Inter- and intra-individual variation in earprints Meijerman, L.

(1)

Citation

Meijerman, L. (2006, February 15). Inter- and intra-individual variation in earprints. Barge's Anthropologica, Leiden. Retrieved from https://hdl.handle.net/1887/4292

Version: Not Applicable (or Unknown)

License: Licence agreement concerning inclusion of doctoral thesis in the_{Institutional Repository of the University of Leiden}

Downloaded from: https://hdl.handle.net/1887/4292

(2)

EXPLORING THE EFFECT OF DURATION OF LISTENING

ON EARPRINTS

L.M eijerman,G.van Antwerpen,R.J.van M unster and G.J.R M aat

(3)

Abstract

(4)

6.1 Introduction

W hilst considering factors that might affect the appearance of an earprint, Dr. Michael Ingleby of the University of Huddersfield (UK) suggested that listening time, i.e., the duration of contact between the ear and the listening surface, could affect the appearance of a print (FearID meeting, Glasgow, February 2003; personal communication). The rationale behind it being that, as listening time increases, fidgeting of the ear may cause an increase in contact area.

The following is an account of a study of earprints to determine if variation in the duration of listening adds to the range of intra-individual variation in earprints. As we know that the force applied by the ear may influence the imprinted surface as well, this aspect of listening was included in the analysis.

6.2 Experimental design

Ten subjects each listened a minimum of seven times at the FearID listening box. All efforts by one subject were made using the same ear; it was left up to the subject to decide which ear to use. Several sound files of various lengths containing spoken word by a male voice were played to serve as target sound. The duration of listening for the seven listening efforts was set at 13, 15, 20, 25, 30, 35 and 45 seconds respectively. The order in which the sound files of various lengths were presented to the subject varied. For three subjects, an additional eighth listening effort that lasted 107 seconds was included; one subject performed an eighth listening effort of 85 seconds.

Five subjects performed all efforts in one continuous session. These are referred to as group B subjects. For the other five subjects – referred to as group A subjects – listening efforts were spread out over two days. The average time interval between listening efforts was 90 minutes for group A subjects. This was done to reduce a possible influence of a decrease in oily substances on the ear.

(5)

each effort was determined by calculating the mean of all recordings after omitting the first two and final two recordings24. Resulting prints were dusted with aluminium powder and lifted using Black Gel Lifters. Prints were scanned at 600 dpi and imported into the FearID Earprint Storage and Analysis System25 after colours had been inverted.

6.3 Analytical methods

Earprints of group A subjects were subjected to a detailed examination by eye. Digitised earprints of all subjects were further used to calculate print-mass. To do so, the FearID Earprint Storage and Analysis System was used to calculate boundaries of imprinted features. Contour boundaries were then manually improved to optimize comparison between prints. The mass of imprinted features was calculated taking both the size of the imprinted area as well as intensity (i.e., grey values of pixels) into account. Intensity was calculated both relative to a calculated intensity median, as well as relative to the background. The following formula was used:

M = - x,y(I(x,y)-B)

were M denotes print-mass, I(x,y) the intensity of a pixel in the imprinted area, and B either the intensity median or the intensity of a (white) background pixel.

We applied a univariate analysis of variance (GLM procedure), using SPSS (version 11.0) to explore the effect of the duration of listening on print-mass. As print-mass may also be influenced by the force that was applied by the ear to the surface during listening, this variable was included in the analysis. Functional force and peak value were analysed separately. Covariates were ‘duration of listening’ and ‘applied force’. ‘Subject’ was included as random factor as both ear-size and ear-coverage (i.e., imprinted surface) varied between subjects.

24

See Meijerman et al., 2005b (chapter 3 of this volume) and Meijerman et al., 2005c (chapter 4 of this volume) for an explanation of the use of both peak values and functional force values to characterise the force applied by the ear to the surface during eavesdropping. The rationale behind omitting force recordings to calculate the functional force of a listening effort is also explained in these chapters.

25

(6)

6.4 Results and discussion

6.4.1 Examination of details

Subject 1 (including additional eighth effort)

See Fig. 6.1 for selection of prints. ‘Fullest’ print with best details after listening 107s: the helix is fairly complete and includes the imprint of an auricular tubercle. In the majority of prints, a (small) part of the helix is missing. Besides the 107s print, only the 45s print and the 20s print have a continuous helix. In the 35s print, the helix is least imprinted. Applied functional force during these four efforts was 1445, 1707, 1700 and 1537g respectively; none of the prints exhibited an obvious shift of pressure distribution. The relative high force values might explain the continuous helix in the 45s and 20s prints. However, as the 107s print was made with relatively low force, its complete helix would support the possible effect of duration of listening.

The auricular tubercle is not always visible, but no apparent correlation to length of listening is observed. Other elements that are very characteristic in prints of this subject are the anthelix appendix, the oblique fold in the anthelix and the pre-auricular creases. The first two elements are visible in all prints; the creases are least apparent in the 107s print. It further appears that increased listening time increases contrast (thereby visibility of details). It has to be noted though that during the 13s and 15s prints, i.e., prints resulting from the shortest duration of ear-surface contact, applied force was comparatively low.

___________________________________________________________________________

___________________________________________________________________________ Fig. 6.1 Selection of prints of subject 1: 13s print (a), 35s print (b), 45s print (c) and 107s print (d).

(7)

See Fig. 6.2 for selection of prints. ‘Fullest’ print after listening 107s; particularly the anthelix surface increased in this print. Changes in imprinted surface are furthermore expressed by changes in the void between anthelix and antitragus. This void is smallest in the 107s print. There is, however, also a relatively small void in the 13s print, while it is relatively large in the 15s print. Though functional force of these efforts was relatively similar, the peak value was not. For the 13s effort it was relatively high (1471g) compared with the 15s effort (1087g). This may have blurred a possible correlation between duration of listening and void size.

Apparent characteristics in prints of this subject are the pre-auricular crease that is easy to distinguish in most prints but not in the 20s and 25s prints, and the small notch in the inner margin of the helix that is visible in all prints. Contrast is good in the 13s and 15s prints of this subject. Imprinted features of the 45s prints are least contrasted. For this print, both values for applied force are comparatively low (567g/696g); for the 13s and 15s prints, they are relatively high (865g/1471g and 928g/1087g respectively). This could point to a greater influence of force on the intensity of imprinted areas for this subject than of duration of ear-surface contact.

___________________________________________________________________________

Subject 3 (no additional eighth effort)

(8)

35s, but the 13s print is also relatively complete and rich in contrast. Particularly functional force was relatively high when this last print was made (1357g, as compared with 841g for the 35s print). The least complete prints (large parts of helix missing) are the 15s and 45s prints. The values for applied force were the lowest among all efforts of this subject. Particularly the recorded force peak value was relatively low during the 15s and 45s listening efforts: 916g and 857g respectively, compared with values varying between 1395g and 2026g for the remaining efforts.

The typical tragus pressure point is visible in all prints, and so is the distinctive crus of helix. The characteristic appendix of the anthelix is also visible in all prints, but least well in the 15s print. The characteristic spot on the cheek was imprinted in all prints that showed this part of the cheek.

___________________________________________________________________________

See Fig. 6.4 for selection of prints. ‘Fullest’ print after 107 seconds. The 25s print is also relatively complete, but has less contrast. This 25s further appears slightly blurred at the helix. The anthelix is particularly thin in the 13s print, but also (to a lesser extent) in the 15s and 20s prints. The characteristic shape of this anthelix therefore becomes more obvious in the prints that resulted from a longer ear-surface contact.

(9)

area in all other prints but for the 25s print, which does have good coverage of other areas. Another characteristic that these two prints have in common is the absence of an imprinted pre-auricular patch near the crus of helix. The lack of contrast in the 25s print, and the occurrence of the void between anthelix and antitragus, is surprising in combination with the relative fullness of the helix and lobe. Absence of the pre-auricular patch near the crus of helix could, however, indicate a different pressure distribution for this print and perhaps also the 13s print. Notably, these two prints were created on the same day, while all other prints were made a day later.

___________________________________________________________________________

Subject 5 (no additional eighth effort)

See Fig. 6.5 for a selection of prints. All prints appear relatively ‘full’ and rich in contrast. An influence of either applied force or duration of listening is not very clear in prints of this subject. The 35s print has slightly more contrast than the remaining prints, though both values for applied force are not particularly high for the 35s effort (1647g/2659g – as compared with values varying between 1542g and 2091g (functional force) and between 2312g and 3253g (peak value) for other efforts). It does appear as if the superior parts of the ear applied a relatively high amount of force when this print was made.

(10)

and 25s efforts (2092g/2959g and 1856g/3253g respectively). The 15s print appears slightly blurred and has greater coverage of the cheek than al other prints of this subject.

___________________________________________________________________________

6.4.2 Analysis of print-mass

Results of the analyses of variance exploring the effect of duration of listening time and applied force (functional force and peak value) on print-mass are provided in Table 6.1. As expected, print-mass significantly differs between subjects. The effect of duration of listening on print-mass appears significant. This is, however, not the case for the effect of (both values for) applied force.

Detailed examination of prints led us to believe that it is possible to detect an influence of duration of listening time in prints of most subjects, but also of the amount of force that was applied by the ear. Other researchers, such as Saddler (1996), have also demonstrated this to be the case. The dimensions of imprinted features frequently tend to increase as applied force increases. Yet, the amount of applied force has not influenced print-mass in prints of these ten subjects to the extent that the assumed effect was confirmed by the results of the analysis. Duration of listening appears to have had a greater influence on print-mass than applied force.

(11)

allowed recovery time between prints, and performed 3 out of the 4 extra long efforts) and those of group B subjects (who’s prints were more likely affected by a reduction of oily substances on the ear), we have repeated the analyses for each of the groups separately. The effect of listening time appeared clearer in group B26. As the order in which efforts of different length were performed varied greatly among subjects, we believe that differences cannot be explained by a confounding effect of a reduction of oily substances available for printing, but are more likely due to individual differences.

___________________________________________________________________________ Table 6.1 P-values for the effect of duration of listening time, applied force (i.e., functional force (model I) and peak value (model II), and the subject on print-mass (PM) relative to white and relative to median. P-values marked with * are considered significant.

___________________________________________________________________________ P-values

Variables

PM relative to white PM relative to median

Model I Duration of listening time 0.008* 0.004*

Functional force 0.778 0.941

Subject 0.000* 0.000*

Model II Duration of listening time 0.008* 0.004*

Peak value 0.365 0.308

Subject 0.000* 0.000*

6.5 Conclusion

Detailed examination of a selection of 38 earprints led us to believe we were able to detect an influence of both duration of listening and the amount of applied force on print surface and intensity. However, only the impact of the duration of listening was supported by

26

(12)

a statistical analysis of print-mass.

Acknowledgements

Colleagues of the Department of Anatomy and Embryology of the LUMC are thanked for donating their earprints.

x The duration of the listening time appears to affect print-mass more than applied force does. x It is advised to vary the duration of the listening time when enrolling prints for the purpose of

(13)