Bernhard Angele
Jon Andoni Duñabeitia- Centro de Investigación Nebrija en Cognición (CINC), Facultad de Lenguas y Educación, Universidad Nebrija, Madrid, Spain
When non-invasive, infrared-based eye-tracking technology combined with real-time data processing first became available in the 1970s (Young and Sheena, 1975; Evans and Gutmann, 1978) it opened up a new way of observing cognitive processing in action. Modern eye-tracking has been used in many areas of cognitive psychology, cognitive science, and neuroscience, but it has been especially important in the field of reading research. It seems fair to say that more has been learned about information processing and eye movement control in the past 40 years since the introduction of infrared-based eye-tracking than in the 100 years of reading research before that, starting with the first experiments in 1879 (Huey, 1908; see also Wade and Tatler, 2008). These advances have led to the development of computational models of word identification and eye movement control during reading, such as E-Z Reader (Reichle et al., 1998, 2003, 2009), SWIFT (Engbert et al., 2002, 2005; Kliegl and Engbert, 2003; Schad and Engbert, 2012), Mr. Chips (Legge et al., 1997), the Rational Model of Reading (Bicknell and Levy, 2010; Duan and Bicknell, 2020), and, most recently, OB1-Reader (Snell et al., 2018) and Chinese Reading Model (Li and Pollatsek, 2020). However, a closer examination of the publications on eye-movement research during reading reveals surprising disparities in terms of the countries where eye-tracking research is taking place.
Publications on eye-movement research in reading
In April 2024, Scopus (Elsevier) listed 7,473 documents (mainly journal articles and conference proceedings) published in the last 50 years related to eye-movements in reading1. When split by country affiliations of the authors, this corresponds to 9,320 country affiliations (one publication could have authors affiliated with more than one country and 271 unclear affiliations were excluded). Out of these affiliations, 2,326 (or 24.3%) corresponded the United States, followed by the United Kingdom (1,195 or 12.5%), Germany (902 or 9.4 %), and China (604 or 6.3%). If we aggregate the results by continents, 47.3% of author affiliations (4,331) were based in Europe, 29.2% of affiliations (2,670) were based in North America, and 18.2% of affiliations (1,663) were based in Asia (apart from China, these publications came mostly from Japan and South Korea). Oceania had 3.39% (310) of publications (of which most were based in Australia and New Zealand). Very few publications had affiliations with authors outside of these areas, with only 64 affiliations (0.7%) from Africa and 116 (1.27%) from South America. Figures 1A, B shows this inequality. Comparing the situation in the first 25 years (1974–2000, Figures 1C, D) with the situation in the last 3 years (2021–2024, Figures 1E, F) this pattern has hardly changed, with one major exception: In the first 25 years, China did not play a major role in reading research using eye movements, but, since 2021, China's contributions (257 affiliations) have only been surpassed by the United States (369). This clearly reflects the massive investment in reading and eye-movement research in China over the last 20 years. Unfortunately, apart from China's enhanced role in eye-movement research, little else has changed: In the past 3 years, most eye movement research still took place in the same Western countries that had dominated the field in 1974–2000.
Figure 1. Author's country affiliations of publications on eye-movements in reading by time period. (A, B) show the entire period from 1974 to April of 2024. (C, D) show the initial phase of eye-movement research in reading from 1974 to 2000, where the United States dominate. (E, F) show the most recent period from 2021 to April 2024, where eye-movement research is more international, but still mainly takes place in a handful of countries. (A, C, E) are choropleth maps with number of affiliations represented by color. (B, D, F) are cartograms showing the same data as the choropleths in which country boundaries are additionally distorted such that their area is proportional to the number of author affiliations. Figures created with ggplot2 (Wickham, 2016) and cartogram (Jeworutzki, 2023). Figure R code and data available at https://osf.io/rcd9g/.
Eye-tracking should be truly global
There are several reasons to pursue a more global approach in the use of eye-tracking research reading. Relying only on results from participants from a scarce number of specific countries harms the field. As Henrich et al. (2010b) pointed out, participants from Western, Educated, Industrialized, Rich, and Democratic (“WEIRD”) countries are not representative of all of humanity (see also Henrich et al., 2010a,c; Schulz et al., 2018). Therefore, drawing conclusions about language processing and reading in general, based only or mostly on results from such participants, is difficult to justify (Andringa and Godfroid, 2020).
Since reading inherently implies a (portion of a) text in a particular language, with its own properties and writing system, notions of “generalized” or “universal reading” are necessarily limited (cf. Rueckl et al., 2015). Reading Up to the early 2000s, reading research was essentially research on reading in English. When Keith Rayner published his highly influential review article (Rayner, 1998), most of the results were on reading in English with few results from other languages. How many important results have we been missing due to limiting ourselves to such a short list of languages? We will not know what proportion of findings we thought to be universal just happen to be common to the languages on this list until we start studying reading in all languages people use to read and write. This could enable global replication efforts similar to #EEGManyLabs (Pavlov et al., 2021), with researchers all over the world collaborating on replicating key effects.
Reasons for the eye-tracking gap
Why is eye-tracking not used more around the world? In theory, apart from behavioral methods, eye-tracking is the ideal technique to use in order to study language in laboratories with limited resources. Unlike EEG, MEG and fMRI, which have extremely high set-up and/or maintenance costs, once an infrared-based eye-tracking system is set up, it requires only electricity to function. Additionally, modern compact eye trackers are completely self-contained and can be set up easily in any environment where a typical laptop computer can be used.
Despite this, as we can see from the publication statistics, eye tracking seems to be used very little to study reading outside of Western and East Asian countries. Why is this? The answer lies almost certainly in the high initial cost of purchasing an eye-tracking system, which, depending on the model and manufacturer, is somewhere between 3,000 and 50,000 US$. Not all eye-trackers are equally suited for reading research, however. Most reading studies are performed with eye-trackers that have a sampling rate of at least 500 Hz. In the last 15 years, recording eye-movements at 1,000 Hz or even 2,000 Hz has become the de-facto standard (e.g., Kliegl and Laubrock, 2017). For many researchers in countries with limited resources, buying such expensive eye-trackers is simply impossible.
In past years, there have been some promising attempts at creating more affordable eye-tracking solutions such as the Eye Tribe eye-tracker (Ooms et al., 2015). However, the Eye Tribe was acquired by Oculus VR/Meta in 2016, stopping development of the product. Similar acquisitions also happened with other companies (e.g., SMI, which was acquired by Apple). Unfortunately, this has, so far, not resulted in devices with research-level eye-tracking capabilities (cf. Valliappan et al., 2020, but this technology has not been made publicly available). Consequently, there are few alternatives for a researcher who does not have the funds for buying a high-precision eye-tracker.
Online eye-tracking using webcams is emerging, with Webgazer.js (Papoutsaki et al., 2016) being the most prominent example. However, using the participant's device naturally comes at the cost of precision and introduces new complications such as accounting for the effect of head movements, screen size, environmental illumination, etc. While there are some encouraging results from recent studies on non-reading tasks (e.g., Slim and Hartsuiker, 2023; Steffan et al., 2024), online eye-tracking has not been formally tested for reading research, since the available sampling rates are extremely low (most webcams can record a maximum of 30 or 60 images per second). In a previous study (Angele et al., 2022), we showed that even behavioral experiments with precise timing requirements can be replicated online. If the same is true for eye-tracking, it would be ideal for enabling eye tracking online all over the world. It is, however, clear that, given the limitations in accuracy (Papoutsaki et al., 2016), the experimental designs possible with this technology will be quite limited.
Discussion
How can we close the eye-tracking gap?
In summary, there are few options for researchers with limited resources who would like to study language processing during reading using eye-tracking. We propose a number of measures, both at the institutional level and at the level of individual research projects:
1. Fund initial investment to make existing eye-tracking systems available in more countries
First, research funders, especially those with a focus on international collaboration and research development, should prioritize projects that make eye-tracking technology accessible in more countries. The simplest way of doing this would be to fund projects that enable researchers in countries with limited resources to make the initial investment to buy a high-precision eye-tracker. The fact that eye-trackers only need a small laboratory space and electricity to work means that this initial investment could provide payoffs for decades to come. This would be the fastest way to enable researchers to use high-precision eye-tracking. Individual researchers in WEIRD countries could contribute to this by establishing collaborations with partners in non-WEIRD countries, applying for grants in WEIRD countries to secure funding for purchasing high-precision eye trackers for non-WEIRD research partners, and committing to work with their new partners to publish research from this new WEIRD-non-WEIRD collaboration.
2. Fund development of open-source eye-tracking systems
In general, leaving eye-tracker development to private companies has been very successful and has led to the availability of mature, highly reliable and precise eye-tracking systems. However, it has not led to the development of lower-cost eye-tracking technology that would be accessible to researchers in countries with limited resources for research funding. Additionally, as the acquisitions of the Eye Tribe and SMI have shown, there is always the danger of privately owned technology to be removed from the market and to become inaccessible to researchers. We believe that there is room for the co-existence of both the premium, high-precision devices that are available on the market today, which are easy to set up and whose manufacturers provide excellent support to those researchers who can afford them, and for a less expensive open-source solution that may not offer the same degree of precision (but sufficient to do reading research) and would rely on researchers to solve problems themselves.
3. Validate existing systems for reading research
What can individual researchers and research groups do to close the eye-movement gap? For most, developing a new eye-tracker is outside of our area of expertise. However, researchers fortunate enough to have access to a high-precision eye-tracking system can use it to validate lower-precision eye-tracking systems, including online eye-tracking solutions. There will inevitably be limits in terms of what phenomena in reading can be observed using eye-trackers with lower precision, but it may well be possible to observe effects of processing on eye-movements that are numerically large and relatively stable. Validation studies already exist (Ehinger et al., 2019; Kaduk et al., 2023), but not for reading research. Publishing such validation studies will encourage researchers with limited resources to invest in eye-trackers that have been shown to be useful in reading research, and, at the same time, will make it easier for such researchers to publish results obtained using already-verified devices.
Conclusions
Publication metrics clearly show a gap between countries where eye-tracking research on reading is done extensively and those where such research is virtually inexistent. The example of China demonstrates that, with sufficient investment on the part of research funders, it is possible for a country to move from the latter group to the former relatively quickly. However, this is likely not an option for many other countries given the cost of initial investment necessary to acquire eye-tracking equipment. As a result, the lack of eye-tracking research in most countries and on most languages threatens the validity and generalizability of reading research overall and reduces its overall usefulness. Funding collaborations involving the deployment of existing high-cost eye-tracking systems in countries with limited resources, funding the development of lower-cost, sustainable open-source solutions for eye-tracking, and, on the level of individual researchers, validating existing low-cost eye-tracking devices against high-quality and high-precision eye-tracking systems of reference may, in the future, contribute to making not just reading research, but also other research that relies on eye-tracking more globally available and more useful and insightful for everyone.
Author contributions
BA: Conceptualization, Writing – review & editing, Writing – original draft, Visualization. JD: Writing – review & editing, Conceptualization.
Funding
The author(s) declare that financial support was received for the research, authorship, and/or publication of this article. This work was supported by Grant 2022-T1/SOC-23980 to BA from the Community of Madrid.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The author(s) declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision.
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Footnotes
1. ^Scopus query: TITLE-ABS-KEY (reading AND ((eye AND track*) OR (eye AND movement) OR (eye AND movements))).
References
Andringa, S., and Godfroid, A. (2020). Sampling bias and the problem of generalizability in applied linguistics. Annu. Rev. Appl. Linguist. 40, 134–142. doi: 10.1017/S0267190520000033
Angele, B., Baciero, A., Gómez, P., and Perea, M. (2022). Does online masked priming pass the test? The effects of prime exposure duration on masked identity priming. Behav. Res. Methods 55, 151–167. doi: 10.3758/s13428-021-01742-y
Bicknell, K., and Levy, R. (2010). “A rational model of eye movement control in reading,” in Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, 1168–1178. Available online at: http://dl.acm.org/citation.cfm?id=1858800
Duan, Y., and Bicknell, K. (2020). A rational model of word skipping in reading: ideal integration of visual and linguistic information. Top. Cogn. Sci. 12, 387–401. doi: 10.1111/tops.12485
Ehinger, B. V., Gro,ß, K., Ibs, I., and König, P. (2019). A new comprehensive eye-tracking test battery concurrently evaluating the Pupil Labs glasses and the EyeLink 1000. PeerJ 7:e7086. doi: 10.7717/peerj.7086
Engbert, R., Longtin, A., and Kliegl, R. (2002). A dynamical model of saccade generation in reading based on spatially distributed lexical processing. Vision Res. 42, 621–636. doi: 10.1016/S0042-6989(01)00301-7
Engbert, R., Nuthmann, A., Richter, E. M., and Kliegl, R. (2005). SWIFT: a dynamical model of saccade generation during reading. Psychol. Rev. 112, 777–813. doi: 10.1037/0033-295X.112.4.777
Evans, J. E., and Gutmann, J. C. (1978). Minicomputer processing of dual Purkinje image eye-tracker data. Behav. Res. Methods Instrum. 10, 701–704. doi: 10.3758/BF03205374
Henrich, J., Heine, S. J., and Norenzayan, A. (2010a). Beyond WEIRD: towards a broad-based behavioral science. Behav. Brain Sci. 33, 111–135. doi: 10.1017/S0140525X10000725
Henrich, J., Heine, S. J., and Norenzayan, A. (2010b). Most people are not WEIRD. Nature 466, 29–29. doi: 10.1038/466029a
Henrich, J., Heine, S. J., and Norenzayan, A. (2010c). The weirdest people in the world? Behav. Brain Sci. 33, 61–83. doi: 10.1017/S0140525X0999152X
Huey, E. B. (1908). The Psychology and Pedagogy of Reading, with a Review of the History of Reading and Writing and of Methods, Texts, and Hygiene in Reading. New York: Macmillan. Available online at: http://archive.org/details/psychologypedago00hueyiala
Kaduk, T., Goeke, C., Finger, H., and König, P. (2023). Webcam eye tracking close to laboratory standards: comparing a new webcam-based system and the EyeLink 1000. Behav. Res. Methods. doi: 10.3758/s13428-023-02237-8
Kliegl, R., and Engbert, R. (2003). “SWIFT explorations,” in The Mind's Eye: Cognitive and Applied Aspects of Eye Movement Research, eds. R. Radach, J. Hyönä, and H. Deubel (North-Holland: Elsevier), 391–411.
Kliegl, R., and Laubrock, J. (2017). “Eye-movement tracking during reading,” in Research Methods in Psycholinguistics and the Neurobiology of Language, eds. A. M. B. de Groot and P. Hagoort (Hoboken, NJ: John Wiley & Sons, Ltd), 68–88.
Legge, G. E., Klitz, T. S., and Tjan, B. S. (1997). Mr. Chips: an ideal-observer model of reading. Psychol. Rev. 104, 524–553. doi: 10.1037/0033-295X.104.3.524
Li, X., and Pollatsek, A. (2020). An integrated model of word processing and eye-movement control during Chinese reading. Psychol. Rev. 127, 1139–1162. doi: 10.1037/rev0000248
Ooms, K., Dupont, L., Lapon, L., and Popelka, S. (2015). Accuracy and precision of fixation locations recorded with the low-cost Eye Tribe tracker in different experimental set-ups. J. Eye Movem. Res. 8, 1. doi: 10.16910/jemr.8.1.5
Papoutsaki, A., Sangkloy, P., Laskey, J., Daskalova, N., Huang, J., and Hays, J. (2016). “WebGazer: scalable webcam eye tracking using user interactions,” in IJCAI'16: Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, 3839–3845. Available online at: https://dl.acm.org/doi/10.5555/3061053.3061156
Pavlov, Y. G., Adamian, N., Appelhoff, S., Arvaneh, M., Benwell, C. S. Y., Beste, C., et al. (2021). #EEGManyLabs: investigating the replicability of influential EEG experiments. Cortex 144, 213–229. doi: 10.1016/j.cortex.2021.03.013
Rayner, K. (1998). Eye movements in reading and information processing: 20 years of research. Psychol. Bull. 124, 372–422. doi: 10.1037/0033-2909.124.3.372
Reichle, E. D., Pollatsek, A., Fisher, D. L., and Rayner, K. (1998). Toward a model of eye movement control in reading. Psychol. Rev. 105, 125–157. doi: 10.1037/0033-295X.105.1.125
Reichle, E. D., Rayner, K., and Pollatsek, A. (2003). The E-Z Reader model of eye-movement control in reading: comparisons to other models. Behav. Brain Sci. 26, 445–476. doi: 10.1017/S0140525X03000104
Reichle, E. D., Warren, T., and McConnell, K. (2009). Using E-Z Reader to model the effects of higher level language processing on eye movements during reading. Psychon. Bull. Rev. 16, 1–21. doi: 10.3758/PBR.16.1.1
Rueckl, J. G., Paz-Alonso, P. M., Molfese, P. J., Kuo, W.-J., Bick, A., Frost, S. J., et al. (2015). Universal brain signature of proficient reading: Evidence from four contrasting languages. Proc. Nat. Acad. Sci. 112, 15510–15515. doi: 10.1073/pnas.1509321112
Schad, D. J., and Engbert, R. (2012). The zoom lens of attention: Simulating shuffled versus normal text reading using the SWIFT model. Visual Cognit. 20, 391–421. doi: 10.1080/13506285.2012.670143
Schulz, J., Bahrami-Rad, D., Beauchamp, J., and Henrich, J. (2018). The Origins of WEIRD Psychology (SSRN Scholarly Paper 3201031).
Slim, M. S., and Hartsuiker, R. J. (2023). Moving visual world experiments online? A web-based replication of Dijkgraaf, Hartsuiker, and Duyck (2017) using PCIbex and WebGazer.js. Behav. Res. Meth. 55, 3786–3804. doi: 10.3758/s13428-022-01989-z
Snell, J., Van Leipsig, S., Grainger, J., and Meeter, M. (2018). OB1-reader: a model of word recognition and eye movements in text reading. Psychol. Rev. 125, 969–984. doi: 10.1037/rev0000119
Steffan, A., Zimmer, L., Arias-Trejo, N., Bohn, M., Dal Ben, R., Flores-Coronado, M. A., et al. (2024). Validation of an open source, remote web-based eye-tracking method (WebGazer) for research in early childhood. Infancy 29, 31–55. doi: 10.1111/infa.12564
Valliappan, N., Dai, N., Steinberg, E., He, J., Rogers, K., Ramachandran, V., et al. (2020). Accelerating eye movement research via accurate and affordable smartphone eye tracking. Nat. Commun. 11, 4553. doi: 10.1038/s41467-020-18360-5
Wade, N. J., and Tatler, B. W. (2008). Did Javal measure eye movements during reading? J. Eye Move. Res. 2. doi: 10.16910/jemr.2.5.5
Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis. New York: Springer-Verlag. Available online at: https://ggplot2.tidyverse.org
Keywords: eye movements and reading, reading, eye movements, eye tracking, equality and diversity
Citation: Angele B and Duñabeitia JA (2024) Closing the eye-tracking gap in reading research. Front. Psychol. 15:1425219. doi: 10.3389/fpsyg.2024.1425219
Received: 29 April 2024; Accepted: 21 May 2024;
Published: 03 June 2024.
Edited by:
Mila Vulchanova, NTNU, NorwayReviewed by:
Fernando Marmolejo-Ramos, University of South Australia, AustraliaCopyright © 2024 Angele and Duñabeitia. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Bernhard Angele, bangele1@nebrija.es