights into reasoning processes that could be used to guide later studies and training.

Environment

The experiments took place in the crime houses of Teesside University in Middlesbrough, a city in the northeast of England. The crime houses consist of three midterrace houses, with briefing and preparation rooms and a dozen or so rooms that have been furnished to represent different locations, including office, bedroom, pub, and lounge. Each room can be “dressed” to stage a particular crime, with damage or markings on point of entry, marks from fingers, feet, or tools around the room, and biological evidence. Each room has CCTV cameras, and the video is passed to a central video recording and editing suite from which it is possible to monitor the search activity.

Setting the Crime Scenes

Two rooms were dressed to simulate burglaries. To control the presentation of initial information and ensure consistency between each participant, one of the experimenters played the part of the key holder who would meet the examiner and provide a brief explanation of when the crime was discovered.

Scene 1: Lounge

A lounge on the ground floor had an open window, which had been forced open from the outside with a flat bladed instrument, causing striation marks in the uPVC frame. The latch of the window was unscrewed and the screws left on the ground under the opened window. A footwear mark had been left on the external window sill. This mark had been covered with black plastic by the first police officer responding to the incident. Inside, the television cabinet had been damaged, a DVD player had been pulled out of the cabinet, and the glass door had been pulled off its hinges. A footwear mark was left on this door, along with glove marks. Glove marks were also left on other surfaces within the room, that is, the multiplug socket behind the television unit, inside the windows transom bar, and on an empty cell-phone box left on a sofa. The police log for the lounge scene was as follows:

06:13: Complainant Simon Pendle states his property has been burgled in the night whilst he was asleep. Attempt made to steal DVD player, wallet containing £50 cash and cards taken. Nokia mobile, Apple iPod touch and Philips DC910 iPod decking system also believed stolen. Total value around £1k. No signs of disturbance in other rooms.

08:13: Update PC 2324 Gibson at scene. Crime Scene resource will be needed, will update with further details.

09:06: PC 2324 updated: Crime report taken and brief house to house, nothing seen. Entry via front lounge window, pushed-to due to inclement weather. Good footwear preserved on the outside of the property window sill. Since arrival complainant couldn’t find vehicle keys and on checking the rear car parking vehicle is believed to have been stolen. Request update PNC, vehicle is a Ford Fiesta registration details T75SPJ value £800, incident reported to have occurred overnight whilst complainant asleep. Complainant needs to go to work and is arranging lift; key holder (@ No2) is a neighbor and will stay with CSE. Availability all day.

Scene 2: Office

The second scene was an office on the first floor of the building. The office window was forced open from the inside and the window opened a few centimeters. Footwear marks on the wooden floor matched footwear inside a plastic bag under a table. The scene could be interpreted as indicating that the complainant had faked the crime for insurance purposes. The simulated police log for the office scene was as follows:

17:00: Complainant Mr Jarvis reporting a burglary, address No 4 King Edward Square. He states that only one room searched as alarm not activated. Computer and electrical items from his office have been taken: New Acer Aspire 7720G laptop computer value £545, a Samsung ML-1640 Mono laser printer value £70, a TomTom One XL Classic SAT NAV along with £425 from the desk drawer.

18:27: PC 2142 updating arrival at address.

19:38: PC 2142 updated, scenes of crime requested, entry believed by forcing top opening window, force marks observed. Footwear marks on floor of office reported not being there the previous day. Confirmed no other rooms disturbed, alarm sensor in landing and alarm not activated. Exit as Entry. House to House completed, elderly occupant next door states nothing seen or heard. Complainant available for a visit all evening and all day tomorrow.

Participants

A total of 12 participants volunteered to undertake searches of staged crime scenes: 6 full-time, experienced CSEs (5 male, 1 female; mean age approximately 37 years; mean experience approximately 12 years) from three different police forces in the northeast of England and 6 first-year students on a forensic degree pathway at Teesside University in Middlesbrough, United Kingdom (3 female, 3 male; mean age approximately 21 years). The students had attended introductory lectures on crime scene examination but had not conducted an exploration of a scene before. Although it was possible to recruit more students (indeed, subsequent trials have been conducted with the complete cohort of the year), it was decided to balance the two groups in terms of numbers. Recruiting serving CSEs required not only their consent but also permission from their manager to release them from planned duties and allow them to spend time on the trials. This severely limited the availability of operational officers to participate in the studies.

Procedure

Participants were briefed on the aims of the study (to explore how crime scenes are searched) and given an explanation on the data collection methods (own-point recording and verbal protocol). If they wished to continue with the study, they donned the head-mounted camera and completed a short introduction session.

Participants were handed a scene log (as shown above) for the first crime scene to be examined. Presentation of scenes was counterbalanced across participants to minimize practice effects. The details follow the conventional format of a crime report that is received by the CSE prior to attending a job. This would indicate the location of the scene, the person to contact, and a brief description of the likely crime to be examined. Once they had read the paper and were satisfied that they understood it, participants were led to the scene.

Participants searched the scene for as long as they wished. Own-point recording was made using the forward-facing camera of the ASL MobileEye system, and a separate head-mounted microphone was used to record the verbal protocol. Audio and video recording was made on the same analogue video recorder that the participant wore. The main objective for the participants was to select objects that they felt would be most useful to recover as evidence. Thus, the focus of the exploration was on evidence identification and selection rather than preparation or recovery (i.e., participants could identify a location that could be used to recover finger marks but did not need to dust for marks and prints).

Data Analysis

Prior to statistical analysis, the recordings of “talk-aloud” commentaries and own-point recording data were transcribed and coded.

Verbal Protocol

Each participant’s “talk-aloud” commentaries were manually transcribed. An initial coding was applied to categorize statements into one of four types:

Reference to the modus operandi of the perpetrator: describing likely point of entry or activity in the room or possible motive

Reference to objects in the room: desk, chair, computer parts, and so on

Reference to analysis: tool marks, fingerprints, footwear marks, fiber

Reference to the room’s features: type of flooring, and so on

The following extract is from a male CSE (age 45 years; experience 15 years).

Er … arrived at the scene … er … of the outside of the house … er … identifying the point of entry. Just confirming that (.) came from, from information from the IP via the OIC who attended the scene. I can see damage (.) … erm … I can also see something’s been covered over with … erm … plastic and I take it this is the footwear mark that was mentioned in the brief, and I can see a skid footwear mark on there. What I am looking at now is how would I actually get into this window, what would I hold onto (0.5+) and I think that footwear mark would be consistent with somebody climbing up onto there so my (0.5+) next action would be to photograph the outside of this (.) point of entry.

The coding of this extract is shown in Table 1. Coding was undertaken by both authors, and an acceptable level of agreement was reached (r² = .93).

Table 1:

Coding of Extract of Verbal Protocol

Coding	Total	Content
Modus operandi	4	… Identifying the point of entry
		… I can see damage
		… How would I actually get into this window, what would I hold onto
		… Consistent with someone climbing us onto there
Objects	0
Evidence	2	… Skid footwear mark
		… Photograph the outside of this (.) point of entry
Scene	1	Something’s been covered over with … erm … plastic

Video Coding

Figure 1 shows a set of stills from the video produced by a female CSE (age 37 years; experience 8 years). This illustrates the categorization of objects checked from the video. In Still a, the object in the center of the frame is the cash tin; in Still b, the center of the frame is the footwear mark; in Still c, the center of the frame is the sole of the shoe; and in Still d, the center of the frame is the edge of the window and one can detect tool marks around this point.

Figure 1.

Stills taken from the head camera of participants inspecting the office scene.

The stills illustrate the problems of ensuring definitive identification of the object that the person was checking. For example, Still b was taken some 2 minutes into the search, and the analysis would suggest that the check was being performed on the footwear marks on the floor, but the check could be on the computer parts, or the floor covering, or some other aspect of the scene. Consequently, a method to allow activity sampling that was sufficiently context dependent to reflect behavior in each scene while allowing a level of consistency in identifying objects checked was required.

All of the objects categorized from the verbal protocols of CSEs and students were collated into a single set of objects (one for each scene). This provided the baseline for identifying checks. It is accepted that there could be objects in the scene that were checked but not mentioned in the verbal protocol, and this was considered during the analysis. However, review of this list of objects relative to the plan used in the setting up of each scene led to the conclusion that the list was exhaustive.

To resolve the question of which object is being checked, the experimenters made a first pass (with the audio recording off) using own-point recording (supplemented by any “eye-tracking” data if available). (Given the variation in illumination levels, the crosshairs, indicating fixation, tended to be lost when the participants approached the windows in either scene but would be present away from the window. There were two participants in the student group and one participant in the CSE group who had no usable eye-tracking data on the video. Of the remaining video, we would estimate that between 30% and 60% of the video had the crosshairs present but that this was not uniformly split across participants and varied with location in the room. Thus, we decided to use this information to supplement ratings when it was available.) This first pass was supported by an application, written in Visual Basic for Applications (VBA) in Microsoft Excel. Each object from the collated verbal protocols was marked on a sketch of the room. When the analyst felt that the participant performed a check on this object, then the corresponding button was clicked (see Figures 2 and 3).

Figure 2.

Activity sampling sheet for the office.

Figure 3.

Activity sampling sheet for the lounge.

The definition of check in this analysis is quite loose but means that the center of the video frame contained an object for longer than half a second. This meant that periods of attention in which the gaze swept around the room were not recorded in this analysis. Following this first pass, a second pass was performed with the audio recording on. If the participant referred to an object in his or her verbal protocol while this object was in frame, this was used to infer that the object was being checked. It is accepted that looking at an object need not indicate that the person is attending to or checking it, but it was felt that this provided a pragmatic means of determining which objects were felt to warrant checking by participants. Coding was undertaken by both authors and a reasonable, if somewhat low, level of agreement was reached (r² = .68). (The main cause of this difference in agreement lies in the definition of checking applied by the two authors. Given the head movements involved in scene inspection, there were times when the gaze passed over objects and it was a matter of conjecture as to whether the participant had “looked” at the object. In some instances, this was resolved through consultation with the audio recording; in other instances, the authors reviewed their independent assessment to reach consensus. We have not included r² following this procedure [because the process resulted in an agreed unanimity]. However, it is worth noting that subsequent analysis is based on this final, agreed set of object checks.)

Results

Time to Conduct Search

The average time to conduct a scene differed between the two scenes, with participants spending less time in the office than the lounge (see Figure 4). The time to conduct the search was defined as the time from the participant entering the room to him or her reporting that his or her search was complete.

Figure 4.

Comparison of mean search time for the two groups in the two scenes.

Mann–Whitney U tests were calculated to compare the two groups on their times to search the office and the lounge. The students spent more time than CSEs searching the office (U = 4, p = .025) and the lounge (U = 0, p = .002). The students spent significantly more time searching the lounge than the office (U = 0, p = .002), but there was no difference in search times for the CSEs (U = 9, p = .15).

Verbal Protocol

The categorized verbal protocol data indicate that the total number of statements produced by the students averaged 38 statements, and the CSEs produced an average of 28 statements. Thus, there was some difference in the total number of statements made. A two independent samples Wilcoxon rank sum test indicates that there was a significant difference between the two groups of participants (Z = −2.309, p = .021). Figure 5 shows the relative distribution of statements across the four categories used in the coding.

Figure 5.

Comparison of the mean proportion of types of comment made by groups in each scene.

Mann–Whitney U tests were conducted on the data to explore differences in categories of statement. There was no difference between the groups in terms of the proportion of statements that related to modus operandi. There was a significant difference in statements relating to objects for the office (U = 5.5, p = .041) but not for the lounge. There also was a significant difference in references to evidence in both scenes (for the office U = 3, p = .015; for the lounge U = 0, p = .002). Finally, there was no difference between groups in terms of scene.

Video Analysis

An initial contrast can be drawn between the number of objects that were checked by the students (M = 135) and by CSEs (M = 48). Thus, the students made almost 3 times as many checks as the CSEs (see Figure 6). This contrasts with the average number of statements made (which were similar for the two groups) but also points to the difference in number of objects mentioned by students and CSEs.

Figure 6.

Comparison of mean number of items checked by the two groups in each scene.

Mann–Whitney U tests were used to compare the groups. Students checked significantly more objects than the CSEs for the office (U = 3, p = .015) and for the lounge (U = 0.5, p = .002). Within groups, there was no difference in the number of objects checked by students across the two scenes (U = 10, p = .24), but there was a difference for CSEs (U = 3, p = .015), who checked more objects for the lounge.

Conclusions

The results indicate differences in strategy between expert and novice CSEs. Although the office was examined in similar times by the two groups, the lounge required a longer search time for the students. One explanation for this difference could lie in the conspicuity of evidence in the two scenes: The office had footwear marks on the floor and scattered components, whereas the lounge had limited disruption and less conspicuous evidence. Thus, there was some ambiguity in this lounge scene, which seemed to confuse the students and led to them spending longer in their examination. The CSEs searched both scenes in similar times, regardless of the number of objects checked.

Second, the verbal protocols of the two groups indicate some differences in diagnostic activity. Although both groups developed a theory of the modus operandi, in terms of likely actions being performed there was little difference in the proportion of statements devoted to this by the two groups. Conversely, the students had a tendency to report significantly more objects in the scene than the CSEs, which is taken to imply a broad, high-level sampling of the scene. This was further demonstrated by the number of objects checked by the students. The CSEs, on the other hand, reported significantly more objects that could be used as evidence, which is taken to imply a focus on subsequent analysis and investigation. One could interpret this as an indication that the CSEs were being selective in their focus on objects and would pick only those objects that could play a role in the investigation because they had the potential to yield evidence.

Discussion

The study explored three hypotheses centered on the cognitive processes of CSEs. The first hypothesis was that the CSEs would differ from students in their ability to target their search on specific objects. This implies differences in framing, which is similar to the ways in which the problem-solving literature sees elements of a problem being used by experts and novices in different ways. Thus, in medical diagnosis, experts generate a small number of working hypotheses (Joseph & Patel, 1990; Kassirer & Gorry, 1978; Patel et al., 1994) and use heuristics to structure the problem space (Gale & Marsden, 1983; Groves, 2008; Medin & Edelson, 1988). For the students, the scenes were packed with objects that could be related to a crime, but for the CSEs the scenes contained objects that might yield recoverable evidence.

The second hypothesis was that the CSEs would seek to complete the search in less time than the students. The CSEs not only were faster than students in completing the search but also were consistent across scenes and focused attention on fewer objects. This reflects the fact that experts tend to focus their attention on relevant cues from a case history (Joseph & Patel, 1990). However, an expert CSE might seek to complete an initial search in a limited period of time because he or she knows that he or she will have to spend time collecting and recording evidence and producing a report before moving on to the next job, and this might have a bearing on his or her decisions as to how many objects to examine (Smith et al., 2008).

The third hypothesis was that the CSEs would be better able to recognize evidential value than students. The primary differences seem to lie less in the appreciation of modus operandi as in the definition of evidence and how it might be subsequently processed (although the differences in numbers of objects checked by the CSEs in the lounge, compared to the office, could suggest that the less obvious the modus operandi, the more effort could be put into examining the scene). Broadly, it is suggested that the students viewed the crime scene as a place in which a crime had been committed and spent their time considering the nature of the crime and the impact that this activity had on the objects in the scene (as indicated by the focus on objects in their verbal protocols, which indicates a focus on the potential for objects to be relevant to the investigation). In contrast, the CSEs used their understanding of the prosecution of the criminal to help decide which objects would be appropriate to recover for subsequent analysis (as indicated by the focus on evidence in the CSEs’ verbal protocols). Thus, both the novice and expert “saw” the rooms as places in which crimes had been committed, but the experts considered this as an initial stage of the process of criminal investigation (seeking to answer the question “What useful evidence can be recovered?”) whereas the students considered it more as a final state of the investigation (seeking to answer the question “What happened here?”). This shows how expertise represents an adaptation to environmental and task demands (Ericsson & Lehmann, 1996), in that the accrued experience of the expert CSEs from many similar scenes influenced their search strategy and ability to attend to specific cues and that experts are better at making inferences about the state of the environment under scrutiny (Cellier, Eyrolle, & Marine, 2001; Ericsson & Smith, 1991) in terms of the selection of “good” places to recover evidence and are better than novices in anticipating future trends in the process (particularly in terms of the subsequent analysis of recovered material).

Implications for Training

Understanding differences between novice and expert CSEs has the potential to steer education in this field. In this section we briefly highlight how our findings guide further research in support of training for CSEs. First, a long line of research shows how expertise develops in terms of “deliberate practice” (e.g., Horn & Masunaga, 2006). It is only through deliberate practice, coupled with early feedback and the construction of goals, that focused development can occur. Our research has helped to identify potential differences in search strategy between the novice and expert, suggesting ways for providing feedback that encourages reflection on the meaning of “evidential value” and the likely analysis that could be applied to an item of evidence. This allows instructional staff to provide advice to learners as they begin the assessment stage of the examination. This is arguably the most important stage as it significantly affects the examination strategy and ultimately the collection of exhibits for forensic analysis. In a learning environment, it opens up opportunities for instructional staff to question learners on the value of specific types of forensic evidence and the associated evidential weighting.

An additional benefit to training stems from illumination of “professional judgment.” Weston (2004), in his account of what skills are needed by CSEs, makes specific mention of professional judgment, but little explanation is provided as to what this is or how it is developed. Our findings, although preliminary, show that it is possible to postulate three key elements that are needed to create training that stimulates the development of professional judgment. First, the simulated environment needs to be sufficiently resourced with background items to challenge the relevance of the objects within its space. Second, the type of forensic samples distributed within the environment need to have varying levels of evidential value in line with the scenario. Third, instructional staff need to be aware of the types of factors constituting expert performance we have helped to identify to implement training that promotes development. Conversely, allowing experts to reflect on their crime scene assessment and examination is also useful in developing their own skills.

Key Points

In comparison to novices, expert crime scene examiners will pay more attention to the evidential value of objects in a crime scene and use this to frame their search.

Both novice and expert crime scene examiners will use the modus operandi that they infer a criminal will have used to make sense of the scene.

The expert crime scene examiner situates his or her work in the context of the broader criminal investigation process, for example, in considering how the material that he or she recovers could be used by other investigators.

Footnotes

Chris Baber gained a BA(Hons) in psychology and English at Keele University in 1987 and a PhD in speech technology at Aston University before joining the University of Birmingham in 1990. He originally joined the University of Birmingham to lecture on the MSc Work Design and Ergonomics course and then, in 2001, moved to the School of Electronic, Electrical and Computer Engineering (where he is currently head of school). He was promoted to reader in interactive systems design in 2004 and chair in pervasive and ubiquitous computing in 2008. His research focuses on the potential human factors issues related to sensor-based and mobile technology for field work.

Mark Butler is a former assistant senior scenes of crime officer with Leicestershire Constabulary and has experience in a wide spectrum of criminal investigations from volume crime right through to serious and organized crime, including murder and terrorist investigation. In 2003 he subsequently took up a position at the National Training Center (now National Police Improvement Agency) as a crime scene instructor. During his two and a half years as an instructor he spent a period of time as a team leader responsible for initial crime scene training. He is a member of the Institution of Fire Engineers, with fire investigation being a particular interest of his. He has been a senior lecturer in crime scene science for the past 2 years at the University of Teesside and is due to start a project examining the expertise of crime scene investigators.

References

Abernethy

Neal

R. J.

Koning

(1994). Visual-perceptual and cognitive differences between expert, intermediate and novice snooker players. Applied Cognitive Psychology, 8, 185–211.

Association of Chief Police Officers. (2002). Investigation of volume crime manual. London, UK: Author.

Baber

(2010). Distributed cognition at the crime scene. AI and Society, 25, 423–432.

Cellier

J.-M.

Eyrolle

Marine

(2001). Expertise in dynamic environments. Ergonomics, 40, 28–50.

Charlton

Fraser-Mackenzie

Dror

I. E.

(2010). Emotional experiences and motivating factors associated with fingerprint analysis. Journal of Forensic Sciences, 55, 385–393.

Charness

(1979). Components of skill in bridge. Canadian Journal of Psychology, 33, 1–16.

Chase

W. G.

Simon

H. A.

(1973a). The mind’s eye in chess. In Chase

W. G.

(Ed.), Visual information processing (pp. 215–281). New York, NY: Academic Press.

Chase

W. G.

Simon

H. A.

(1973b). Perception in chess. Cognitive Psychology, 1, 55–81.

de Groot

A. D.

(1965). Thought and choice in chess. The Hague, Netherlands: Mouton.

10.

Didierjean

Fernand

(2008). Sherlock Holmes—An expert’s view of expertise. British Journal of Psychology, 99, 109–125.

11.

Dror

Charlton

(2006). Why experts make errors. Journal of Forensic Identification, 56, 600–616.

12.

Dror

I. E.

Péron

Hind

Charlton

(2005). When emotions get the better of us: The effect of contextual top-down processing on matching fingerprints. Applied Cognitive Psychology, 19, 799–809.

13.

Eccles

Walsh

Ingledew

(2006). Visual attention in orienteers at different levels of experience. Journal of Sports Sciences, 24, 77–87.

14.

Ericsson

K. A.

Lehmann

A. C.

(1996). Expert and exceptional performance: Evidence of maximal adaptation to task. Annual Review of Psychology, 47, 273–305.

15.

Ericsson

K. A.

Simon

H. A.

(1993). Protocol analysis: Verbal reports as data (2nd ed.). Cambridge, MA: MIT Press.

16.

Ericsson

K. A.

Smith

(1991). Towards a general theory of expertise: Prospects and limits. Cambridge, UK: Cambridge University Press.

17.

Fox

M. C.

Ericsson

K. A.

Best

(2011). Do procedures for verbal reporting of thinking have to be reactive? A meta-analysis and recommendations for best reporting methods. Psychological Bulletin, 137, 316–344.

18.

Gale

Marsden

(1983). Medical diagnosis: From student to clinician. Oxford, UK: Oxford University Press.

19.

Groves

(2008). Clinical reasoning and medical expertise: The development of diagnostic skill in medical students. Saarbrucken, Germany: VMD Verlag.

20.

Horn

Masunaga

(2006). A merging theory of expertise and intelligence. In Ericsson

K. A.

Charness

Feltovich

P. J.

Hoffman

R. R.

(Eds.), The Cambridge handbook of expertise and expert performance (pp. 600–604). Cambridge, UK: Cambridge University Press.

21.

Joseph

G. M.

Patel

V. L.

(1990). Domain knowledge and hypothesis generation in diagnostic reasoning. Medical Decision Making, 10, 31–46.

22.

Kassirer

J. P.

Gorry

G. A.

(1978). Clinical problem solving: A behavioral analysis. Annals of Internal Medicine, 18, 247–248.

23.

McRobert

A. P.

Williams

A. M.

Ward

Eccles

D. W.

(2009). Tracing the process of expertise in a simulated anticipation task. Ergonomics, 52, 464–483.

24.

Medin

Edelson

(1988). Problem structure and the use of base-rate information from experience. Journal of Experimental Psychology: General, 117, 68–85.

25.

Omodei

Wearing

McLenna

(1997). Head-mounted video-recording: A methodology for naturalistic decision making. In Flin

Salas

Strub

Martin

(Eds.), Decision making under stress: Emerging themes and application (pp. 137–146). Avebury, UK: Ashgate.

26.

Patel

V. L.

Arocha

J. F.

Kaufman

D. R.

(1994). Diagnostic reasoning and medical expertise. In Medin

(Ed.), The psychology of learning and motivation (Vol. 31, pp. 187–252). San Diego, CA: Academic Press.

27.

Raab

Johnson

J. G.

(2007). Expertise-based differences in search and option-generation strategies. Journal of Experimental Psychology: Applied, 13, 158–170.

28.

Santilla

Korpela

Häkkänen

(2004). Expertise and decision-making in the linking of car crime series. Psychology, Crime and Law, 10, 97–112.

29.

Schraagen

J. M.

Leijenhorst

(2001). Searching for evidence: Knowledge and search strategies used by forensic scientists. In Salas

Klein

(Eds.), Linking expertise and naturalistic decision making (pp. 263–274). Mahwah, NJ: Lawrence Erlbaum.

30.

Smith

P. A.

Baber

Hunter

Butler

(2008). Measuring team skills in crime scene examination: Exploring ad-hoc teams. Ergonomics, 51, 1463–1488.

31.

Weston

(2004). The crime scene. In White

P. C.

(Ed.), Crime scene to court: The essentials of forensic science (2nd ed., pp. 26–35). Cambridge, UK: Royal Society of Chemistry.