Log in / Register
Home arrow Business & Finance arrow Chronic Regulatory Focus and Financial Decision-Making
< Prev   CONTENTS   Next >

3.2.2 Eye Tracking

As indicated in Sect. 3.1, the eye tracker was used as the measure to collect behavioural data regarding the participants' asset and portfolio allocation choices. The eye tracker is an objective tool (Duchowski 2007) and thus using the eye tracker and the self-report as measures to test the hypotheses will limit the effects of social desirability bias and common method variance (Podsakoff et al. 2003). This section outlines the use of the eye tracker as a measure.

After the chronic regulatory focus of participants was assessed, participants

underwent a distraction scenario (Sect. 4.3.2) to prevent them from determining the hypotheses (Shavit et al. 2010). They were then seated in front of the Tobii T120 Eye Tracker to determine which assets and portfolios they looked at for a pro- portionately longer time. The asset and portfolio that they looked at for a propor- tionately longer time is deemed to be the selection. Using proportionate viewing time aids in controlling for individual differences (Shavit et al. 2010).

Eye tracking is a technique that collects behavioural data whereby an individ- ual's eye movements are measured so that the researcher knows where a person is looking at any given time, along with the sequence in which their eyes are shifting from one location to another (Poole and Ball 2005). Eye tracking was first used in reading research over 100-years ago (Rayner et al. 2011), and a multitude of studies across various disciplines have been conducted since. Past research has utilised eye tracking to gain insight into various aspects of consumer behaviour (Janiszewski 1998; Rosbergen et al. 1997), indicating the veracity of eye tracking methods in collecting behavioural data.

Most systems today use a 'corneal-reflection/pupil centre' method (Goldberg

and Wichansky 2003), consisting of a standard desktop computer with an infrared camera mounted below the display monitor. The infrared light from an LED embedded in the infrared camera is first directed in the eye to create strong reflections in target eye features to make them easier to track. The light enters the retina and a large amount of it is reflected back, making the pupil appear bright and well-defined. The corneal reflection is generated by the infrared light, appearing as a small glint. Once the software has identified the centre of the pupil and the location of the corneal reflection, the vector between is measured, and the point of interest can be found. Video-based eye trackers have to be fine-tuned to the particularities of each person's eye movements by a 'calibration' process. This process is slightly different depending on the system that is being used, and for the Tobii T120, in this book, a 9-point monitor calibration was done.

Eye tracking technology possesses several limits, and these may affect data that is collected. For example, if one is interested in analysing fixations, [1] the equipment must be optimised for such purposes (Karn et al. 2000). Sampling rate is also important, as a 60 Hz rate may be suitable for usability studies, but not for reading research, for example (Rayner et al. 2011). However, the 120 Hz rate of the eye tracker in this experiment is more than sufficient. Eye trackers are generally somewhat sensitive, and can have difficulty tracking participants who wear hard contact lenses and bifocals, as these will interrupt the normal path of a reflection, and thus care was taken in this book to exclude such participants. Those with unusually large pupils or 'lazy eye' may also be incompatible with the hardware. In addition, once calibration is completed, this process should be repeated at regular intervals to maintain an accurate point-of-regard measurement. However, as par- ticipants were only in front of the display screen for less than half an hour, this is unnecessary. To ensure that eye movements are properly attributed to actual cog- nitive processing, the tasks that participants undergo should be well-defined (Just and Carpenter 1976), and visual distractions such as colourful objects around the screen should be eliminated lest when they alter the eye-movement data (Goldberg and Wichansky 2003). It is recommended that the researcher stays out of sight of the participants, and refrains from speaking to them (Strandvall 2013). Moreover, given that eye tracking results in huge amounts of data, it is essential to perform filtering and analysis automatically, to save time and minimise errors through manual data processing. These principles have been taken into account in this book, resulting in a clear and to-the-point experimental design.

Several qualitative data analysis methods exist within the Tobii Eye tracker software, to display the data collected, such as gaze plots, heat maps, cluster vi- sualisations and bee swarm visualisations. Gaze plots show the sequence and positions of fixations on a static media, with the size of the dots indicating the fixation duration and the numbers in the dots representing the order of the fixations, are used to illustrate the gaze patterns of participants. Heat maps use different colours to show the aggregated number of fixations participants made in certain areas, to analyse how long, where and how many times participants look at pre- defined areas on the screen. Cluster visualisations are aggregated graphic repre- sentations of areas with high concentrations of gaze data points. Bee swarm visualisations are dynamic representations for the fixations of a group of partici- pants on top of a selected video, with the fixation of each participant illustrated with a coloured dot (Strandvall 2013). As these are qualitative analysis methods, they were not used in the book, as results from qualitative eye tracker tests are suitable for describing problems and making improvements, not for describing behaviour (Strandvall 2013).

Quantitative eye tracker tests are more suited for generalisation and comparisons between groups, as in this book. Varieties of quantitative methods are used in eye tracking, such as total scanpath length, mean fixation duration and various area-of-interest[2] (AOI) techniques. An AOI technique was used for this book, to see where and how long participants affix their gaze on the various situations displayed. Of the methods available, number of visits and percentage of participants fixating on an AOI have been deemed the most suitable, as these give the most accurate representation of the areas in the display that individuals find most interesting (Strandvall 2013). The number of fixations on an AOI was not used as this can be influenced by AOI size. There still exist limitations with the AOI method. As any point within the target area classifies as a fixation point, saccade [3] points leading into a fixation may be included in the final fixation, resulting in misleadingly long durations for identified fixations. Moreover, long saccades through large AOI regions may be identified as fixations, leading to incorrect data analysis (Salvucci and Goldberg 2000), although this effect may be controlled for by recruiting a large number of participants.

Research has shown that fixations and interests are related, and that eye movements reflect the thought process (Yarbus and Riggs 1967). It is hypothesised that when a participant looks at an object, the former thinks about the object as long as the recorded fixation (Just and Carpenter 1980). Visual attention is slightly ahead of the eye, and when attention moves to a new location, the eyes follow suit (Hoffman 1998). As such, the length of fixations recorded on the system may give an indication of the participants' interest in a given allocation scenario. However, given the nature of covert attention,[4] the resulting fixation patterns may not indicate one's attention, but only where the eye has been viewing (Wright and Ward 2008). It is still not possible to infer specific cognitive processes solely from fixation on a particular object in a display (Holsanova 2006). Thus, the purpose of the eye tracker in this book is not to act as a sole test for allocation choice, but to act as a complementary instrument to the self-report. This is hoped to reduce social desir- ability bias (Podsakoff et al. 2003) and may give insight into the participants' thought process (Yarbus and Riggs 1967).

As detailed earlier, the Tobii T120 Eye Tracker was utilised for this book. It possesses a 120 Hz data rate, freedom of head movement of 30 × 22 × 30 cm, binocular tracking, maximum of 33 ms latency and 300 ms time to tracking recovery. Participants entered the lab individually, and were seated in front of a computer monitor. The next section outlines the research regarding measures of financial literacy, along with how financial literacy was tested.

  • [1] Fixation is the maintaining of the visual gaze on a single location
  • [2] AOI techniques refer to displays wherein an area on the screen has been highlighted beforehand, and all fixations within this area will be recorded by the eye-tracking software
  • [3] Saccades are quick, simultaneous movements of both eyes in the same direction
  • [4] Overt attention is the direction of one's senses toward a certain stimulus. Covert attention is the act of focusing on one of several possible stimuli
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science