Template Ageing and Longitudinal Verification in Fixed-Text Keystroke Dynamics: A Subject-Disjoint Study Across Eight Weeks

Simon Parkinson; Saad Khan; Na Liu; Qing Xu

cs.CRSep 24, 2026

Template Ageing and Longitudinal Verification in Fixed-Text Keystroke Dynamics: A Subject-Disjoint Study Across Eight Weeks

Authors: Simon Parkinson, Saad Khan, Na Liu, Qing Xu

Organizations: Department of Computer Science, University of Huddersfield, HD1 3DH Huddersfield, U.K. · College of Intelligence and Computing, Tianjin University, Tianjin, China

Abstract

Behavioural biometric templates are widely believed to degrade as the gap between enrolment and verification grows, but few studies measure this template ageing effect directly under controlled conditions. We collected a longitudinal dataset of 40 fixed passwords, each typed four times per weekly session over eight consecutive weeks. We compare a scaled-Manhattan matcher (M1), a gradient-boosted classifier (M2), a TypeNet-style recurrent embedding model (M3), and a TypeFormer-style Transformer (M4) under a 5-fold subject-disjoint protocol and a design that jointly varies mechanism and the enrolment-to-query gap, from 0 to 7 weeks. Template ageing proves large and systematic. Error increases monotonically with the gap for every mechanism, from an EER of 14.6-27.2% at a gap of zero to 25.5-37.1% at seven weeks, or 1.7% of decision error per week elapsed (p < 0.001). However, the choice of mechanism matters more than its rate of ageing. Baseline accuracy spans 12.6 percentage points across the four mechanisms, the degradation each accumulates over seven weeks spans only 2.3 points, and ageing never reorders them. A matcher can therefore be chosen on same-session accuracy, with ageing managed by re-enrolment scheduling rather than by matcher selection. The two properties are nonetheless distinct, as M3 is the least accurate mechanism yet ages significantly more slowly than M1 under every specification tested. Training randomness also matters differently by architecture, with 58% of the recurrent model's fold-to-fold variance attributable to seed noise against 19% for the Transformer. Because the smaller ageing-rate differences are sensitive to modelling choices, while the accuracy differences and the ageing effect are not, we recommend that comparative ageing-rate claims be supported by seed-level score fusion, independent replication, and an alternative outcome-model specification.

Figures & tables

Study	Longitudinal (weeks+)	Cross-week template test	Learned embedding model	Subject-disjoint evaluation	Inferential statistics
Montalvão et al. [ 27 ]	$\times$	$\times$	$\times$	$\times$	$\times$
Acien et al. (TypeNet) [ 9 ]	$\times$ (free-text)	$\times$	✓	✓	$\times$
Parkinson et al. [ 12 ]	✓	$\times$	$\times$	$\times$	$\times$
Giot/Pisani (template update) [ 6 , 7 ]	$\sim$ (multi-session)	$\sim$ (adaptation active)	$\times$	$\times$	$\times$
Yang et al. [ 8 ]	✓ (drift-detection)	$\times$	$\times$	$\times$	$\times$
This paper	✓	✓	✓	✓	✓

TABLE I: Positioning relative to selected prior work (qualitative; ✓ = present,

\times

= absent,

\sim

= partially addressed).

Hyperparameter	Value
Hidden size	64
Embedding dimension	64
Encoder layers	1
Gaussian range-encoding bins ( $n_{\text{gaussians}}$ , M4 only)	24
Max duration for range encoding	3.0 s
Triplet margin	0.3

TABLE II: Architecture and optimisation hyperparameters for M3 (LSTM) and M4 (Transformer), fixed a priori and shared across folds and replicates.

Fig. 1: EER versus enrolment-to-query gap $\Delta$ (weeks), by mechanism and outlier-clipping condition, at $k{=}5$ enrolment sessions. Shaded bands are bootstrap 95% confidence intervals over test participants. Error increases monotonically with $\Delta$ for every mechanism.

Trials (all mechanisms): genuine / impostor / total
Mechanism	EER at $\Delta{=}0$	EER at $\Delta{=}7$
M1 (scaled Manhattan)	21.0 [19.7, 22.4]	33.1 [30.1, 36.0]
M2 (shallow ML)	17.2 [15.9, 18.5]	27.0 [23.7, 30.5]
M3 (TypeNet-style LSTM)	27.2 [25.6, 28.8]	37.1 [34.7, 39.5]
M4 (TypeFormer-style)	14.6 [13.5, 15.9]	25.5 [21.8, 29.8]
$\Delta{=}0$	23,827 / 352,560 / 376,387

TABLE III: EER (%) at

\Delta{=}0

and

\Delta{=}7

(

k{=}5

, clipped, M3/M4 three-seed fusion, with 95% bootstrap CI in brackets), together with the underlying genuine/impostor trial counts (identical across mechanisms).

Fig. 2: DET curves at $k{=}5$ , pooled across $\Delta$ , for the outlier-clipped (left) and unclipped (right) conditions.

Mechanism	FNMR @ FMR=1%	FNMR @ FMR=0.1%
M1 (scaled Manhattan)	91.5	98.6
M2 (shallow ML)	79.7	93.9
M3 (TypeNet-style LSTM)	95.2	99.3
M4 (TypeFormer-style)	86.4	97.8

TABLE IV: FNMR (%) at fixed FMR operating points,

k{=}5

, clipped, pooled across

\Delta

Term	Coef.	$z$	$p$
$\Delta$ (per week, centred)	$+0.017$	21.60	$<0.001$
Mechanism: M2 vs. M1	$-0.040$	$-18.30$	$<0.001$
Mechanism: M3 vs. M1	$+0.052$	$23.90$	$<0.001$
Mechanism: M4 vs. M1	$-0.065$	$-30.15$	$<0.001$
Length (per character, centred)	$-0.020$	$-11.27$	$<0.001$
Hardware-change (week 5)	$+0.005$	$2.08$	$0.037$

TABLE V: Mixed-effects model: selected fixed-effect estimates (main-effects model, no interaction).

N=299{,}984

, with participant and password variance components of 0.003 and 0.001 respectively.

	Linear probability model			Logistic (cluster-robust)
Mechanism vs. M1	Run 1	Run 2	Run 3	Run 1	Run 2	Run 3
M2	$-0.002$ (.029)	$-0.003$ (.010)	$-0.002$ (.030)	$-0.004$ (.652)	$-0.004$ (.650)	$+0.003$ (.713)
M3	$-0.002$ (.047)	$-0.004$ ( $<$ .001)	$-0.004$ (.001)	$-0.019$ (.032)	$-0.021$ (.016)	$-0.015$ (.056)
M4	$-0.000$ (.678)	$-0.001$ (.485)	$-0.000$ (.954)	$+0.016$ (.093)	$+0.016$ (.090)	$+0.024$ ( .008 )

TABLE VI:

\Delta\times

Mechanism interaction under the linear-probability and logistic (cluster-robust) specifications, across three independent end-to-end pipeline replicates. Cells show the coefficient, with the

p

-value in parentheses. In the linear specification, coefficients are expressed on the weekly probability-of-error scale per unit of

\Delta

, whereas in the logistic specification, coefficients are measured in log-odds. Consequently, only the direction and statistical significance of the coefficients, rather than their magnitudes, are directly comparable across the two model formulations. For the joint test in the linear model, the test statistics are

\chi^{2}(3)=7.27,\ 17.92,\ \text{and }16.22

, with corresponding

p

-values of

0.064,\ 0.0005,\ \text{and }0.0010

for Runs 1, 2, and 3, respectively.

Mechanism	Within-fold seed SD	Across-fold SD	Noise share
M3 (LSTM)	0.0181	0.0313	$\sim$ 58%
M4 (Transformer)	0.0062	0.0325	$\sim$ 19%

TABLE VII: Decomposition of fold-to-fold EER variance into training-seed noise (same data, different training run) versus total across-fold spread.

$\Delta$	Genuine trials	Impostor trials	Total
0	23,827	352,560	376,387
1	79,994	293,714	373,708
2	68,527	249,512	318,039
3	56,387	203,999	260,386
4	45,332	163,875	209,207
5	34,393	123,530	157,923

TABLE VIII: Genuine and impostor trial counts by

\Delta

(

k{=}5

, clipped; identical across mechanisms).

Mechanism	Condition	$\Delta{=}0$	$\Delta{=}1$	$\Delta{=}2$	$\Delta{=}3$	$\Delta{=}4$	$\Delta{=}5$	$\Delta{=}6$	$\Delta{=}7$
M1 (scaled Manhattan)	Clipped	21.0	24.2	25.7	27.0	28.3	29.3	30.7	33.1
M1 (scaled Manhattan)	Unclipped	24.0	26.3	27.7	28.8	30.0	31.2	32.2	34.3
M2 (shallow ML)	Clipped	17.2	20.6	22.1	23.5	24.5	25.2	25.6	27.0
M2 (shallow ML)	Unclipped	19.8	23.1	24.5	25.9	27.0	27.6	27.8	28.6
M3 (TypeNet-style LSTM)	Clipped	27.2	29.0	30.5	32.1	33.4	34.0	35.2	37.1
M3 (TypeNet-style LSTM)	Unclipped	21.4	24.1	25.7	27.3	28.9	30.1	31.8	34.1

TABLE IX: Full EER (%) by mechanism, clipping condition, and temporal gap

\Delta

k{=}5

Mechanism	$k$	$\Delta{=}0$	$\Delta{=}1$	$\Delta{=}2$	$\Delta{=}3$	$\Delta{=}4$	$\Delta{=}5$	$\Delta{=}6$	$\Delta{=}7$
M1 (scaled Manhattan)	1	28.9	31.4	32.3	33.2	34.0	34.6	35.0	36.5
	3	21.9	25.6	26.5	27.9	29.2	30.0	30.6	33.1
	5	21.0	24.2	25.7	27.0	28.3	29.3	30.7	33.1
M2 (shallow ML)	1	15.8	20.5	22.0	23.3	24.2	24.8	24.7	26.5
	3	16.1	20.7	22.1	23.5	24.8	25.2	25.6	26.9
	5	17.2	20.6	22.1	23.5	24.5	25.2	25.6	27.0

TABLE X: Full EER (%) by mechanism, enrolment size

k

, and temporal gap

\Delta

, clipped condition.

Mechanism	Estimator	$\Delta{=}0$	$\Delta{=}1$	$\Delta{=}2$	$\Delta{=}3$	$\Delta{=}4$	$\Delta{=}5$	$\Delta{=}6$	$\Delta{=}7$
M1 (scaled Manhattan)	Pooled	21.0	24.2	25.7	27.0	28.3	29.3	30.7	33.1
M1 (scaled Manhattan)	Particip.-balanced	20.9	24.2	25.6	26.7	28.2	29.4	30.8	32.2
M2 (shallow ML)	Pooled	17.2	20.6	22.1	23.5	24.5	25.2	25.6	27.0
M2 (shallow ML)	Particip.-balanced	16.2	19.2	20.7	21.9	22.9	23.8	24.2	26.0
M3 (TypeNet-style LSTM)	Pooled	27.2	29.0	30.5	32.1	33.4	34.0	35.2	37.1
M3 (TypeNet-style LSTM)	Particip.-balanced	27.0	28.6	30.2	31.6	32.8	33.7	34.6	37.7

TABLE XI: EER (%) at

k{=}5

, clipped: pooled-trial vs. per-participant-balanced estimator, all eight values of

\Delta

Fold	Mechanism	Seed 0	Seed 1	Seed 2
0	M3	29.7	35.2	29.6
0	M4	20.6	20.8	21.5
1	M3	31.0	34.5	33.7
1	M4	18.9	17.7	17.2
2	M3	35.5	32.5	32.6
2	M4	18.6	19.4	21.0

TABLE XII: Pooled EER (%) for three independently-trained (unfused) seeds of M3 and M4, by test fold.

Explore similar work

Position: Age Estimation Models Do Not Process Biometric Data
May 17, 2026Nikita MarshalkinBiometrics Age
Sumsub GmbH, Berlin, Germany.
ECG-biometrics-bench: A Unified Framework for Reproducible Benchmarking of ECG Biometrics
May 2, 2026Milad ParvanSmall-Sample Electroencephalogram Datasets Biometrics
Independent Researcher · Milan, Italy
Face Age Verification Vulnerabilities Under Simple Appearance Manipulations
Jul 27, 2026Ioannis Sarridis, Ioannis Kompatsiaris, Symeon PapadopoulosFace Recognition Model Appearance
Information Technologies Institute, Centre for Research and Technology Hellas, 6th km Charilaou-Thermi Rd, Thessaloniki, 57001, Greece

May 17, 2026, cs.CYJ/K move · Enter open · S save

Position: Age Estimation Models Do Not Process Biometric Data

Nikita Marshalkin

Sumsub GmbH, Berlin, Germany.

Biometrics Age Automatic Identification System Position

Open paper AskPDF arXiv

When a neural network estimates someone's age from a photograph, does it process biometric data? The answer depends on whether identity-discriminative representations arise within the network during inference, a question that may seem trivial to ML researchers but triggers consent requirements under GDPR, statutory damages under BIPA, or high-risk AI classification under the EU AI Act. Yet no regulatory guidance addresses it. This position paper provides empirical evidence: 14 models evaluated across 3 face verification benchmarks show age estimators fall orders of magnitude short of identification thresholds. Age estimation models cannot identify individuals. We call on researchers to provide transparency about what systems store and can do, and on regulators to distinguish transient processing from template storage.

Abstract

Explore similar work

Position: Age Estimation Models Do Not Process Biometric Data

ECG-biometrics-bench: A Unified Framework for Reproducible Benchmarking of ECG Biometrics

Face Age Verification Vulnerabilities Under Simple Appearance Manipulations