A paper about the eye-tracking drawing project Id-Dgħajjes tal-Fidili, published by the drawing journal Drawing: Research, Theory, Practice is now available at the following:
Attard, Matthew (2022), ‘Eye (re)drawing historical ship graffiti: Tracing ex-voto drawings with eye-tracking technology’, Drawing: Research, Theory, Practice, 7:2, pp. 185–98, https://doi.org/10.1386/drtp_00088_1
Figure 22: 5 superimposed eye-drawings of my right hand
Figure 23: Generative development of 30 tween curves between the 5 eye-drawings in Figure 22
What universe of drawing does the practice of eye drawing really fall under? I feel that the performative element while eye-drawing my chosen subject matter is embedded into my previous knowledge of drawing. I seem to automatically adapt certain methodologies which are comparable to techniques that have a long history in drawing (and which I should both highlight and challenge). A primary example is the intention of attempting to ‘contour’ along the subject being looked at and to present them as such (as in Figure 22).
At the same time, the adaptation of the eye-tracker into a drawing medium is in itself allowing me to question several acts of looking while drawing, and by choosing the hand (and body) as subject matter, the practice is being given a specific context. Therefore, during the act of eye drawing and during my first reactions to the data processing, the general mood is that of drawing ‘as drawing’.
However, examples like Figure 22 contain more information other than the eye-drawn contours. Within the virtual space, they exist as 5 individual curves in a three-dimensional field constituting in several possible viewpoints (Figure 22 is closest to a conventional drawing result). Each curve (eye-drawing) consists in a set of time-based coordinates, along with several other data provided by the eye-tracker.
The mood of drawing ‘as drawing’ starts to fade within this context and Figure 23 is essentially a product of this condition. It becomes a computer-assisted drawing, or a computer-generative drawing. In Figure 23, the curves stop being referred to as eye-drawings and become information for a set of instructions — specifically, 30 tween curves in the space between every 2 curves (eye-drawings). In this scenario, the eye-drawings become data/information which guide the algorithm in order to create the instructed tween curves. The generation of the tween curves is generative and I therefore do not have full control on its appearance. However, I can vary several inputs like; the number of tween curves, viewpoint of the exported image, distance between curves (eye-drawings) and positioning of the latter within the virtual space, among other options (which I will continue to explore through the practice). The linear form of the eye-drawings, are themselves the primary influence for results such as Figure 23.
Figure 13: Three eye-drawings of the same rotating hand-gesture
After repeatedly eye-drawing my left hand from the same perspective, I tried eye-drawing my right hand from different points of view by rotating the same gesture (Figure 13 and Figure 14). The hand and drawing have a long history, and not just because of the manuality within the hand-eye coordination. The hand is the ‘skill’ and the means by which to ‘imprint’ while drawing, but it has also been regarded as one of the most difficult body parts to draw as a subject matter; not just for its anatomical rendition but also due to the cultural and symbolic values carried by its gestures. It has been therefore studied and drawn extensively (Petherbridge, 2010, pp. 251-259).
The hand carries innate verbal ways of communication, and it has been crucial in the evolutionary steps of our body, making us the humans of today (it allowed us the use of tools). The hand is also a sign of presence. Perhaps these are some of the reasons why it has been widely represented, from primitive art to contemporary times. Yet, in our daily routine we are barely aware of the hand.
It is worth mentioning that the hand has also been the vehicle with which to challenge the conventional ways of drawing (eg. automatism). While I am eye drawing, the hand is free as it is the body part which has been by-passed from the process. It is there and it can be eye-drawn. It is therefore very instinctive to do so.
__________
References:
Petherbridge, D. (2010). The Primacy of Drawing. New Haven and London: Yale University Press.
Figure 14: Three eye-drawings of the same rotating hand-gesture
Figure 11: Superimposition of the 6 eye-drawings of my left hand
Figure 11 is a drawing made up of a superimposition of the six eye-drawings in my previous post. When I am eye-drawing from life, I cannot refer to the visual of the ‘physical drawing’ being created as this is only generated during post-processing. Many drawers highlight the importance of this reference in their practice, as does Richard Talbot who specifically talks about perspective drawing and its thinking about the orientation; “to the relationship with the surface of the paper – the picture plane” (Talbot, 2008, p. 45).
The methodology I have been using so far is to import the recorded eye-tracking coordinates (as a set of points) into a 3D software like Rhino, from where I can process them into drawings using a polyline curve (which follows the temporal value of each point). At this stage, the eye-drawing sits in a virtual 3-dimensional space, and I navigate this with the same understanding of when physically drawing on a picture plane. I had not noticed that this was my reasoning while developing these eye-drawings until I recently read an interview of Michael Kidner where he talks about how he governed 3D Studio Max through knowledge coming from his; “previous practical pursuits in the material world” (Eames, 2008, p. 139).
Perhaps it was also this way of thinking/doing that drove my intuition to superimpose the hand eye-drawings. When merged into one image, I see the eye-drawings become an extended representation of moments in time, highlighting the fact that even though while eye-drawing I was attempting to follow a ‘contour’ through an invisible trace, outlines are a synthetic construct and do not exist in nature. There are, therefore, infinite possibilities to its representation as the history of drawing illustrates for us.
In this sense, the image of Figure 11 becomes a continuous eye-drawing (a kind of seismography of the drawing eye) other than 6 superimposed eye-drawings of 15 seconds each.
Is the practice I am dealing with a process of relearning, learning or unlearning (with regards to drawing/seeing/looking/gesturing)? The practice is resulting in new visions and visuals, and the technologies of the eye-tracker together with the 3D software are making them possible. I cannot however mistake the technology for the idea. The machine is not aiding my way of drawing or looking at the world, but is making an image/views possible, which was/were previously unnoticeable. In this context, I elaborated Figure 12, which includes a generative development between the 6 eye-drawings in Figure 11. These algorithmic/generative/computer-aided drawings will be tackled in future posts.
__________
References:
Eames, A., (2008). Embedded Drawing. In S. Garner (Ed.), Writing on Drawing, Essays on Drawing Practice and Research (pp. 125 – 140). Bristol, UK: Intellect Books.
Talbot, R., (2008). Drawing Connections. In S. Garner (Ed.), Writing on Drawing, Essays on Drawing Practice and Research (pp. 43 – 58). Bristol, UK: Intellect Books.
Figure 12: Computational development between the superimposed 6 eye-drawings of my left hand
This experiment came right after my previous posts, using the same set-up, eye-tracker and calibration. Just as in my previous recordings, my left hand was placed on the table in front of me, and the intention was now to eye-draw it (repeatedly). The scanpaths in my previous posts dealt with eye-tracking other than actual eye-drawing, as both scanpaths are the traces of my eye while drawing with my right hand. This post illustrates six eye-drawings resulting from the set-up mention above, and they lasted approximately 15 seconds each to record.
This is 9 seconds less than the previous blind-contouring exercise, which does not seem to be a great difference until one calculates that they are 37.5% of the (earlier) time spent blind-contouring. I am flagging this up not because I am interested in the mathematical calculation of the act of drawing (or eye-drawing), but because this can lead me to several new questions about my own practice. It can be argued that the blind-contouring exercise took longer due to the hand co-ordination involved while drawing (eye-drawing is more ‘direct’), but I think there can be other variables to consider.
What happens if I try to consciously go slower? Can I actually go slower? The eye movements are quite difficult to control and saccades are always in motion. How can this ‘trace’ relate to the line/mark/contour in drawing, and how does it compare to the notion of ‘reducing’ the mark before it quickly transforms into a defined image? I am not (spatially) aware of the mark while eye-drawing, or rather, the mark is in my head (not on a support). I can only tangibly relate to it during post-processing (virtually).
Does the distance I was eye-drawing my hand from naturally influence the speed (and other factors)? Distance has always been a crucial issue to take into account when drawing; both physically and mentally.
I am also intrigued by the fact that it seems to have automatically taken about 14-16 seconds for every eye-drawing to manifest (six times in a row). Is this an indication of habituality? All six eye-drawings unconsciously started from the back of my hand, moving clockwise through my fingers and ended below the wrist. When hand-drawing I am aware that I make a lot of conscious (and sometimes habitual) decisions; where to start on my page, where to start on the object I am gazing at, etc. What happens if I train myself to be more aware of where to start and how to ‘move’ while eye-drawing (or should I not)?
Another important mention are the fixations. Arguably the fixations in the eye-drawings represent some kind of hesitation in figuring out where I am in the ‘drawing’ (relating directly to somewhere in my mental imagery). It is also interesting to note that the 6th eye-drawing manifests an unconscious ‘inattention’ towards the completion of the drawing, resulting in a ‘misdirection’ of the saccades. These unpredictable registrations can be of most interest when treating the practice of eye drawing in context of its unique perceptive moments in time and space.
Figure 5: Eye-drawing 1 of my left hand
Figure 6: Eye-drawing 2 of my left hand
Figure 7: Eye-drawing 3 of my left hand
Figure 8: Eye-drawing 4 of my left hand
Figure 9: Eye-drawing 5 of my left hand
Figure 10: Eye-drawing 6 of my left hand
This experiment came right after Experiment 1, using the same set-up, eye-tracker and calibration. The left hand was again posed on the table in front of me in the same way as in my previous post, and I blind-contoured it with my right hand.
Three mentions need to be noted. Firstly, during my past eye-drawing attempts I associated some experiential aspects of the practice of eye-drawing to blind-contouring (Attard, 2018, pp. 133 – 136). I felt
confident in doing so due to the extensive blind-contouring sessions I did in the past as a drawer myself. I feel that the most important similar characteristics to note are; how the eye behaves like a pencil (with the difference that in blind-contouring it coordinates with the hand), and how the eye experiences ‘synthetic’ contouring while gazing at the subject (in this case – my left hand).
Secondly, I saw that in some cases this technique has been referred to as blind-contouring (Nicolaïdes, 1969; among others), while in others as blind-drawing (Tchalenko and Miall, 2017; Petherbridge, 2010; among others).
Thirdly, blind-contouring is a practice which has been observed to be automatically implied by artists when drawing from life after they gain several experience (Tchalenko and Miall , 2017). Together with the latter point it is important to emphasise that I am not referring to the technique’s automatic induction for unconscious imagery, but to an enhanced seeing-drawing understanding.
Figure 3 shows my blind-contoured hand and Figure 4 illustrates my eye-scanpath while drawing it. The blind-contouring exercise lasted 24 seconds. Unlike in my previous post, the eye-scanpath (my gaze) does not show any reference made to the ongoing drawing on paper, and instead of several ‘traces’ projecting outward from the ‘shape’ of the hand, Figure 4 evidences a more linear/contained/controlled quality. This includes several fixation points along the hand’s ‘outline’. Some agitation occurs at the central area of the hand, which is presumably the area where the eye had to move along the superimposed fingers (therefore needing more perceptual understanding). I also suspect that some fixation occurred at points when my (drawing) right hand was ‘catching-up’ with the eye-movements. How much does this eye-tracking result vary in quality and in representation from an actual eye-drawing of the same hand?
__________
References:
Attard, M. (2018). Datafication as a contemporary artistic process [Unpublished Masters Dissertation]. University of Malta.
Nicolaïdes, K. (1969). The Natural Way to Draw (M. Harmon, Ed.). Houghton Mifflin. (Original work published 1941)
Petherbridge, D. (2010). The Primacy of Drawing. New Haven and London: Yale University Press.
Tchalenko, J., & Miall, R. (2017). Auguste Rodin Draws Blind: An Art and Psychology Study. Leonardo, 1-17.
Figure 3: My blind-contoured left hand
Figure 4: Eye-scanpath result while blind-drawing Figure 3
This experiment consisted in eye-tracking my eye-movements while drawing my left hand. A monocular Pupil Labs Core eye-tracker was used, tracking the gaze of my left eye. Figure 1 shows my hand-drawing of my left hand and Figure 2 shows my eye scanpath while drawing it. The drawing took 50 seconds to complete (and consequentially Figure 2 is the result of 50 seconds of data).
My practice-based research question consists in the attempt to explore ways of drawing with my eyes, but this exercise visualises my eye movements during my usual hand-drawing practice (which would be useful for future comparisons).
I feel that wearing the eye-tracker does not influence me much due to its unobtrusive nature, and even though at first I am aware and cautious about the fact that I am recording my acts, this immediately wears away as I concentrate on the practice of drawing. My main questions when drawing from life concern the what and where to look at. Many definitions of drawing have been attempted, and the ones I acknowledge most are those which seem to allude to the inconclusive and deliberately attempt a ‘non-definition’ of drawing — perhaps due to the ‘ghostly’ and ‘unfinished’ nature of drawing (and the practice) itself. It is worth noting that I am looking at drawing with my eyes for the sake of the act of drawing per se, and not to regard it as a sketch, as a way to problem-solve or any other possible drawing adaptation.
Figure 2 reminds me of several important points. There is more gazing activity happening around the hand area, which intensifies along its ‘contours’. This evidences the looking and seeing quest of identifying an outline/contour while drawing. Perhaps it is also evident of the induced/habitual ‘rule’ when drawing from life; to draw what you see and not what you know. Drawing from life is both a hesitant and yet, precise activity.
Figure 1: Hand-drawing of my left hand
Figure 2: Eye-tracking result while drawing Figure 1