June 2016
Volume 57, Issue 7
Open Access
Visual Neuroscience  |   June 2016
A Preliminary Study on Normalized Pattern-Reversal Peripheral Field SSVEPs as a Potential Objective Indicator of Useful Field of View Performance
Author Affiliations & Notes
  • Sheng Tong Lin
    Department of Electrical and Computer Engineering, National University of Singapore, Singapore
    Defence Medical and Environmental Research Institute, DSO National Laboratories, Singapore
  • Lian Kheng Tey
    Defence Medical and Environmental Research Institute, DSO National Laboratories, Singapore
Investigative Ophthalmology & Visual Science June 2016, Vol.57, 3248-3256. doi:10.1167/iovs.15-18439
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Sheng Tong Lin, Lian Kheng Tey; A Preliminary Study on Normalized Pattern-Reversal Peripheral Field SSVEPs as a Potential Objective Indicator of Useful Field of View Performance. Invest. Ophthalmol. Vis. Sci. 2016;57(7):3248-3256. doi: 10.1167/iovs.15-18439.

      Download citation file:


      © 2017 Association for Research in Vision and Ophthalmology.

      ×
  • Supplements
Abstract

Purpose: Currently, the test for useful field of view (UFOV) typically relies on subjective feedback for detecting peripheral targets while attending to a central target. Steady-state visual evoked potential (SSVEP) recorded from the occipital region of the brain has been frequently used as an objective alternative to replace subjective responses. Our aim was to investigate if a short screening test can approximate UFOV performance objectively.

Methods: We recorded SSVEPs during a simple peripheral task and a UFOV task with and without distractors as well as 30-second offline SSVEPs for eight participants. In all tasks, a peripheral annulus ring of checkered reversal pattern was concurrently presented to the participants to elicit SSVEPs.

Results: Orthogonal linear regression analysis found that normalized SSVEP during the UFOV task is predictive of peripheral target detection accuracy (F(1,6) = 16.250, P < 0.01, n = 8) for a task with peripheral distractors. However, this is not the case for a task without distractors (F(1,6) = 0.397, P = 0.552), possibly due to near-perfect accuracy across most participants. Offline normalized SSVEPs also only predicted the peripheral target detection accuracy (F(1,6) = 26.799, P < 0.01) for the UFOV task with peripheral distractors.

Conclusions: Difficult peripheral target detection with distractors is required to differentiate UFOV performance in healthy young adults. The results suggested that offline normalized SSVEPs elicited by 30 seconds of pattern-reversal checkered annulus from the peripheral vision field can identify individuals with good UFOV performance under the stress of a difficult peripheral task.

Typically, the current means of assessing attention is to put individuals through psychophysical tests. These tests rely on subjective feedback and critical reaction-time response; hence time must be set aside for multiple trials to gather accurate averages and to minimize effects of noise on data. When a test has to be repeated to confirm results, the lengthy testing time multiplies with each repetition. Also, a subjective test tends to rely on the motivational level of the individual. In the case in which a subjective test is used to select suitable candidates for sports or military vocation, those mandated by rules to participate may respond suboptimally to avoid good scores while those who are keen but are not confident about passing the test may try guessing tricks to achieve high scores. Recruiters would prefer shorter testing times while avoiding subjectivity by having objective markers. 
In this study, the aim was to investigate steady-state visual evoked potential (SSVEP) as a neurosignal marker that can quickly and objectively assess individuals for their attentional span around their primary gaze, or, to use a more recognized term, the useful field of view (UFOV). Useful field of view is defined as the visual area from which one captures visual information with a single glance without any eye movement.14 It indicates attentional performance toward the peripheral visual field while visual information is concurrently processed from a fixation point. It has been found that driving accidents are correlated with poorer UFOV in elderly drivers.3,4 On the other hand, SSVEP is the neuronal signal typically from the occipital brain region that oscillates at the same frequency as the flicker/flashing rate of the visual stimulus presented, and this oscillation is modulated by how and where visual attention is deployed.5 The SSVEPs can be identified by recording brain signals using electroencephalography (EEG) and then transforming the recorded data into the frequency domain where SSVEP is seen as an increase in amplitude or power in the same frequency as the stimulus presentation frequency.5,6 Relationships between SSVEP and psychophysical assessment of attention have often been studied. Mishra et al.7 found that action game players have increased suppression of SSVEP amplitudes to unattended peripheral stimuli. Many other effects of attention phenomena in modulating SSVEP have also been studied.812 Not surprisingly, objective feedback using SSVEP have become potential objective assessments for retinal functions,13 visual acuity,14 amblyopia,15 stereoscopic vision,16 binocular rivalry,1719 and fatigue.20 In the past two decades, frequency-tagging technique for SSVEPs has been used to study localized attention.2124 This technique requires the visual stimulus of interest to flash at a specific frequency; hence the stimulus is said to be “tagged” with a frequency. The amplitude of SSVEP is modulated by attention to the flashing stimulus regardless of whether eye gaze is directed to the stimulus or not.23,24 Consequently, many studies have used changes in SSVEPs as the neuronal response during experiments in place of or to support the occurrence of specific subjective responses.2124 
While frequency tagging provides an excellent objective alternative to subjective responses in studying overt and covert attention, it is not the optimal way to reduce psychophysical testing time. Perhaps one of the ways to use SSVEPS objectively and at the same time reduce testing time is to look at offline SSVEPs for their potential to approximate attentional performance in actual psychophysical testing, that is, as a fast screening tool. This satisfies the ideal requirements of shorter testing time and objectivity. Thus this preliminary study specifically examined a screening task that is a shorter and reduced version of an experimental task that elicits the relevant SSVEP for predicting UFOV performances. 
The approach is to first identify UFOV task(s) that can show differences in UFOV performance between individuals and at the same time contains a flashing stimulus that elicits SSVEPs effectively. To do so, we designed experimental paradigms with a pattern-reversal annulus ring covering only the peripheral visual field to collect peripheral field SSVEP levels. Since SSVEPs are indicative of attentional resources and UFOV is about the extent of “spotlight” of attention, hypothetically, SSVEPs stimulated by pattern-reversal stimulus at or near the peripheral regions of this spotlight should carry information about the outer boundary of the attentional field, in a way similar to frequency tagging. Hence, we distributed peripheral stimuli just outside this annulus to collect psychophysical data to verify if accuracy in detecting peripheral stimuli could be predicted by this peripheral field SSVEP. If an association between performance and SSVEPs could be demonstrated, then one could extract the relevant elements of the experiment task to design a screening task capable of estimating UFOV performance. 
Methods
Eight healthy participants (6 males, 2 females, ages 18–37) with vision correctable to 6/6 for both far and near in both eyes were recruited. The research protocols were consistent with the 2008 Declaration of Helsinki and all participants gave informed consent before the experiments. The experiments were conducted using a 120-Hz light-emitting diode (LED) monitor (Asus VG278HR; ASUSTeK Computer, Inc., Taipei, Taiwan) with a typical computer equipped with a GTX 570 Nvidia (Santa Clara, CA, USA) graphics card. The monitor screen was positioned 70 cm from the participant's eyes. A commercial off-the-shelf EEG system (ASA LAB waveguard64; ANT-Neuro, Enschede, The Netherlands) was used to record brain signals noninvasively by placing the sensors on the scalp at electrode positions O1, O2, and Oz of the occipital region according to the international 20–20 system standard, referenced to M1 and M2 positions on the mastoid. These electrode positions have been previously demonstrated to collect strong SSVEPs.2527 All electrodes attained an impedance level of 10 kΩ or less before the start of the experiments. All participants completed the following four experiments, and each participant was familiarized using 10 trials of each experiment prior to the commencement of the experiment. 
Experiment 1: Baseline Peripheral Attentional Task
The purpose of this task was to ensure that participants had good covert attention to the peripheral visual field to begin with. Each trial consisted of a baseline phase and an activity phase (Fig. 1). During the baseline phase, participants were asked to fixate on a 0.83-Hz blinking cross (0.4° visual angle in size, 100 ms on, 1100 ms off) at the center of a gray screen. Around the blinking cross was an annulus ring of checkered pattern at 7.5 reversals/s (each checkered box = 1.5°). The mean illuminance of the checkered pattern was 150 cd/m2. The thickness of the ring spanned between 3.5° and 7° from the fixation cross center, leaving a center gray zone extending from the fixating point to the inner edge of the ring. The baseline phase lasted for 10 seconds, followed by the activity phase, which was the same as the baseline phase except for two aspects. Firstly, just outside and around the annulus ring were eight equally distributed small peripheral rings (1.4° visual angle in size). Only one of these circles was randomized to be filled black for 100 ms (onset of peripheral stimulus) every 1.2 seconds in synchrony with the fixation cross. Secondly, the activity phase was randomized to last for 16 to 32 seconds to allow sufficient time to adapt to the peripheral stimuli. The end of the activity phase was indicated by a white-noise screen, which was presented for 2 seconds; then participants had to indicate which one of the peripheral circles last seen was filled black by pressing a response button corresponding to the peripheral circle position (e.g., numeric keypad number 1 = bottom left circle, 9 = top right circle). The purpose was to exclude anticipatory deployment of attention to the periphery. Participants were told to hold their blink during the baseline and activity phase but were encouraged to blink to their satisfaction during the time when they made their response for the peripheral target after the white noise. Participants were also told to take their time to make this response so as to have ample time for blinking and sufficient rewetting of their eyes. Scores were recorded as the percentage of correct responses for the peripheral stimulus that occurred just before the white-noise onset. There were a total of 20 trials. An eye tracker (SMI RED 250; SensoMotoric Instruments Gmb, Teltow, Germany) was used in this study to ensure that participants' eyes were fixating on the central stimulus. Trial results were discarded if a subject was found to fixate beyond 1° away from the central stimulus during the entire trial or when a blink occurred during the last 5 seconds of the baseline or activity phase. 
Figure 1
 
A single trial illustration for experiment 1.
Figure 1
 
A single trial illustration for experiment 1.
Experiment 2: Central–Peripheral Dual Task
The purpose of this task was to investigate the participant's ability to direct simultaneous attention to both central and peripheral stimulus. In other words, it measured the likelihood of detecting a target at a given eccentricity while attending to a central stimulus, thus assessing the breadth of allocated attention in space (i.e., UFOV performance). The trial design was the same as for experiment 1 except that during activity phase, the center fixating stimulus was a small circular ring, 0.167° (10 arcmin) in size (Fig. 2). The circular ring presented a 0.083° (5 arcmin) broken gap every 1.2 seconds in synchrony with the peripheral stimulus onset. The presentation of the gap was randomized to occur at either the up, down, left, or right part of the circular ring. The participant had to actively respond in real time to the gap direction using keyboard arrow buttons, while response to the peripheral stimulus occurred after the white noise when each trial ended. There were a total of 20 trials. The UFOV performance scores were recorded as the percentage of correct responses for the peripheral stimulus that occurred just before the white-noise onset. Note that trial results were excluded from score calculations if a participant did not make a correct response to the last central stimulus during its simultaneous onset with the last peripheral stimulus just before the white noise. 
Figure 2
 
A single trial illustration for experiment 2.
Figure 2
 
A single trial illustration for experiment 2.
Experiment 3: Central–Peripheral Dual Task With Active Distractors
The purpose of this task was to investigate that participant's ability to perform a central–peripheral attention task in the presence of peripheral distractors. The trial design was the same as for experiment 2 except that during the activity phase, extra distractors between the peripheral circles in the form of black spots (1.4° visual angle in size) were presented along with the onset of peripheral stimulus (Fig. 3). 
Figure 3
 
A single trial illustration for experiment 3.
Figure 3
 
A single trial illustration for experiment 3.
Experiment 4: Single-Trial Central Task
In this experiment, there was only one trial, consisting of three phases instead of two. The first phase was the “empty” phase with a central blinking cross identical to those used in the baseline phase against an empty gray background without a checkered annulus. The second phase was the baseline phase, exactly the same as in the previous three experiments. The third phase was the activity phase, exactly the same as the activity phase in experiment 2 except that there was no peripheral stimulus. Hence, all phases had no peripheral task. All phases lasted for a fixed 10 seconds each. The purpose was to examine if the task could be an effective screening method for producing SSVEPs that could predict psychophysical accuracies in experiments 1 to 3. 
Data Processing
The raw signal data from O1, O2, and Oz channels underwent noise removal and band pass between 1 and 40 Hz using the zero-phase Hamming-windowed sinc Finite Impulse Response (FIR) filters28 from the EEGLAB toolbox (Version 13.1.1) for MATLAB R2013b (The MathWorks, Inc., Natick, MA, USA). For each channel and for each experiment, the last 5-second filtered data before the end of baseline phase were selected and transformed into frequency domain from which 7.5-Hz neural oscillation power (in dB = 10Log10[μV2]) was obtained. The same procedure was done for the activity phase in each experiment. Then the obtained power was averaged across all three channels in their respective phases and experiments. The normalized SSVEP (nSSVEP) power for each experiment was calculated by taking the average 7.5-Hz power of the accumulated activity phase minus its accumulated preceding baseline phase's power of the same frequency. Statistical analyses on psychophysical and nSSVEPs were conducted using GraphPad Prism 7.0 (La Jolla, CA, USA). 
Results
All participants fixated within 1° away from the stimulus center in all trials, and no blinks during the last 5-second window during the baseline and activity phases (where SSVEP signals were used for processing) were detected by the eye tracker. All participants attained a perfect score for responses to the last central stimulus in experiments 2 and 3 during their simultaneous onset with the last peripheral stimulus before the white noise; hence no trials were excluded. In each of experiments 1 to 3, the activity phase lasted for an average of 24 seconds. 
Behavioral Results
Responses to central stimuli were consistent with good central fixation behavior for all participants in this preliminary study. Hence, all correct responses here were simplified to consider only correct responses to peripheral stimulus. On the other hand, accuracy scores differed between different experiments for response to peripheral stimuli/targets. 
Figure 4 shows the percentage accuracy for peripheral targets for each experiment. All participants scored nearly perfect if not perfect in experiments 1 and 2, except for one who scored 70% in experiment 2. The accuracy scores started to spread within the participants in experiment 3, where the task difficulty was the greatest with the peripheral distractors. Repeated measures ANOVA with Greenhouse-Geisser correction indicated a significant difference in percentage correct for peripheral targets among the experiments (F(1.122, 7.855) = 32.471, P < 0.01). A post hoc test with Bonferroni correction revealed that there were significant differences only between experiment 1 and experiment 3 (mean differences = 0.544, 95% confidence interval [CI] [0.257, 0.831], P < 0.01) and experiments 2 and 3 (mean differences = 0.5, 95% CI [0.224, 0.776], P < 0.01). There were no significant differences between experiments 1 and 2 (mean differences = 0.044, 95% CI [−0.037, 0.124], P = 0.4). The analysis suggested that experiments 1 and 2 were easily accomplished and exhibited near-ceiling effects for all participants except one. Hence, these two experiments did not show difference in performance between the sampled normal individuals, in contrast to experiment 3, which showed a spread of performance differences between individuals. It was also noted that the top three performers in this experiment were a semiprofessional volleyball player, a competitive basketball player, and a frequent video gamer (role-playing game [RPG] player), respectively, in descending order of top scores, while the rest of the participants were office workers with negligible engagement in sports and video games. 
Figure 4
 
Percentage accuracy for peripheral targets in experiments 1 to 3 for (a) individual and (b) group mean. Error bar: SE.
Figure 4
 
Percentage accuracy for peripheral targets in experiments 1 to 3 for (a) individual and (b) group mean. Error bar: SE.
Repeated measures ANOVA indicated no significant difference in nSSVEPs between the experiments (F(2,14) = 0.607, P = 0.559). Figure 5 shows the mean values of nSSVEP for experiments 1 to 3. However, looking at the small sample size here and the high variance observed in Figure 5, there is a possibility of decreasing nSSVEPs with increasing stimulus difficulties given a case of low statistical power. Nevertheless, until proven with a larger sample size, the difference in nSSVEPs between experiments 2 and 3 seems to be small if even existent. 
Figure 5
 
Mean normalized SSVEPs in experiments 1 to 3. Error bar: SE.
Figure 5
 
Mean normalized SSVEPs in experiments 1 to 3. Error bar: SE.
Regression Analysis Results
Orthogonal linear regression (Deming regression) and Pearson correlation analyses were done to study the relationship between peripheral target accuracies and the corresponding nSSVEPs for each experiment (Fig. 6). 
Figure 6
 
Percentage accuracy versus normalized SSVEPs in experiment 1 (a), 2 (b), and 3 (c).
Figure 6
 
Percentage accuracy versus normalized SSVEPs in experiment 1 (a), 2 (b), and 3 (c).
Regression analysis showed a significant relationship between target accuracies in experiment 3 and its nSSVEPs (F(1,6) = 16.250, P < 0.01), and their correlation coefficient, r = 0.854, indicates a strong positive relationship. There was no significant relationship and correlation found between the accuracy and nSSVEPs for experiment 1 (F(1,6) = 0.065, r = 0.103, P = 0.808) and experiment 2 (F(1,6) = 0.397, r = 0.249, P = 0.552). It was also noted that the previously mentioned top three performers had nSSVEPs larger than the rest of the participants. 
The strong relationship between nSSVEPs and behavioral performance in experiment 3 triggers potential questions. If nSSVEPs during a dual task with distractors are indeed a strong indicator of performance and there is a high chance that different peripheral target difficulties (with or without distractors) have a small effect on nSSVEP as seen in Figure 5 (experiment 2 versus experiment 3), can nSSVEPs derived from a simple central stimulus task approximate dual-task performance in experiment 3? And given the stability of the 7.5-Hz SSVEPs strength here, can such a task be shortened to a single trial to examine the possibility of a neuronal screening procedure to quickly approximate UFOV performance? Hence, experiment 4 was designed to investigate the possibility of such a screening method. 
Experiment 4 included an “empty” phase for comparison of its SSVEPs with those in the baseline and activity phase. A very significant difference in SSVEPs was observed between the three phases (F(2,14) = 6.802, P < 0.01). Note that this empty phase was left out in experiments 1 to 3 to shorten their testing duration. A post hoc test with Bonferroni correction revealed that there were significant differences between baseline phase and empty phase (mean differences = 4.275, 95% CI [1.542, 7.007], P < 0.01). These results ensure that SSVEPs were actively produced by the presentation of checkered annulus rings during the baseline phase. There were no significant differences between the activity phase and empty phase (mean differences = 2.645, 95% CI [−1.95, 7.240], P = 0.344) or between activity phase and baseline phase (mean differences = −1.630, 95% CI [−5.033, 1.773], P = 0.534). The differences between activity phase and the other two phases are not significant, possibly due to the lack of statistical power with the small sample size. The mean SSVEPs for all phases in experiment 4 are shown in Figure 7
Figure 7
 
Mean SSVEPs for “empty,” baseline, and activity phase for experiment 4. Error bar: SE.
Figure 7
 
Mean SSVEPs for “empty,” baseline, and activity phase for experiment 4. Error bar: SE.
The mean nSSVEP in the single-trial experiment 4 was −1.63 dB, and there were no significant differences between the nSSVEPs from experiments 3 (Median = −1.755) and 4 (Median = −2.03), Wilcoxon test, z = 0.0, P = 0.99. This suggests that nSSVEPs were not substantially affected by the occurrence of peripheral targets. There is also the possibility that the nSSVEPs between experiments 2 and 3 might be different by a small percentage compared to the bigger percentage difference in UFOV performance, but the significance cannot be determined with the small sample size in this preliminary study. 
Orthogonal linear regression analysis was conducted to predict percentage accuracy of peripheral targets in all three experiments based on the single-trial nSSVEP in experiment 4 (Fig. 8). Pearson correlation analysis was also done to study the strength of the relationship between peripheral target accuracies in experiments 1 to 3 and the nSSVEPs from experiment 4. 
Figure 8
 
Percentage accuracy in experiments 1 (a), 2 (b), 3 (c) versus normalized SSVEPs in experiment 4.
Figure 8
 
Percentage accuracy in experiments 1 (a), 2 (b), 3 (c) versus normalized SSVEPs in experiment 4.
Regression analysis showed a significant relationship between experiment 1 and single-trial nSSVEP (F(1, 6) = 8.421, P = 0.027), with r = 0.764 indicating a strong positive relationship. There was no significant relationship found between percentage accuracy for experiment 2 and single-trial nSSVEP (F(1,6) = 1.341, r = 0.427, P = 0.291). A significant relationship was found between percentage accuracy in experiment 3 and single-trial nSSVEP (F(1,6) = 26.800, P < 0.01), with r = 0.904 indicating a strong positive relationship similar to what was shown in the regression analysis in Figure 6c. 
Discussion
The first experiment demonstrated that all participants had good peripheral vision to start with. The second experiment demonstrated that all participants could perform the central–peripheral task well. The third experiment increased peripheral task difficulty by throwing in peripheral distractors on top of the central–peripheral dual attention task to examine how individuals handle multiple pieces of information closer to the boundaries of UFOV. The peripheral distractors in experiment 3 were shown to be effective in increasing peripheral task difficulty, providing a wider spread of performance differences between participants. Noticeably, the relationship between UFOV performance and nSSVEP in experiment 3 is driven by two clusters of data in the current small sample size. Hence, at the moment, implementing peripheral distractors in experiment 3 could differentiate the participants' performance into only two different levels. Interestingly, peripheral task difficulty was not accompanied by a corresponding significant difference in nSSVEPs when comparing experiments 3 and 2 (with and without peripheral distractor comparison) and experiments 3 and 4 (with and without peripheral task comparison). This could be a case of a small difference in nSSVEPs between experiments 2 and 4 compared to a larger difference in their UFOV performances; this small difference could not reach significance due to small sample size. Nevertheless, experiment 4, a case of an extremely easy single-trial central task without any peripheral task, produced nSSVEPs similar to those in experiment 3 in identifying good UFOV performers over poorer ones for a difficult peripheral task. This suggested that the single-trial nSSVEP has applicable value as a screening tool. Another strong contributing factor to the success of experiment 4 is how the baseline was designed. Both SSVEPs were stimulated and acquired during baseline phase (no task) and activity phase (task). Taking the signal differences between them to associate with effects arising from the corresponding task performance is more meaningful than purely having an SSVEP extracted and baselined to a controlled condition without any checkered stimulus. 
Another observation in experiment 3 was that most nSSVEPs in this study were negative in value, and those with good UFOV had positive nSSVEP values. This may be attributed to the fact that good UFOV performers in experiment 3 maintained their attention better under the stress of a difficult peripheral task, hence suffered less peripheral field SSVEP attrition when executing central–periphery dual detection during the activity phase. Given that SSVEP strength is relatively stronger for attended stimuli,5,6 this explanation is plausible. 
The last observation worth noting concerns the top three performers in experiment 3 (mentioned earlier) who included two sportsmen and a video gamer. The effect of video gaming has previously been demonstrated, with specific differences in the SSVEPs of video gamers compared to normal individuals.7,29 Krishnan et al.29 found that RPG players have increased SSVEPs for attended flashing stimuli accompanied by an increased detection accuracy in a multiregion visual search task. Hence, they are thought to be deploying some form of attention enhancement. Findings for the participant in experiment 3 who was a frequent RPG player seem to concur with the findings of Krishnan et al.29 as her relatively higher nSSVEP is accompanied by better peripheral detection accuracy than most of the other participants. Unfortunately, this preliminary study did not have participants who were active/frequent action video game players. It is known that action video gamers perform better in UFOV tasks,30,31 and they probably ignore distractors better using a unique neural strategy revealed by the suppression of SSVEPs from irrelevant flashing stimuli.7 Therefore, it will certainly be of interest to repeat experiments 3 and 4 with action video gamers. 
On the other hand, the influence of sports training on SSVEPs is less well studied. Also, the findings in the literature regarding better UFOV in ball sport athletes are rather mixed. Matos and Godinho32 and Memmert et al.33 highlighted that UFOV for these sportsmen is not significantly better than in nonsportsmen. Memmert et al. suggested that the commercial version of the UFOV test is designed for screening elderly persons for risk of driving accidents and may not have the sensitivity to find differences between sportsmen and nonsportsmen among healthy young individuals. However, Schwab and Memmert34 found that sports vision training improves UFOV for hockey players but did not report sports performance to suggest any corresponding improvement in actual sports activity. A study by Störmer et al.27 suggested that individuals who are better at tracking multiple flashing moving targets within their visual field exhibit higher SSVEPs than those elicited by the frequency-tagged targets. If these individuals are analogous to ball game sportsmen, then there is a good possibility that better UFOV in sportsmen could be accompanied by higher levels of SSVEPs. 
In conclusion, in this preliminary study, we have demonstrated that changes in peripheral field SSVEP can potentially be used to identify an individual with good UFOV performance in the presence of distractors. Particularly, as demonstrated in experiment 4, it is possible that a short 30-second objective SSVEP test can differentiate those individuals with good UFOV from those with poor UFOV. For the purpose of a quick screening test, this could be a reasonable trade-off of assessment accuracy for shorter testing time. While the testing time is short, the setup time for experiment 4 requires approximately 3 minutes to place the six electrodes (O1, O2, Oz, M1, M2, GND) and to ensure their impedance quality. This setup time can be reduced in the future when dry electrodes with conductivity nearly equivalent to that of wet types are available. Future work should look at larger sample size to avoid the possibility of type 1 error, as well as various eccentricities of peripheral stimulus to estimate the distribution of attentional performance toward the periphery. Extensive efforts in testing and retesting of the screening task in experiment 4 must be made to ascertain its reliability in predicting UFOV performance. It will also be worthwhile to recruit sportsmen and video gamers to study their differences using current experimental paradigms. 
Acknowledgments
Supported by Future Systems and Technology Directorate (FSTD), Singapore. Grant number 9012102527. 
Disclosure: S.T. Lin, None; L.K. Tey, None 
References
Sanders AF. Some aspects of the selective process in the functional visual field. Ergonomics. 1970; 13: 101–117.
Ball K, Beard B, Roenker D, Miller R, Griggs D. Age and visual search: expanding the useful field of view. J Opt Soc Am A. 1988; 5: 2210–2219.
Ball K, Owsley C, Stalvey B, Roenker DL, Sloane ME, Graves M. Driving avoidance and functional impairment in older drivers. Accid Anal Prev. 1988; 30: 313–322.
Ball KK, Wadley VG, Edwards JD. Advances in technology used to assess and retrain older drivers. Gerontechnology. 2002; 1: 251–261.
Beverina F, Palmas G, Silvoni S, Piccione F, Giove S. User adaptive BCIs: SSVEP and P300 based interfaces. PsychNology Journal. 2003; 1: 331–354.
Herrmann CS. Human EEG responses to 1–100 Hz flicker: resonance phenomena in visual cortex and their potential correlation to cognitive phenomena. Exp Brain Res. 2001; 137: 346–353.
Mishra J, Zinni M, Bavelier D, Hillyard SA. Neural basis of superior performance of action videogame players in an attention-demanding task. J Neurosci. 2011; 31: 992–998.
Morgan ST, Hansen JC, Hillyard SA. Selective attention to stimulus location modulates the steady-state visual evoked potential. Proc Natl Acad Sci U S A. 1996; 93: 4770–4774.
Müller MM, Teder-Sälejärvi W, Hillyard SA. The time course of cortical facilitation during cued shifts of spatial attention. Nat Neurosci. 1998; 1: 631–634.
Müller MM, Picton TW, Valdes-Sosa P, Riera J, Teder-Sälejärvi WA, Hillyard SA. Effects of spatial selective attention on the steady-state visual evoked potential in the 20–28 Hz range. Brain Res Cogn Brain Res. 1998; 6: 249–261.
Belmonte M. Shifts of visual spatial attention modulate a steady-state visual evoked potential. Brain Res Cogn Brain Res. 1998; 6: 295–307.
Kim YJ, Grabowecky M, Paller KA, Muthu K, Suzuki S. Attention induces synchronization-based response gain in steady-state visual evoked potentials. Nat Neurosci. 2007; 10: 117–125.
Herbik A, Geringswald F, Thieme H, Pollmann S, Hoffmann MB. Prediction of higher visual function in macular degeneration with multifocal electroretinogram and multifocal visual evoked potential. Ophthalmic Physiol Opt. 2014; 34: 540–551.
Mackay AM, Bradnam MS, Hamilton R, Elliot AT, Dutton GN. Real-time rapid acuity assessment using VEPs: development and validation of the step VEP technique. Invest Ophthalmol Vis Sci. 2008; 49: 438–441.
Baker DH, Simard M, Saint-Amour D, Hess RF. Steady-state contrast response functions provide a sensitive and objective index of amblyopic deficits. Invest Ophthalmol Vis Sci. 2015; 56: 1208–1216.
Johansson B, Jakobsson P. Fourier analysis of steady-state visual evoked potentials in subjects with normal and defective stereo vision. Doc Ophthalmol. 2000; 101: 233–246.
Zhang P, Jamison K, Engel S, He B, He S. Binocular rivalry requires visual attention. Neuron. 2011; 71: 362–369.
Sutoyo D, Srinivasan R. Nonlinear SSVEP responses are sensitive to the perceptual binding of visual hemifields during conventional “eye” rivalry and interocular “percept” rivalry. Brain Res. 2009; 1251: 245–255.
Jamison KW, Roy AV, He S, Engel SA, He B. SSVEP signatures of binocular rivalry during simultaneous EEG and fMRI. J Neurosci Methods. 2015; 243: 53–62.
Cao T, Wan F, Wong CM, da Cruz JN, Hu Y. Objective evaluation of fatigue by EEG spectral analysis in steady-state visual evoked potential-based brain-computer interfaces. Biomed Eng Online. 2014; 13: 28.
Ding J, Sperling G, Srinivasan R. Attentional modulation of SSVEP power depends on the network tagged by the flicker frequency. Cereb Cortex. 2006; 16: 1016–1029.
Toffanin P, de Jong R, Johnson A, Martens S. Using frequency tagging to quantify attentional deployment in a visual divided attention task. Int J Psychophysiol. 2009; 72: 289–298.
Müller MM, Malinowski P, Gruber T, Hillyard SA. Sustained division of the attentional spotlight. Nature. 2003; 424: 309–312.
Malinowski P, Fuchs S, Müller MM. Sustained division of spatial attention to multiple locations within one hemifield. Neurosci Lett. 2007; 414: 65–70.
Pastor MA, Artieda J, Arbizu J, Valencia M, Masdeu JC. Human cerebral activation during steady-state visual-evoked responses. J Neurosci. 2003; 23: 11621–11627.
Zhang Y, Jin J, Qing X, Wang B, Wang X. LASSO based stimulus frequency recognition model for SSVEP BCIs. Biomed Signal Process Control. 2012; 7: 104–111.
Störmer VS, Winther GN, Li SC, Andersen SK. Sustained multifocal attentional enhancement of stimulus processing in early visual areas predicts tracking performance. J Neurosci. 2013; 33: 5346–5351.
Widmann A, Schröger E, Maess B. Digital filter design for electrophysiological data–a practical approach. J Neurosci Methods. 2014; 250: 34–46.
Krishnan L, Kang A, Sperling G, Srinivasan R. Neural strategies for selective attention distinguish fast-action video game players. Brain Topogr. 2013; 26: 83–97.
Green CS, Bavelier D. Action video game modifies visual selective attention. Nature. 2003; 423: 534–537.
Hubert-Wallander B, Green CS, Bavelier D. Stretching the limits of visual attention: the case of action video games. Wiley Interdiscip Rev Cog Sci. 2011; 2: 222–230.
Matos R, Godinho M. Influence of sports practice in the useful field of vision in a simulated driving test. In: Active Lifestyles: The Impact of Education and Sport, Book of Abstracts of AIESEP 2005 World Conference. Lisbon. 2005; 262: 557–566.
Memmert D, Simons DJ, Grimme T. The relationship between visual attention and expertise in sports. Psychol Sport Exerc. 2009; 10: 146–151.
Schwab S, Memmert D. The impact of a sports vision training program in youth field hockey players. J Sports Sci Med. 2012; 11: 624.
Figure 1
 
A single trial illustration for experiment 1.
Figure 1
 
A single trial illustration for experiment 1.
Figure 2
 
A single trial illustration for experiment 2.
Figure 2
 
A single trial illustration for experiment 2.
Figure 3
 
A single trial illustration for experiment 3.
Figure 3
 
A single trial illustration for experiment 3.
Figure 4
 
Percentage accuracy for peripheral targets in experiments 1 to 3 for (a) individual and (b) group mean. Error bar: SE.
Figure 4
 
Percentage accuracy for peripheral targets in experiments 1 to 3 for (a) individual and (b) group mean. Error bar: SE.
Figure 5
 
Mean normalized SSVEPs in experiments 1 to 3. Error bar: SE.
Figure 5
 
Mean normalized SSVEPs in experiments 1 to 3. Error bar: SE.
Figure 6
 
Percentage accuracy versus normalized SSVEPs in experiment 1 (a), 2 (b), and 3 (c).
Figure 6
 
Percentage accuracy versus normalized SSVEPs in experiment 1 (a), 2 (b), and 3 (c).
Figure 7
 
Mean SSVEPs for “empty,” baseline, and activity phase for experiment 4. Error bar: SE.
Figure 7
 
Mean SSVEPs for “empty,” baseline, and activity phase for experiment 4. Error bar: SE.
Figure 8
 
Percentage accuracy in experiments 1 (a), 2 (b), 3 (c) versus normalized SSVEPs in experiment 4.
Figure 8
 
Percentage accuracy in experiments 1 (a), 2 (b), 3 (c) versus normalized SSVEPs in experiment 4.
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×