Making the Work

We can identify three layers to OTE that clarify the making of the work: the planned and proposed live art installation Only the Envelope; the pre-existing video work Telephone that was viewed by visitors to OTE during the ‘experiment’; and the audiovisual documents that were created during the installation, also grouped under the umbrella of the work OTE. Such a dissection is somewhat arbitrary, as these layers overlap and inform one another, as we will show, but each will be addressed briefly in this section. 

Only the Envelope: An artistic exercise in data retention was funded by Edith Cowan University via its eResearch Technology Funding Scheme (ETFS), which supports pilot research projects employing information technologies held by the university. OTE is artistic research, but since its aesthetic intentions also require the participation of others it is also human research. To meet the requirements of the university’s ETFS, as well as the university’s Human Research Ethics Committee (HREC), the artist ensured her applications made some clear claims about the data to be gathered. Namely, when viewing a video and wearing eye-tracking glasses, who do the participants focus on? Where do they look? How long is their attention span? These data are summarised below.

Sixty-five participants were recorded wearing eye-tracking glasses while watching Telephone, resulting in sixty-two recordings. Of these, thirty-two could be used for comparative analysis: they had complete demographic data with clear recordings, in which participants watched the entire video (watching the whole video was not a requirement). Eleven recordings were discarded because they were patchy or incomplete; in these cases, the hardware was not able to keep track of the participants’ eyes for large parts of the session. In one case, calibration was inaccurate and so the recording was discarded. The analysis focused on ten participant recordings drawn from the thirty-two useable recordings, where half were male participants and half female. 

The test video Telephone is a reflexive and non-narrative digital video artwork in which four performers directly address the viewer in the intimate style of a dating video.[6] McKenzie led a team of four artists in making the work, whose dramaturgy reflects the difficulties of forming authentic interpersonal connections in the age of the digital social network, where sharing personal information in the public sphere is the norm. In this way, the themes of OTE build on McKenzie’s earlier artistic research in Telephone.

For the purposes of analysing the eye-tracking data generated in OTE, two scenes from Telephone – the first and the last – were chosen for comparison. The first was selected because it depicts a single face that fills the screen, offering an opportunity to clearly indicate where participants mainly focus. The eye-tracking technology visualises information in two ways: heat maps average out the areas of concentrated looking, while gaze plots reveal the spread of looks over the image. Unsurprisingly, analysis of the first scene of the test video shows that the focus ‘hot spots’ were on the eyes, nose and mouth areas. In some cases participants looked at the background, particularly the line dividing the dark and light areas of the background, suggesting that viewers’ eyes are drawn to areas of contrast.

Telephone builds in a structurally recursive (rather than narrative) fashion, and the last scene depicts an array of sixteen faces in a four-by-four grid who address the viewer simultaneously, thus providing an opportunity to learn where OTE participants mainly focus when there are different faces on one screen. Analysis of the last scene revealed gazes resting on the column of faces on the right-hand side of the screen; however, the ‘hottest’ spot was a face in the upper right-hand corner of the lower left quadrant. 

These interpretations were made from a small sample, though the data accord with expected findings. Further analyses might repeat the processes of mapping hot spots and gaze plots in the first and last scenes for the remaining twenty-two usable recordings to see if findings are consistent in a quantitative comparative analysis. Additional analysis could investigate whether the demographic information collected reveals a relationship with eye-tracking data. For example, an initial analysis of the length of time participants submitted to surveillance suggests that gender is not significant here. 

Given that OTE uses eye-tracking in an exploratory way, nothing new was added to the store of knowledge regarding viewing behaviour that has not already been noted by behavioural psychologists. Nor was it the intention for the project to challenge expected findings regarding viewing behaviour. Rather, we sought to create an encounter in which the eye-tracking device played an important psychological role in the unfolding drama of gazes. It is by framing OTE as creative arts research that space can be made for findings beyond those anticipated through research design. The paradigm of arts research methodologies recognises that knowledge is ‘often unstable, ambiguous and multidimensional’.[7] Smith and Dean’s practice-as-research model, the iterative cyclic web (a name that illustrates the varieties of relationships and processes involved), notes that process-driven creative practice and research can be ‘directed towards emergence, that is the generation of ideas which were unforeseen at the beginning of the project’.

Thus, when the artist discovered during the course of the project that the eye-tracking videos could potentially identify participants, as their reflections can be momentarily glimpsed when Telephone depicts dark (and therefore reflective) surfaces, OTE adapted to accommodate this finding. These accidentally recorded reflections immediately render the eye-tracking videos more visually interesting and meaningful in the context of a work about data capture, but they cannot be shared as this could potentially breach participant anonymity. Anonymity is guaranteed as a condition of consent and the participants’ eye-tracking data are captured as de-identified digital video files. To abide by the requirements of the HREC, the artist re-enacted the eye-tracking videos she wished to interpret and share by playing the role of participant within her own work. This re-enactment and its associated audiovisual materials became the third layer of OTE and the rationale for this exposition. It is important to note that under the requirements of the university’s HREC, all data gathered in the course of this project are protected by the researchers for a period of five years, after which time they will be destroyed. In this exposition, the term ‘participant’ refers to visitors who consented to wear eye-tracking glasses; the terms ‘visitor’ or ‘viewer’ refer to participants prior to consent, or for those who didn’t consent.