Eye movements and brain oscillations to symbolic safety signs with different comprehensibility
© Siswandari and Xiong. 2015
Received: 11 September 2015
Accepted: 8 December 2015
Published: 10 December 2015
The aim of this study was to investigate eye movements and brain oscillations to symbolic safety signs with different comprehensibility.
Forty-two young adults participated in this study, and ten traffic symbols consisting of easy-to-comprehend and hard-to-comprehend signs were used as stimuli. During the sign comprehension test, real-time eye movements and spontaneous brain activity [electroencephalogram (EEG) data] were simultaneously recorded.
The comprehensibility level of symbolic traffic signs significantly affects eye movements and EEG spectral power. The harder to comprehend the sign is, the slower the blink rate, the larger the pupil diameter, and the longer the time to first fixation. Noticeable differences on EEG spectral power between easy-to-comprehend and hard-to-comprehend signs are observed in the prefrontal and visual cortex of the human brain.
Sign comprehensibility has significant effects on real-time nonintrusive eye movements and brain oscillations. These findings demonstrate the potential to integrate physiological measures from eye movements and brain oscillations with existing evaluation methods in assessing the comprehensibility of symbolic safety signs.
KeywordsSafety sign Comprehensibility Eye movement Brain activity Physiological measures
Out of all the substages in receiver stage, comprehension can be regarded as the most substantial since the correct information flow from this substage plays a big role in transmitting the intended information to the receiver. The faulty design of safety signs could lead to increased potential risks or hazards in public places. For example, a study by Gazmarian et al.  with 2659 hospital patients showed that 54 % patients with low health literacy could not understand instructions to take medication on an empty stomach, and 48 % did not understand to take medication every 6 h. Another study by Kirmizioglu and Tuydes-Yaman  reported that one major factor affecting safe driving is the comprehensibility of traffic signs by drivers. The results of these studies have affirmed the fact that safety sign comprehension is of utmost importance. Therefore, comprehensibility of safety signs should be ensured prior to being implemented in various public places.
Various evaluation methods have been developed in the past to measure the comprehensibility of the safety sign [4–10]; most of them rely on the guessability score and sometimes cognitive sign features. ISO 9186-1 specifies a method for testing the comprehensibility of graphical symbols and uses the guessability score (GS) from open-ended responses as a measure of the degree to which a graphical symbol communicates its intended message to recipients . In ISO 9186-1, each open-ended response needs be assigned to fully correct, partially correct, or wrong to assess the comprehensibility in terms of GS, as a standard in pictorial comprehension test. However, there is unavoidable subjectivity of the scoring method for evaluating the participants’ open-ended responses in the process of obtaining GS, especially for “partially correct” answers . More importantly, this evaluation method involves questions which require the users to give their opinions or ratings on the tested stimuli after they have “experienced” the stimuli. However, these opinions may be distorted by memory and consciousness effects .
Previous studies showed that changes in various physiological processes and states covary with changes in cognitive load elicited by stimuli or objects of attention [13, 14]. There are several advantages of utilizing physiological measures to infer cognitive load: (1) these measures are relatively unobtrusive, (2) these measures do not require overt performance, and (3) most of physiological measures are continuously recorded; they provide us the chance to present measures that respond relatively quickly to phasic shifts in cognitive process . Relating to the aforementioned C-HIP model, some physiological measures seem to correspond with certain stages. Eye tracking data and measures were investigated in this study since people use their visual modality (stage 2: channel) to capture information conveyed by safety signs. Moreover, eye tracking has been used as an effective tool in a wide variety of studies since it serves as a “mirror” to understanding human attention and behavior when engaged in a specific activity [16, 17]. Brain activity was also investigated since it is closely linked with comprehension in the receiver stage and it provides a “window” into the human mind [17–21].
The primary purpose of the current study was to investigate eye movements and brain oscillations to symbolic safety signs with different comprehensibility. The findings of this study could provide preliminary evidences on integrating potential physiological measures from eye movements and brain oscillations with existing evaluation methods in assessing the comprehensibility of the symbolic safety signs.
Forty-two young Korean adults (24 males: 21.2 ± 1.44 years old; 18 females: 20.2 ± 1.31 years old) participated in this study. Each participant provided written informed consent on a protocol approved by the university institutional review board (IRB No.14-17-01-A). The selection criteria are that they should have normal or corrected-to-normal vision, have no colorblindness and red-green deficiencies, and are right-handed. Additionally, participants were instructed not to consume any drinks containing alcohol or caffeine 24 h before their scheduled experiment time, to decrease the likelihood of participants being intoxicated during the experiment.
The EEG and eye tracker started to record each participant’s data right after stimuli onset, and stopped after 10 s had ended for each sign. No physiological data was recorded when participants were giving their answers verbally. This procedure was repeated for ten signs. The whole session was recorded by a video camera and also an android-based voice-recorder app, Easy Voice Recorder Pro. The whole experiment lasted for approximately 1 h.
Data processing and statistical analysis
To calculate the guessability score of each sign, a scoring procedure was carried out as follows. Two judges independently assessed the accuracy of the responses given by the participants. Correct understanding of the actual meaning of the sign symbol (over 80 % understood) was given one point, getting very close to the actual meaning (understood 66–80 %) was given 0.75 points, and getting close to the actual meaning (understood 50–65 %) was given 0.5 points. Giving the opposite answer to the actual meaning caused a one-point deduction, and zero points were awarded for any other answer . Analysis of variance (ANOVA) was conducted to see whether there was significant difference in the comprehensibility among signs. A Bonferroni post hoc test was conducted to group the signs based on their comprehensibility. Signs with a guessability score of at least 85 % were regarded as easy-to-comprehend signs, ie., good signs (ANSI Z535.3 ), and signs with a guessability score below 40 % were regarded as hard-to-comprehend signs (bad signs).
Three eye-movement measurements from eye tracking data were investigated: blink rate, pupil diameter, and time to first fixation. For each measurement of a particular participant, the result was averaged across different signs within the same comprehension group (good or bad signs). Pupil diameters for good and bad signs were investigated for 0.5–1s time interval after stimulus onset. The 0.5–1 s was chosen as the observed time interval since the pupil can react to stimuli in 0.2 s, with the response peaking in 0.5–1 s after stimuli onset [26, 27].
Power spectral analysis was conducted to analyze EEG data . First, raw data from six investigated channels for each sign was extracted using an interactive Matlab toolbox for EEG signal processing (EEGLAB). Fast Fourier transform (FFT) was then applied to transform the raw data to a frequency spectrum, resulting in frequency (Hz) on the x-axis and power (μV2/s) on the y-axis. After that, the FFT-ed data was divided into different bands according to the natural frequencies of the brain: delta (0.5–3 Hz), theta (3.5–7 Hz), alpha (8–13 Hz), and gamma (30–70 Hz) . For each participant, the average power of each frequency band across signs which belong to the same comprehension group was calculated for each channel. Therefore, the EEG data of each participant for further statistical analysis was the averaged delta, theta, alpha, and gamma signal power for the two groups of signs with different comprehensibility, for each investigated channel. To examine the hemispheric lateralization during sign comprehension, the right (R) hemisphere vs. left (L) hemisphere power asymmetry indices for each frequency band were computed for three homologous sites (F4–F3, O2–O1, T8–T7), using the formula of (R − L)/(R + L) . Laterality itself was described as qualitatively different functions from homologues areas in the left and right hemispheres .
Paired t tests were conducted to check whether different comprehensibility levels (easy-to-comprehend and hard-to-comprehend) of the tested signs affect eye tracking and EEG measures. An additional one-sample t test was performed on inter-hemispheric power asymmetry indices to check the significance of the hemispheric lateralization. SAS JMP 10 was used for statistical analyses at a significance level of 0.05.
Descriptive statistics of guessability scores for ten symbolic traffic signs
Guessability score (mean ± SD) (%)
Bonferroni grouping resulta
Do not turn left
100 ± 0.0
A (easy-to-comprehend group—good signs)
100 ± 0.0
Do not turn right
99.4 ± 3.9
Do not go straight
98.2 ± 8.5
U turn is prohibited
97.6 ± 15.4
11.3 ± 25.4
B (hard-to-comprehend group—bad signs)
No vehicle carrying explosives
6.8 ± 39.2
Headphone users may be lost in music
6.3 ± 15.5
Risk of grounding
2.9 ± 9.9
2.4 ± 15.6
Eye tracking measures
Time to first fixation
Effects of sign comprehensibility on investigated physiological measures from eye tracking and EEG data (p values less than 0.05 are shown in italics)
p value of two-group comparison (good signs vs. bad signs)
Time to first fixation
Prefrontal cortex (F3, F4)
Visual cortex (O1, O2)
Auditory cortex (T7, T8)
Overall results of eye movements and brain oscillations to traffic symbols with different comprehensibility
Overall results for the investigated physiological measures are summarized in Table 2, which showed that 13 out of a total of 27 investigated physiological measures evoked significant changes when the participants were exposed to signs with different levels of comprehensibility (good and bad signs).
In the past few decades, a large body of safety sign research has examined how sign characteristics (such as color, shape, symbol, incongruent information) and receiver personal factors (such as age, gender, belief, perception of risk, stress) impact warning effectiveness . These studies provide basic principles and guidelines for the design and implementation of more effective safety signs; however, the present study takes a further step by investigating the underlying human attention and cognitive processes that affect sign comprehension. This is the first study, to our knowledge, to investigate safety sign comprehensibility by utilizing physiological measures from eye tracking and spontaneous brain activity data. Real-time nonintrusive monitoring of human eye movements and brain oscillations during sign comprehension can provide us detailed and objective information on human attention and cognitive processes [17, 32].
The results of this study demonstrated that the physiological measures of eye movements and the brain’s natural oscillations were able to reflect the changes from different comprehensibility levels of traffic symbols. The first and most prominent indicator found in this study was the blink rate, which was inversely related with comprehension difficulty. This result is consistent with previous studies. Holland and Tarlow  stated that blink occurs at the moment of cognitive change. For example, in some cases when participants were concentrating on finding an answer to a specific problem, they tend not to blink, and when they found the answer, blink occurs. Telford and Thompson  and Broadbent  reported that participants tend to suspend blinking while something interesting catches their attention and when information in memory is being operated on. In the case of safety sign comprehension, participants tend to perform less number of blinks while being exposed to a hard-to-comprehend sign, since they had to concentrate and focus their attention to the displayed sign in order to guess the actual meaning of the sign correctly. In addition, when compared to a normal blink rate of 24.36 blinks/min obtained from Cardona et al. , a decreased blink rate (18.0 blinks/min) was observed when participants were trying to comprehend a bad sign, suggesting that more cognitive load was required during this task. On the contrary, when participants were exposed to an easy-to-comprehend sign, it was expected that they would spend less cognitive load to guess the meaning of the sign, thus allowing faster blinks (33.6 blinks/min) when they set their attention loose.
In terms of pupil diameter, results showed that the increasing cognitive load was accompanied by an increasing pupil diameter within the time interval of 0.5–1 s following stimuli onset, resulting in larger pupil diameters for signs with low comprehensibility, in comparison with signs with high comprehensibility. This result is in line with previous studies which reported that pupil dilation is positively associated with increasing cognitive load [37–39]. Different comprehensibility levels of signs were also found to significantly affect time to first fixation. Results from our study suggested that participants had difficulties fixating their gaze on one part of the pictogram when exposed to signs with a low comprehensibility level, resulting in a longer time to first fixation. This result should be reasonable and can be supported by findings from previous studies [40, 41]. Salience could be another factor that affects time to first fixation [42, 43]. Pictograms used in the good signs probably contain a more prominent aspect that can be easily detected by the viewers compared with the pictograms used in the bad signs.
With regard to EEG signal, alpha, theta, delta, and gamma oscillations were examined in this study since they have been reported to govern cognitive processes . Significant alpha, theta, and gamma power differences between good and bad signs were found in channels F3 and F4, which were parts of the prefrontal cortex of the brain. These results were reasonable, since the association between functions of the prefrontal cortex and various cognitive behaviors including orchestration of thoughts and actions and access to working memory had been acknowledged in many previous studies [44–49]. Miller and Cohen  reported that the prefrontal cortex plays a very important role in cognitive performance. They stated that representative areas in the prefrontal cortex can function as attentional templates by providing top-down signals to other parts of the brain which will guide the flow of activities needed to perform a cognitive task. More specifically, alpha power was observed to be higher for easy-to-comprehend signs in this study; this result was in line with a previous study by Klimesch  which suggested that an overall decrease in alpha power indicated an increasing cognitive load in general. According to Klimesch’s study in 1996 , alpha power was blocked or attenuated by attention and mental effort, which also explains the results obtained in the current study. Our study showed 9.9~13 % decreases of alpha power in both the prefrontal and visual lobes. Similar tendencies were also observed in a study with Alzheimer patients , where upper alpha power showed about 26 % decrease for tasks which require higher mental effort. Interestingly, about 9 % increase was also reported in the lower alpha power for tasks with higher mental effort. This finding is also noteworthy for future research, since dichotomy in the alpha frequency range could have led scientists to discover a more interesting nature of cognitive processes while encoding visual stimuli. Higher theta power was found for easy-to-comprehend signs since these signs can be encoded better compared to the bad signs [51, 52]. This result is also supported by a comprehensive study of brain oscillation from Klimesch , which suggested that theta oscillation is linked to an encoding process of new information. Lower gamma power in the prefrontal cortex (channels F3 and F4) was found for easy-to-comprehend signs; this could be explained by increasing gamma oscillations being possibly associated with the cognitive processing of attended stimuli . In the current study, the gamma power in the prefrontal cortex showed 10.2~11.7 % increases, when participants were exposed with hard-to-comprehend signs. A similar pattern was also reported by Başar-Eroglu et al. , who investigated the visual perception of ambiguous patterns and reported 40~50 % increases in human frontal gamma activity, in comparison to spontaneous EEG recordings. The differences on the amount of changes in their study and our current study might be due to different cognitive tasks administered and the participants in the experiments. Many previous studies have showed that brain oscillations are correlated with multiple functions and are highly dependent on tasks, sensation, and individuals .
Aside from investigating channels in the prefrontal cortex, the present study also investigated the visual cortex of the human brain, represented by channels O1 and O2, since visual stimuli was utilized to capture the participant’s attention during the experiment. Higher alpha and gamma power was observed for easy-to-comprehend signs in both channels O1 and O2. Since overall decrease in alpha power indicated increasing demands of attention, alertness, and task load in general , the hard-to-comprehend signs could be associated with reduced alpha power. As to the gamma frequency band, higher gamma power was observed in channels O1 and O2 for the good signs since the occipital lobe is closely linked with visual saliency  and good signs likely evoke a higher visual saliency than the bad signs. However, it is also worthy to note that the increased gamma power observed in the visual cortex for the good signs in the current study might also be caused by other factors, such as stimulus properties or interindividual differences, which are often the causes of diverse findings in brain oscillation-related studies .
It was hypothesized in our study that differences between signs with different comprehensibility levels would exist only in specific brain regions related to visual stimuli processing and cognitive performance. Therefore, to prove this hypothesis, additional analysis was done for channels T7 and T8 located on the auditory cortex, which were supposed to be related with auditory stimuli processing . The result showed no significant difference between the two groups of signs. This finding supported our hypothesis that spatial analysis focusing on brain regions related to attention, cognitive, and visual stimuli processing was an efficient approach to analyze brain oscillatory phenomena when comprehending the symbolic signs.
In coherence to cognitive processing of visual stimuli, laterality is one aspect that often draws interests. In this study, the brain oscillations in the delta, theta, alpha, and gamma tend to be more prominent in the right hemisphere of the visual cortex (O1 and O2). This result is relevant when compared to the previous studies, which mentioned that the right hemisphere of the human brain is specialized for information related to perception of a global shape, while the left hemisphere is specialized to process more detailed information, for example, when two or more modalities were involved . Another study by Goldberg and Costa  also mentioned that the right hemisphere has a greater ability to process novel stimuli. It is noteworthy to mention that the tasks administered in the current study were stimulus-driven (bottom-up) and the obtained laterality pattern may differ when (1) instruction-driven tasks (top-down) are also incorporated into the experiment and (2) two or more modalities are incorporated into the experiment .
There are several limitations inherent in the current study. Firstly, the significant results obtained for several physiological measures investigated in this study were elicited by signs which had been proven to be very different in terms of comprehensibility level (easy to comprehend vs. hard to comprehend). Whether or not the significant measures found in this study can reflect human attention and cognitive processes elicited by signs whose comprehensibility levels are in between is definitely worthy of further investigation. Secondly, the stimuli used in the present study were traffic signs, and they were tested with young participants; whether the findings from this study can be generalizable to other types of safety signs and other populations should be studied in the future. Thirdly, a simple two-group comparison (easy-to-comprehend vs. hard-to-comprehend signs) is used in this study to provide some preliminary evidences that signs of different comprehensibility could induce some changes on simultaneously recorded physiological data. A further study with the addition of a control condition of “no sign” could be conducted to compare with sign conditions so that the effect of displaying sign stimuli on human physiological differences can also be examined. The last, but not least, advanced analysis techniques should be developed to link eye movement to the ongoing EEG for giving more information about what regions of the brain are activated while people engage in sign comprehension.
This study investigated eye movements and brain oscillations to symbolic traffic signs with different comprehensibility. Results showed that the comprehensibility level of traffic signs significantly affects eye movements and EEG spectral power. The harder to comprehend the sign is, the slower the blink rate, the larger the pupil diameter, and the longer the time to first fixation. Noticeable differences on EEG spectral power between easy-to-comprehend and hard-to-comprehend signs can be observed in the prefrontal and visual cortexes of the human brain.
Taken together, these findings demonstrate the potential to integrate physiological measures from eye movements and brain oscillations with existing evaluation methods in assessing the comprehensibility of the symbolic safety signs. Real-time nonintrusive monitoring of human attention and the brain’s spontaneous electrical activity may enable researchers a deeper and advanced understanding the underlying cognitive processes that affect sign comprehension.
Analysis of variance
Communications–human information processing
This study was funded by the Basic Science Research Program through the National Research Foundation of Korea (NRF 2011-0022185, NRF-2014R1A1A2056193) and UNIST Creativity and Innovation Fund (1.150103.01). The authors would like to give special thanks to Woojoo Kim and Hayoung Oh for their help on experimental data acquisition.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Wogalter M S, DeJoy DM, Laughery KR. Organizing theoretical framework: a consolidated communication-human information processing (C-HIP) model. Warnings and Risk Communication, 1999: 15–23.
- Gazmarian JA, Baker D, Williams M, Parker R, Scott TL, Green DC. Health literacy among Medicare enrollees in a managed care organization. J Am Med Assoc. 1999;281:545–51.View ArticleGoogle Scholar
- Kirmizioglu E, Tuydes-Yaman H. Comprehensibility of traffic signs among urban drivers in Turkey. Accid Anal Prev. 2011;45:131–41.View ArticleGoogle Scholar
- King LE. Laboratory comparison of symbol and word roadway signs. Traffic Eng Control. 1971;12:518–20.Google Scholar
- Easterby RS, Zwaga H. Evaluation of public information symbols, ISO Tests: 1975 series (AP report 60). Birmingham, England: Applied Psychology Dept., University of Aston, cited in Lerner, ND & Collins, BL (1980); 1976.Google Scholar
- Dewar RE, Ells JG. The semantic differential as an index of traffic sign perception and comprehension. Hum Factors. 1977;19(2):183–9.PubMedGoogle Scholar
- Wogalter MS, Conzola VC, Smith-Jackson TL. Research-based guidelines for warning design and evaluation. Appl Ergon. 2002;33:219–30.View ArticlePubMedGoogle Scholar
- ISO 9186–1. Graphical symbols -- test methods -- part 1: methods for testing comprehensibility. Geneva: International Standards Organization (ISO); 2007.Google Scholar
- Ng AWY, Chan AHS. Cognitive design features on traffic signs. Eng Lett. 2007;14(1):13–8.Google Scholar
- Siswandari Y, Kim W, Xiong S. Comprehension and redesign of recently introduced water-sport prohibitive symbols in South Korea. Int J Ind Ergon. 2015;50:196–205.View ArticleGoogle Scholar
- Lerner ND, Collins BL. The assessment of safety symbol understandability by different testing methods. Washington: National Bureau of Standards (NBSIR 80–2088); 1980.Google Scholar
- Schultheis H, Jameson A. Assessing cognitive load in adaptive hypermedia systems: physiological and behavioral methods. In: Adaptive hypermedia and adaptive web-based systems. Berlin Heidelberg: Springer; 2004. p. 225–34.View ArticleGoogle Scholar
- Paas FG, Van Merriënboer JJ. Instructional control of cognitive load in the training of complex cognitive tasks. Educ Psychol Rev. 1994;6(4):351–71.View ArticleGoogle Scholar
- Rowe DW, Sibert J, Irwin D. Heart rate variability: indicator of user state as an aid to human-computer interaction. In Proceedings of the SIGCHI conference on Human factors in computing systems. ACM Press/Addison-Wesley Publishing Co. New York, NY, USA; 1998. p. 480–87
- Kramer AF. Physiological metrics of mental workload: a review of recent progress. In: Damos DL, editor. Multiple-Task performance. London: Taylor and Francis; 1991. p. 279–328.Google Scholar
- Duchowski AT. A breadth-first survey of eye-tracking applications. Behav Res Methods Instrum Comput. 2002;34(4):455–70.View ArticlePubMedGoogle Scholar
- Popa L, Selejan O, Scott A, Mureşanu DF, Balea M, Rafila A. Reading beyond the glance: eye tracking in neurosciences. Neurol Sci. 2015. doi:10.1007/s10072-015-2076-6.PubMedGoogle Scholar
- Klimesch W. EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis. Brain Res Rev. 1999;29(2):169–95.View ArticlePubMedGoogle Scholar
- Ergenoglu T, Demiralp T, Bayraktaroglu Z, Ergen M, Beydagi H, Uresin Y. Alpha rhythm of the EEG modulates visual detection performance in humans. Cogn Brain Res. 2004;20(3):376–83.View ArticleGoogle Scholar
- Sauseng P, Klimesch W, Stadler W, Schabus M, Doppelmayr M, Hanslmayr S, et al. A shift of visual spatial attention is selectively associated with human EEG alpha activity. Eur J Neurosci. 2005;22(11):2917–26.View ArticlePubMedGoogle Scholar
- Ma Q, Jin J, Wang L. The neural process of hazard perception and evaluation for warning signal words: evidence from event-related potentials. Neurosci Lett. 2010;483:206–10.View ArticlePubMedGoogle Scholar
- Wolff JS, Wogalter MS. Comprehension of pictorial symbol: effects of context and test method. Hum Factors. 1998;40:173–86.View ArticleGoogle Scholar
- Siswandari, Y., Kim, W., and Xiong, S (2014). Comprehension of newly introduced water-sport prohibitive signs in Korea by westerners. 2014 HFES International Annual Meeting, October 27–31, Chicago, Illinois, USA.
- Ou YK, Yung CL. Effects of sign design features and training on comprehension of traffic signs in Taiwanese and Vietnamese user groups. Int J Ind Ergon. 2012;42:1–7.View ArticleGoogle Scholar
- ANSI Z535.3. Criteria for Safety Symbols National Electrical Manufactures Association. Washington: American National Standards (ANSI); 2002.Google Scholar
- Lowenstein O, Loewenfeld IE. The pupil. Eye. 1962;3:231–67.Google Scholar
- Guyton DL. Prescribing cylinders: the problem of distortion. Surv Ophthalmol. 1977;22:177–88.View ArticlePubMedGoogle Scholar
- Başar E, Başar-Eroglu C, Karakaş S, Schürmann M. Gamma, alpha, delta, and theta oscillations govern cognitive processes. Int J Psychophysiol. 2001;39(2):241–8.PubMedGoogle Scholar
- Yuvaraj R, Murugappan M, Ibrahim NM, Omar MI, Sundaraj K, Mohamad K, et al. On the analysis of EEG power, frequency and asymmetry in Parkinson’s disease during emotion processing. Behav Brain Funct. 2014;10:12.
- Hugdahl K. Lateralization of cognitive processes in the brain. Acta Psychol. 2000;105(2):211–35.View ArticleGoogle Scholar
- Rogers WA, Lamson N, Rousseau GK. Warning research: an integrative perspective. Hum Factors. 2000;42(1):102–39.View ArticlePubMedGoogle Scholar
- Mehta RK, Parasuraman R. Neuroergonomics: a review of applications to physical and cognitive work. Front Hum Neurosci. 2013;7:889.PubMed CentralView ArticlePubMedGoogle Scholar
- Holland MK, Tarlow G. Blinking and thinking. Percept Mot Skills. 1975;41(2):403–6.View ArticleGoogle Scholar
- Telford CW, Thompson N. Some factors influencing voluntary and reflex eyelid responses. J Exp Psychol. 1933;16(4):524.View ArticleGoogle Scholar
- Broadbent, D. E. (1958). The selective nature of learning. Perception and communication. , (pp. 244–267). Elmsford, NY, US: Pergamon Press. -->
- Cardona G, García C, Serés C, Vilaseca M, Gispets J. Blink rate, blink amplitude, and tear film integrity during dynamic visual display terminal tasks. Curr Eye Res. 2011;36(3):190–7.View ArticlePubMedGoogle Scholar
- Beatty J, Kahneman D. Pupillary changes in two memory tasks. Psychon Sci. 1966;5(10):371–2.View ArticleGoogle Scholar
- Pomplun M, Sunkara S. Pupil dilation as an indicator of cognitive workload in human computer interaction. In: Proceedings of the International Conference on HCI. 2003.Google Scholar
- Bailey BP, Iqbal ST. Understanding changes in mental workload during execution of goal-directed tasks and its application for interruption management. ACM Trans Computer-Human Interaction (TOCHI). 2008;14(4):21.View ArticleGoogle Scholar
- Rayner K, Duffy SA. Lexical complexity and fixation times in reading: effects of word frequency, verb complexity, and lexical ambiguity. Mem Cogn. 1986;14(3):191–201.View ArticleGoogle Scholar
- Tatler BW, Baddeley RJ, Gilchrist ID. Visual correlates of fixation selection: effects of scale and time. Vis Res. 2005;45(5):643–59.View ArticlePubMedGoogle Scholar
- Field DJ. Relations between the statistics of natural images and the response profiles of cortical cells. J Opt Soc Am A Opt Image Sci Vis. 1987;4:2379–94.View ArticleGoogle Scholar
- Itti L, Koch C. A saliency-based search mechanism for overt and covert shifts of visual attention. Vis Res. 2000;40(10):1489–506.View ArticlePubMedGoogle Scholar
- Bianchi L. The mechanism of the brain. Edinburgh: E&S Livingstone; 1922.Google Scholar
- Fuster JM. Prefrontal cortex. Boston: Birkhäuser; 1988. p. 107–9.Google Scholar
- Cohen JD, Braver TS, O’Reilly RC. A computational approach to prefrontal cortex, cognitive control and schizophrenia: recent developments and current challenges. Philos Trans R Soc Lond Ser B Biol Sci. 1996;351(1346):1515–27.View ArticleGoogle Scholar
- West RL. An application of prefrontal cortex function theory to cognitive aging. Psychol Bull. 1996;120(2):272.View ArticlePubMedGoogle Scholar
- Miller EK, Cohen JD. An integrative theory of prefrontal cortex function. Annu Rev Neurosci. 2001;24(1):167–202.View ArticlePubMedGoogle Scholar
- Yang Y, Raine A. Prefrontal structural and functional brain imaging findings in antisocial, violent, and psychopathic individuals: a meta-analysis. Psychiatry Res Neuroimaging. 2009;174(2):81–8.View ArticlePubMedGoogle Scholar
- Klimesch W. Memory processes, brain oscillations and EEG synchronization. Int J Psychophysiol. 1996;24(1):61–100.View ArticlePubMedGoogle Scholar
- Klimesch W, Doppelmayr M, Russegger H, Pachinger T. Theta band power in the human scalp EEG and the encoding of new information. Neuroreport. 1996;7(7):1235–40.View ArticlePubMedGoogle Scholar
- Başar E, Başar-Eroğlu C, Karakaş S, Schürmann M. Brain oscillations in perception and memory. Int J Psychophysiol. 2000;35(2):95–124.PubMedGoogle Scholar
- Keil A, Gruber T, Müller MM. Functional correlates of macroscopic high-frequency brain activity in the human visual system. Neurosci Biobehav Rev. 2001;25(6):527–34.View ArticlePubMedGoogle Scholar
- Başar-Eroglu C, Strüber D, Kruse P, Başar E, Stadler M. Frontal gamma-band enhancement during multistable visual perception. Int J Psychophysiol. 1996;24:113–25.
- Rodriguez E, George N, Lachaux JP, Martinerie J, Renault B, Varela FJ. Perception’s shadow: long-distance synchronization of human brain activity. Nature. 1999;397(6718):430–3.View ArticlePubMedGoogle Scholar
- Johnson KL, Nicol TG, Kraus N. Brain stem response to speech: a biological marker of auditory processing. Ear Hear. 2005;26(5):424–34.View ArticlePubMedGoogle Scholar
- Springer SP, Deutsch G. Left brain, right brain. San Francisco: Freeman; 1981.Google Scholar
- Goldberg E, Costa LD. Hemisphere differences in the acquisition and use of descriptive systems. Brain Lang. 1981;14(1):144–73.View ArticlePubMedGoogle Scholar
- Ivry RB, Robertson LC. Two sides of perception. Cambridge: MIT Press; 1998.Google Scholar