Dynamic signatures: A review of dynamic feature variation and forensic methodology

Does cognitive decline influence signing?

OBJECTIVE

The study explored the change in handwritten signature in neurodegenerative diseases by using of a rater-based approach.

METHODS

Four independent observers were required to compare a pair of signatures (on average, 5 years elapsed between the two signatures) made by 103 patients (mean age 72 years) with Alzheimer's disease (AD) or frontotemporal dementia (FTD) and by 31 healthy participants (HC; mean age 73 years), judging their change according to a 0-1 rating scale (0 = similar or 1 = different). If a signature change was detected, the rater had also to report which signature features (spatial layout, omitted/added/switched letters or names, shape of letter, pen-flow) changed on the same 0-1 scale. For the AD and FTD groups, one signature was collected prior to the diagnosis of dementia, the other subsequent.

RESULTS

A signature change was reported by raters in 36% of AD patients, 44% of FTD, and 17% of HC, with significant differences between both clinical groups and HC (vs. AD, p = .01; vs. FTD, p = .001). There was not a distinctive marker of the signature change (i.e., feature change) in patients with dementia. Moreover, the signature changes in neurological patients were unrelated to their clinical and demographic characteristics (age, sex, education, time elapsed between the two signatures, Mini-mental State Examination score).

CONCLUSION

The findings suggest a resistance of handwritten signature in neurodegenerative diseases and in physiological aging, also suggesting that the signature may be an unreliable indicator of the cognitive status in AD and FTD, at least if subjectively evaluated.

Bayesian evaluation of dynamic signatures in operational conditions

Forensic handwriting examiners (FHE) activities are focused on comparative analysis of handwritten objects such as signatures. Their role is to provide and evaluate evidence for and against the authenticity of a questioned signature. In recent years, cases involving handwritten signatures captured on electronic devices have become more commonplace. These so-called 'dynamic signatures' (also known as 'digitally captured signatures') are much different from paper-based signatures. Not only does the medium of recording differ, but also the type, volume of data and features are different from the pattern-based evidence that makes up paper-based signatures. Recent developments in forensic science - including signature examination - have led to the adoption of evaluative probabilistic methodologies in many disciplines [see, e.g. ENFSI 1915 Guidelines]. In the current paper, a probabilistic model to evaluate signature evidence in the form of multivariate data, as proposed and described in Wacom Europe GmbH (2019), is adopted. Topics like data sparsity, joint evaluation of multiple features and feature selection are investigated. Performed experimental studies showed an accuracy rate above 90% even when a limited number (5) of reference signatures was available. The performances of a multivariate approach are compared with those characterizing a so-called multiplicative approach where variables (features) are taken as independent and the Bayes' factor (BF) is obtained as the product of univariate BFs associated to each selected feature. The simplicity of this latter approach is, however, accompanied by severe issues about the reliability of results. The use of a multivariate approach is therefore highly recommended. Finally, the evidential values in correspondence of alternative feature sets are compared. Results suggest that discriminative features are writer-related and necessitate a case-specific selection.

Shifting forensic science focus from means to purpose: A path forward for the discipline?

Forensic science is facing a persistent crisis that is often addressed by organizational responses, with a strong focus on the improvement and standardisation of means and processes. However, organisations and processes are highly dependent on the political, economical and legal structures in which they operate. This may explain why most proposed solutions had difficulties in addressing the crisis up to now, as they could hardly be applied transversally to all forensic science models. Moreover, new tools and technologies are continuously developed by a quasi-infinite number of different scientific disciplines, thus leading to further diversity and fragmentation of forensic science. In this paper, it is proposed to shift the focus from means to purpose and consider forensic science current challenges in terms of discipline, before addressing organisations' specific issues. As a distinct discipline, forensic science can refocus research and development on shared principles and purposes, such as reconstructing, monitoring, and preventing crime and security issues. This focus change will facilitate a better understanding of the trace as the object of study of forensic science and eventually lead to a more impactful and long-lasting effect. This approach will also foster the development of a forensic science culture (instead of a primarily technological culture) unified by purpose rather than means through more relevant education and research.

The challenge of comparing digitally captured signatures registered with different software and hardware

Along with the growing popularity of electronic documents authorised with digitally captured signatures, such evidence has appeared in the work of forensic practitioners. Many different vendors offer signature pads with varying specifications. It is therefore expected that forensic handwriting experts will be called upon to compare questioned and known samples captured with completely or partially different hardware and software combinations. Such cases may be challenging as numerical handwriting data produced by various equipment may differ not only in the type of information captured and its quality, but also in its structure and coding. In this research, numerical data of handwriting - i.e. spatial coordinates, force, and time values - were acquired with 26 different combinations of hardware and software to study characteristics of their coding. The analysis of samples revealed that scaling of numerical data is not only hardware but also software dependent. Therefore, their compliance with the ISO/IEC 19794-7 standard is recommended to improve the data interoperability. This standard emphasizes the importance of supplementing numerical signature data with scaling ratios of the used signing solution. The paper also includes descriptions of several phenomena observed in the acquired data to highlight possible pitfalls in performing inter-solution comparisons in casework.

Interpol review of questioned documents 2016-2019

This review paper covers the forensic-relevant literature in questioned documents from 2016 to 2019 as a part of the 19th Interpol International Forensic Science Managers Symposium. The review papers are also available at the Interpol website at: https://www.interpol.int/content/download/14458/file/Interpol Review Papers 2019.pdf.

Bayesian multivariate models for case assessment in dynamic signature cases

Dynamic signatures are recordings of signatures made on digitizing devices such as tablet PCs. These handwritten signatures contain both dynamic and spatial information on every data point collected during the signature movement and can therefore be described in the form of multivariate data. The management of dynamic signatures represents a challenge for the forensic science community through its novelty and the volume of data available. Much as for static signatures, the authenticity of dynamic signatures may be doubted, which leads to a forensic examination of the unknown source signature. The Bayes' factor, as measure of evidential support, can be assigned with statistical models to discriminate between competing propositions. In this respect, the limitations of existing probabilistic solutions to deal with dynamic signature evidence is pointed out and explained in detail. In particular, the necessity to remove the independence assumption between questioned and reference material is emphasized.

Recommendations for capturing signatures digitally to optimize their suitability for forensic handwriting examination

The use of digitally captured signatures in everyday course of business increases annually and, compared to pen and paper signatures, provides various advantages concerning the administration of documents. These signatures may also become subjects of a forensic handwriting examination and, therefore, in order to optimize their suitability for this purpose, they should satisfy several requirements. This paper presents recommendations drawn up by forensic handwriting examiners associated with ENFHEX (ENFSI) in a project aimed at defining best practices in forensic examination of digitally captured signatures. The paper is dedicated mainly to hardware and software developers, providers, and user institutions of digitally captured signature technologies to improve their practice to a level optimized for forensic handwriting examination. The most important requirements outlined in this paper concern digitally captured signature data, hardware, and software used to acquire these data, as well as optimized signing conditions. Following these requirements ensures the suitability of signature data for forensic handwriting examination and, consequently, increases the reliability of the associated electronic documents. In spite of rapidly evolving technology, they can serve as a solid basis for understanding and consideration of the optimal use of digitally captured signatures for signing electronic documents.

Features of digitally captured signatures vs. pen and paper signatures: Similar or completely different?

The question of whether digitally captured signatures and conventional signatures executed with a pen on paper differ in their characteristics is of practical relevance for forensic handwriting examiners. Due to gaps in the current literature, the present research is dedicated to this issue. Eighty persons signed in three conditions: a) with a stylus on a pad, b) with an inking pen on a sticky note attached to a signature pad allowing to obtain a digital and an analogue version on paper of one and the same writing simultaneously, and c) with a pen on paper. The first step was to investigate to what extent the character shape and number of pen lifts differ between the digital and analogue representation of one and the same signature. This revealed minor differences which are due to technical characteristics of the devices used. The observed distortions are of minor practical relevance according to ratings by eight participating forensic handwriting examiners. Subsequently, signature characteristics were compared between the three different writing conditions in a casework-oriented way. Statistical multi-level models indicate significant differences between the three signature types, but minor effect sizes in most of the examined characteristics. From the point of view of the participating handwriting examiners, these factors do not fundamentally restrict the comparability between digitally captured and conventional signatures in practice. However, caution should be exercised when generalising the results, as several factors, such as the usage of different signature pads as well as signatures made with the finger instead of a stylus, could result in more important differences compared to pen and paper signatures.

The potential of collaborative learning as a tool for forensic students: Application to signature examination

Transferring theoretical knowledge to practical skills remains a big challenge in forensic science, especially in questioned documents. The examination of handwriting and signatures requires years of practice to develop the necessary skills. While students (and to some extent the general population) often have the impression that it is easy to differentiate handwriting from different persons, in practice, particularly when dealing with simulated signatures, there is a high risk of reaching a wrong conclusion when questioned document experts do not use a systematic approach and/or are not sufficiently experienced (see for example the famous French Dreyfus case). Thus, a novel teaching approach, based on collaborative learning, has been introduced in a theoretical handwriting class to improve the students' theoretical knowledge, and additionally make them aware of the limitations of their practical skills and give them tools to improve them in their future practice. Through five activities, the students took the roles of victims, forgers, teachers and experts and created their own learning materials (i.e. signatures and mock casework). During those interactive activities, they learned to describe their signature's characteristics, intra-variability and complexity, and thus evaluate their own signature's vulnerability (as potential victims). They learned techniques to simulate signatures and detect the resulting forgeries' characteristics (in the role of forgers). In the role of teachers, they prepared mock casework scenarios and gave feedback to their colleague's examination of the produced material. As experts, they carried out signature examination as they would in a proficiency test and were exposed to the difficulties an actual expert may encounter in practice. The evaluation of this novel teaching scenario was very positive, as students learned more extensively the possibilities and limitations of signature comparison. They were more active and motivated in their learning experiences. The teaching team also had an improved experience. Some students complained of an increased workload and imprecise instructions. Improvements were tested and are discussed in this paper.

On-Line Signature Partitioning Using a Population Based Algorithm

The on-line signature is a biometric attribute which can be used for identity verification. It is a very useful characteristic because it is commonly accepted in societies across the world. However, the verification process using this particular biometric feature is a rather difficult one. Researchers working on identity verification involving the on-line signature might face various problems, including the different discriminative power of signature descriptors, the problem of a large number of descriptors, the problem of descriptor generation, etc. However, population-based algorithms (PBAs) can prove very useful when resolving these problems. Hence, we propose a new method for on-line signature partitioning using a PBA in order to improve the verification process effectiveness. Our method uses the Differential Evolution algorithm with a properly defined evaluation function for creating the most characteristic partitions of the dynamic signature. We present simulation results of the proposed method for the BioSecure DS2 database distributed by the BioSecure Association.