Forensic Investigation: Fallacies in Forensic Sciences

The term "forensic" is utilised as a description of something that is linked to a crime. In every crime, there is a perpetrator. The current legal system's aim is to assign a perpetrator to a crime and ensure that an appropriate punishment is carried out. In order for a crime to be thoroughly explored, physical evidence needs to be collected from a crime scene and analysed. Hence, crime investigation is dependent on the forensic sciences (Olivier Delémont et al., 2014).

Forensic sciences are usually viewed as a single discipline. Yet, it can be more correctly perceived as a collection of sub-specialities that employ knowledge to handle evidence obtained at a crime. The forensic sciences are extremely varied, ranging from ballistics to forensic botany, and yet, they are furthermore segmented by a large volume of sub-fields (Morgan, 2019). Forensic scientists make use of the scientific method, which is grounded in problem-solving and the search for evidence to support observations and hypotheses. However, there are aspects of the field that are questionable in regard to how scientific they truly are (Tibbett, 2020).

Figure 1: Forensic scientist observing crime scene evidence (American Academy of Forensic Sciences, n.d.).

Hence, to evaluate the processes used in forensic investigations, it is essential to give a general overview of the techniques employed in forensic sciences and address their limitations. This article will specifically discuss fingerprinting and DNA analysis. Although the forensic sciences are often criticised, the article will acknowledge that they are imperative for crime investigations and will propose directions for how the field can advance in the future.

Current Forensic Techniques and Their Limitations

Fingerprint Analysis

Fingerprints are one of the main forms of evidence that can be collected at a crime scene. They are valuable in forensic investigations because they allow one to differentiate between multiple suspects and identify the perpetrator. This is because fingerprints are viewed as unique and personalised, meaning the print left by one person will not be identical to the fingerprint of any other person. Three types of fingerprint evidence exist: patent, latent and plastic prints. Patent prints are the most obvious as they can be observed by the naked eye, such as when a print has been left in blood or on a glass pane. Latent prints are what forensic scientists most commonly encounter. These are fingerprints that are not visible to a person in the absence of procedures that would make the print detectable, like the use of powder or fluorescent chemicals. Finally, plastic prints are left as impressions in soft, mouldable materials, like clay or wax (Yamashita et al., 2010). When identified at a crime scene, fingerprint collection will depend on the surface upon which they are found. A method may prove ineffective when collecting fingerprints from a soft, absorbent surface (such as paper) but can be ideal for non-porous materials, like plastic. Once the prints have been collected, with the use of protective measures to avoid contamination, the evidence is transferred to a laboratory for analysis (Dhaneshwar et al., 2021).

Figure 2: Fingerprint collection and analysis (Gibb & Riemen, 2023).

The analysis stage is heavily criticised for the lack of scientific foundation and standardisation. An analyst will first assess if the fingerprint harvested at the crime scene is viable for analysis. This is an integral step, as fingerprints very often can be incomplete, smudged or faint. This is because the transfer and durability of a fingerprint are dependent on multiple factors. The pressure that was applied by the finger to leave the print; the surface on which the print was left and the environmental conditions of the crime scene can all impact the integrity of a fingerprint. With so many external factors affecting the quality of the evidence, the accuracy of fingerprinting techniques is disputed (Gibb & Riemen, 2023).

If a suitable fingerprint is identified, the analyst will compare the sample print to a fingerprint set from a suspect or run it through the Automated Fingerprint Identification System (AFIS). The AFIS is a digital program that can search a database of fingerprints to find the ones that have similar characteristics to the sample print. The system can present an analyst with hundreds of prints that share features of the reference fingerprint. The analyst then goes on to manually compare the fingerprint collected at the crime scene to the top results presented by the system, or to the fingerprint set of a suspect (Gibb & Riemen, 2023). Fingerprint analysis is a step-wise process (involving observation at three levels) that requires searching for patterns and specific identification markers. The fingerprint would first be looked at as a complete specimen, eliminating prints that do not share the general, crude features of the reference print. After, the fingerprints would be examined for more specific and less obvious signs, such as the presence of a bifurcation or scars (Dhaneshwar et al., 2021) [refer to Figure 3].

Figure 3: Fingerprint features used for analysis (Dhaneshwar et al., 2021).

Yet, fingerprinting analysis is a flawed procedure. Primarily, fingerprint analysis cannot be considered a scientific procedure. It is a subjective process that relies on how the examiner perceives a reference fingerprint. The way an analyst approaches the examination can be heavily biased if they are made aware of certain details of the crime. For example, knowing that the suspect has been previously convicted of a similar crime may influence the analyst to match the fingerprint from the crime scene to the fingerprint of the suspect (Edmond et al., 2014). Additionally, studies have shown that errors can occur when analysts prioritise the first few fingerprints listed by the AFIS, being unconsciously more likely to match them to the reference fingerprint (Gibb & Riemen, 2023). Furthermore, there is no universally agreed threshold for deeming two fingerprints as being a match. In other words, finding seven markers in two fingerprints can constitute a match in one country while in another country, fifteen points need to be similar between fingerprints for a match to be made (Ezegbogu & Omede, 2022). The lack of standardisation places fingerprint analysis under scrutiny. This is because there is no scientific evidence to indicate a point at which two fingerprints can reliably be identified to come from one source (Stevenage & Pitfield, 2016). Despite all this, studies continue to show that fingerprinting analysis mostly yields reliable results, with false positives (when a match is made without there being one) and false negatives (when an actual match is missed) being minimal. However, a single error in the forensic sciences can have detrimental consequences, either convicting an innocent person or allowing a perpetrator to avoid punishment (Morgan, 2023). Therefore, it is important to standardise fingerprinting analysis, develop techniques to avoid bias and find scientific evidence to support the technique.

DNA analysis

Deoxyribonucleic acid (DNA) is the genetic material that instructs the activity of cells. Each half of DNA is randomly inherited from both parents and hence, the product is an entirely new and unique combination of genetic code. Therefore, like fingerprints, DNA can be a way to identify and link a person to a crime. Biological materials from an individual left at a crime scene, like hair, blood, saliva and skin, can all act as sources of DNA. Genes, sequences that code for the synthesis of certain proteins, are comprised of sections of DNA. Yet, a large proportion of DNA does not encode anything (Lakshmi et al., 2021). Short tandem repeats (STR) can be examples of non-coding DNA regions. Although they can be perceived as "junk" DNA, STRs are imperative in forensic investigations because they are the few parts of the DNA that differ between individuals. STRs can be found at different points along the DNA and thirteen of these areas (loci) are used in forensic DNA analysis (Norrgard, 2008). As STRs are made up of repeating DNA sequences, the number of times that a sequence repeats itself can be different at each of the thirteen sites, creating a distinct combination of STRs for each person (Jordan & Mills, 2021; Udogadi et al., 2020).

Figure 4: Short Tandem Repeats (STRs) (Udogadi et al., 2020).

For DNA analysis to occur, it must first be extracted from biological samples found at a crime scene. However, it often happens that the extraction process yields an inconsequential amount of DNA, an amount that cannot be reliably analysed. Hence, regions of DNA that consist of the STRs can be amplified using molecular techniques, such as the polymerase chain reaction (PCR). The genetic material can then be visualised on gels using fluorescent dyes with a technique known as gel electrophoresis (Jordan & Mills, 2021). The results are digitised so that the STRs extracted from DNA found at the crime scene can be compared with STRs from databases or from samples provided by suspects. The number of repeats occurring at each STR locus has a certain frequency at which they can appear in a population. A statistical analysis is then performed to derive the probability of the examined DNA profiles coming from the same person (Udogadi et al., 2020).

Although DNA analysis is the gold-standard technique in the forensic sciences, its reliability and efficacy can still be doubted. DNA can be found in any biological sample left at a crime scene. Yet, it is common for the quality of the evidence to be hindered by environmental factors or for it to be present in minute quantities (Lakshmi et al., 2021). If only a small amount of DNA is available, even after amplification, researchers may be forced to use only four or five STR loci for the analysis, rather than the preferred thirteen loci. This has a negative impact on the evidence, making it statistically less powerful and necessitating other forms of evidence to connect a perpetrator to a crime (Udogadi et al., 2020). Moreover, there is the risk that DNA evidence can become contaminated. At a crime scene, biological samples of the perpetrator can be mixed with those of the victim, a witness or an individual who was never at the scene (indirect transfer). Hence, when DNA is analysed, the single, complete sample may in fact be a combination of two DNA samples from entirely different individuals. This error may lead to a unique DNA sequence being examined, which may actually be traced to a person who is unrelated to the crime (van Oorschot et al., 2021).

Furthermore, fallacies can occur during DNA analysis if appropriate protective measures are not taken. There have been cases of false incriminations caused by DNA evidence contaminating physical evidence once in the laboratory (Sai, 2022). Finally, if a suspect is not detained and if the perpetrator has not committed any previous crime, there could be no DNA samples in a database that can match the one found at a crime scene. Such circumstances may render DNA evidence useless until further progress in the forensic investigation has been made. Though this technique has limitations, it is considered to be the most reliable and accurate forensic methodology. However, it is of great importance that DNA evidence is not used to override other forms of evidence due to the misconception that it is a procedure with perfect accuracy (Gill, 2012).

Figure 5: Gel electrophoresis being used to compare DNA profiles (Kramer, 2023).

Future of Forensic Investigation

Despite being an area of criticism, the forensic sciences continue to expand and garner novel methods capable of giving further insights into crimes. Additionally, current techniques come under constant reevaluation, with studies being performed to define ways that could facilitate evidence collection and eliminate the issue of bias.

Improving fingerprint analysis

A gap remains in the scientific evidence that backs up the use of fingerprint analysis in the forensic sciences but recent developments have shown ways of advancing the technique. Currently, latent fingerprints (which are invisible at the crime scene) require extensive manipulation for them to be visualised. This usually consists of a manual process, which can be prone to human error and require the use of toxic substances. The final result may often lack clarity and be distorted by the chemical processes, preventing third-level details from being observed in the fingerprint analysis (Yamashita et al., 2010). Researcher Wang and colleagues propose the use of carbon dot powder or solution to create bio images of latent fingerprints. Carbon dots are photo-luminescent materials. They are able to absorb specific wavelengths of light and emit different wavelengths, producing various colours (Jacques et al., 2007). The researchers found that by using carbon dots, red, orange and yellow images of latent fingerprints could be produced. The images had high contrast and could be produced from various surfaces, allowing third-level details (like pores) to be analysed (Wang et al., 2022).

A study conducted by researcher Costa and colleagues portrayed the utility of mass spectrometry imaging in the analysis of fingerprints in drug-related crimes (Costa et al., 2021). Mass spectrometry is a powerful tool that can detect the identity of molecules by considering their charge and mass (Huang et al., 2022). In the study, mass spectrometry imaging was able to map the location of cocaine (and its metabolite) on fingerprints. The researchers concluded that the technique could help distinguish between fingerprints that belonged to people who used cocaine and those who had only come into contact with the drug through touch. This was because a much higher concentration of cocaine’s metabolite, which would have been excreted through sweat, was found in the fingerprint ridges of cocaine users, and not in those who had accidental contact with it. This shows that combining scientific techniques (mass spectrometry imaging) with fingerprint analysis can help gather more information about a crime scene (Costa et al., 2021).

Figure 6: The use of carbon dots (CDs) to visualise latent fingerprints (Verhagen & Kelarakis, 2020).

Cognitive biases can affect the performance of fingerprint analysts and alter the outcome. Therefore, to avoid this bias from occurring, it is important for individuals who examine crime scene evidence not to overindulge in information about a crime. This can involve not giving away details about a suspect to the fingerprint analyst and having analysts work independently of the police force, to minimise the influence that comes from individuals working closely with the crime investigation (Edmond et al., 2014). An experiment was conducted with 36 expert fingerprint analysts in Australia. The researchers found that when examiners were working as a group and had to make a collective decision about a fingerprint match, the error rate fell by up to 12%. This suggests that shared decision-making could make the process of fingerprint analysis more objective, in turn improving the reliability of the results (Tangen et al., 2020).

Proteomics

Unlike genetic analysis of DNA, proteomics focuses on the study of the proteins that make up an organism. Proteins are comprised of chains of amino acids, which are themselves synthesised from specific sequences of DNA (Sacco & Aquila, 2023). DNA analysis is valuable in the identification of a perpetrator, but it may not always assist in the reconstruction of the events of a crime. In contrast, protein analysis can help biological samples to be traced back to specific regions and tissues in the body (Wilke, 2021). In a study conducted in China, researchers determined various biomarkers that could allow hair samples to be differentiated by their origin (scalp, underarm or pubic). This information can be imperative in deciding the context of the crime as pubic hairs would be more indicative of a sexual crime than hairs derived from the scalp (Xiao et al., 2023).

Researcher Bonicelli and colleagues have also suggested the utility of proteomics in the determination of the time of death. Various proteins had been identified in cadavers that could act as a biological clock, giving an insight into the time of a person’s death. Albumin is one such protein; its level in bones was found to diminish as the time after death increased (Bonicelli et al., 2022). Although proteomics is a relatively new research field in the context of the forensic sciences, it has great potential in aiding crime investigations. Future research needs to focus on testing the reliability of proteomics and peer-reviewing the techniques used (Sacco & Aquila, 2023).

Figure 7: Proteomic analysis (Díaz Martín et al., 2019).

Conclusion

The forensic sciences have faced scrutiny for utilising methods that lack scientific basis, for being subjected to cognitive biases and for having a high risk of human error. Fingerprinting and DNA analysis are the two most practised techniques in forensic investigations. Fingerprint analysis is largely subjective and does not have a standardised universal guideline. However, future research into the scientific basis of why fingerprints are unique and the use of quantitative methods in the analysis can help the forensic sciences progress. DNA analysis may often be perceived as infallible but contamination of DNA evidence can lead to false positive results. As well as enforcing protective measures in the laboratory, future studies should focus on developing methods for evidence collection that eliminate human error and detect indirect DNA transfers. Proteomics may become integral to the forensic sciences so it is important that proper research and funding is dedicated towards this area of study.

Bibliographical References

Bonicelli, A., Di Nunzio, A., Di Nunzio, C., & Procopio, N. (2022). Insights into the Differential Preservation of Bone Proteomes in Inhumed and Entombed Cadavers from Italian Forensic Caseworks. Journal of Proteome Research, 21(5), 1285–1298. https://doi.org/10.1021/acs.jproteome.1c00904

Costa, C., Jang, M., de Jesus, J., Steven, R. T., Nikula, C. J., Elia, E., Bunch, J., Bellew, A. T., Watts, J. F., Hinder, S., & Bailey, M. J. (2021). Imaging mass spectrometry: a new way to distinguish dermal contact from administration of cocaine, using a single fingerprint. The Analyst, 146(12), 4010–4021. https://doi.org/10.1039/d1an00232e

‌

Dhaneshwar, R., Kaur, M., & Kaur, M. (2021). An investigation of latent fingerprinting techniques. Egyptian Journal of Forensic Sciences, 11(1). https://doi.org/10.1186/s41935-021-00252-4

Edmond, G., Tangen, J. M., Searston, R. A., & Dror, I. E. (2014). Contextual bias and cross-contamination in the forensic sciences: the corrosive implications for investigations, plea bargains, trials and appeals. Law, Probability and Risk, 14(1), 1–25. https://doi.org/10.1093/lpr/mgu018

‌

Ezegbogu, M. O., & Omede, P. I.-O. (2022). The admissibility of fingerprint evidence: An African perspective. Canadian Society of Forensic Science Journal, 1–19. https://doi.org/10.1080/00085030.2022.2068404

‌Gibb, C., & Riemen, J. (2023). Toward better AFIS practice and process in the forensic fingerprint environment. Forensic Science International: Synergy, 7, 100336. https://doi.org/10.1016/j.fsisyn.2023.100336

Gill, P. (2012). Misleading DNA Evidence: Reasons for Miscarriages of Justice. International Commentary on Evidence, 10(1). https://doi.org/10.1515/ice-2014-0010

‌

Huang, L., Nie, L., Dai, Z., Dong, J., Jia, X., Yang, X., Yao, L., & Ma, S. (2022). The application of mass spectrometry imaging in traditional Chinese medicine: a review. Chinese Medicine, 17(1). https://doi.org/10.1186/s13020-022-00586-8

‌

Jacques, L., Maruyama, S., & Finnie, P. (2007). Photoluminescence: Science and Applications. Topics in Applied Physics, 287–319. https://doi.org/10.1007/978-3-540-72865-8_9

‌

Jordan, D., & Mills, D. (2021). Past, Present, and Future of DNA Typing for Analyzing Human and Non-Human Forensic Samples. Frontiers in Ecology and Evolution, 9. https://doi.org/10.3389/fevo.2021.646130

‌Lakshmi, J. B., Tejasvi, M. L. A., Avinash, A., P., C. H., Talwade, P., Afroz, M. M., Pokala, A., Neela, P. K., Shyamilee, T. K., & Srisha, V. (2021). DNA Profiling in Forensic Science: A Review. Global Medical Genetics, 8(4). https://doi.org/10.1055/s-0041-1728689

‌

Morgan, J. (2023). Wrongful convictions and claims of false or misleading forensic evidence. Journal of Forensic Sciences, 68(3). https://doi.org/10.1111/1556-4029.15233

‌

Morgan, R. M. (2019). Forensic science. The importance of identity in theory and practice. Forensic Science International: Synergy, 1, 239–242. https://doi.org/10.1016/j.fsisyn.2019.09.001

Norrgard, K. (2008) Forensics, DNA fingerprinting, and CODIS. Nature Education 1(1):35 https://www.nature.com/scitable/topicpage/forensics-dna-fingerprinting-and-codis-736/#

Olivier Delémont, Lock, E., & Olivier Ribaux. (2014). Forensic Science and Criminal Investigation. Springer EBooks, 1754–1763. https://doi.org/10.1007/978-1-4614-5690-2_145

Sacco, M. A., & Aquila, I. (2023). Proteomics: A New Research Frontier in Forensic Pathology. International Journal of Molecular Sciences, 24(13), 10735. https://doi.org/10.3390/ijms241310735

‌

Sai, Y. (2022). Advancing forensic science: Addressing challenges and embracing emerging technologies. Forensic Science Today, 8(1), 001–005. https://doi.org/10.17352/fst.000023

‌

Stevenage, S. V., & Pitfield, C. (2016). Fact or friction: Examination of the transparency, reliability and sufficiency of the ACE-V method of fingerprint analysis. Forensic Science International, 267, 145–156. https://doi.org/10.1016/j.forsciint.2016.08.026

‌

Tangen, J. M., Kent, K. M., & Searston, R. A. (2020). Collective intelligence in fingerprint analysis. Cognitive Research: Principles and Implications, 5(1). https://doi.org/10.1186/s41235-020-00223-8

‌

Tibbetts , J. H. (2020). Is Forensic Science Scientific? Crime lab errors and privacy issues raise concerns. BioScience, 70(5), 377–382. https://doi.org/10.1093/biosci/biaa022

‌

Udogadi, N. S., Abdullahi, M. K., Bukola, A. T., Imose, O. P., & Esewi, A. D. (2020). Forensic DNA Profiling: Autosomal Short Tandem Repeat as a Prominent Marker in Crime Investigation. Malaysian Journal of Medical Sciences, 27(4), 22–35. https://doi.org/10.21315/mjms2020.27.4.3

van Oorschot, R. A. H., Meakin, G. E., Kokshoorn, B., Goray, M., & Szkuta, B. (2021). DNA Transfer in Forensic Science: Recent Progress towards Meeting Challenges. Genes, 12(11), 1766. https://doi.org/10.3390/genes12111766

Wang, R., Huang, Z., Ding, L., Yang, F., & Peng, D. (2022). Carbon Dot Powders with Cross-Linking-Based Long-Wavelength Emission for Multicolor Imaging of Latent Fingerprints. ACS Applied Nano Materials, 5(2), 2214–2221. https://doi.org/10.1021/acsanm.1c03901

Wilke, C. (2021). Proteomics Offers New Clues for Forensic Investigations. ACS Central Science, 7(10), 1595–1598. https://doi.org/10.1021/acscentsci.1c01232

‌Xiao, D., Chen, J., Xu, L., Zhou, C., Mei, S., Qiu, Q., Xie, Q., & Liu, Y. (2023). Protein Biomarkers for the Identification of Forensically Relevant Human Hair from Different Body Parts in Intimate Contact Cases. Journal of Proteome Research, 22(7), 2391–2399. https://doi.org/10.1021/acs.jproteome.3c00129

‌

Yamashita, B., French, M., Bleay, S., Cantu, A., Inlow, V., Ramotowski, R., Sears, V., & Wakefield, M. (2010). Latent Print Development. https://www.ojp.gov/pdffiles1/nij/225327.pdf

Visual Sources