Profiling the profiler : exploring the role of cues in the development of expertise within the domain of offender profiling

Abstract

Cues have been identified as important precursors to successful decision-making amongst expert practitioners. Consequently, the use of expert cues in novice training programs is an attractive option for skill development. This thesis adopted the naturalistic approach to decision-making theory and research to investigate potential differences in cue activation (and presumably use) across two stages of expertise (expert and novice). The aim of the research was to determine whether it is possible to improve novice decision-making performance based on a modelling of expert cue activation. The research used the practice of Offender Profiling as a domain for investigating cue-use across expertise and involved three studies. Study 1 employed an interview strategy to identify a number of potential concepts of interest to expert and novice profilers when formulating decisions. Using content and thematic analyses, the findings yielded a number of crime-related and offender-related concept labels to be used as a basis for discriminating between expert and novice cue activation in Study 2. Study 2 incorporated two phases of research, the first of which (Study 2a) presented pairs of the concept labels identified in Study 1 as part of a Paired-Concept Association Task (P-CAT), which recorded participants' response latency in recognising associations between the concepts. The second phase (Study 2b) involved the distribution of a survey to further test participants' perceptions of the associations between concepts in relation to their frequency of use, strength, diagnosticity, and domain specificity. The findings of Study 2 revealed that the P-CAT was able to accurately discriminate between expert and novice cue activation and yielded a number of cue-based associations that were recognised/activated consistently and rapidly by, predominantly, the expert profilers. Further, the experts' perceptions of the associations reflected their response latencies. However, novice perceptions did not differentiate between associations, suggesting that the ability to discriminate between cues is a function of expertise. Finally, Study 3 investigated whether novice acquisition of the expert-cue-based associations was beneficial to their decision-making performance. Two phases were reported, the first of which (Study 3a) employed an experimental training program, Associated Concepts Exposure (ACE), to aid na ve participants' acquisition of a number of expert cue-based associations. Study 3b, again, used ACE training; this time with novices. Additionally, this phase was designed to determine whether potential improvements in novices' recognition of expert associations were matched by concomitant improvements in several facets of decision-making performance, including decision accuracy, time, and information acquisition. Finally, this phase sought to determine whether potential improvements in novice performance were impacted by the extent to which the decision scenarios engaged reflected a naturalistic scenario. Study 3b employed a decision assessment interface which functioned as a surrogate for the operational environment, enabling users to explore, acquire, and integrate cue-based information in quasi-realistic profiling decision tasks. The interface was administered prior to, and following, ACE training. The outcomes of Study 3 revealed that a set of expert cue-based associations could be acquired by both na ve and novice participants using ACE training. Moreover, exposing novice profilers to a set of expert cue-based associations improved their decision-making performance in a profiling decision task. However, the results also revealed that the novices were unable to improve their decision performance within a task consistent with a naturalistic scenario. It was concluded that, although the acquisition of valid cue-based associations is vital to the development of expertise, a cognitive gap remains between expert and novice performance that cue-use alone cannot account for. Ultimately, the key to closing this gap appears to hinge on a number of additional cognitive skills (e.g., the construction of mental models based on the retrieval of decision cases from long-term memory) which many decision-makers will most likely acquire as a result of extensive domain-specific operational experience. The primary limitation associated with each of the three studies detailed was the predominant use of text-based descriptions to represent cue-based information. Future studies may seek to explore more naturalistic representations. The thesis findings have implications for a number of key areas, including; knowledge elicitation strategies, cue-based training, offender profiling, and the naturalistic decision-making paradigm.

Date of Award	2010
Original language	English