Applying eye-fixation-related brain potentials for a concealed information test

doi:10.21203/rs.3.rs-3008560/v1

Download PDF

Short Report

Applying eye-fixation-related brain potentials for a concealed information test

https://doi.org/10.21203/rs.3.rs-3008560/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

A concealed information test (CIT) is a memory detection technique that is commonly used in criminal investigations. Although event-related brain potentials (ERPs) have been used as a measure in CIT studies, they are not typically measured in criminal investigations due to their susceptibility to noise resulting from eye movement. The present study focused on eye-fixation-related brain potentials (EFRPs), which are a type of brain potentials that are time-locked to the onset of eye-fixations, in order to avoid contamination from eye movement, and investigated whether EFRPs could be a useful measure for CIT. Nineteen participants selected a photograph of one person out of five and created a fictional story about a crime they committed with the person in the selected photograph. The participants provided information about the selected photograph to an experimenter in a no-secret condition, but they did not disclose any information about it in a concealed condition. After the story was created a photograph of a person and a target were presented continuously on alternating sides of the monitor and participants were asked to follow them with their eyes and to press a button when their eyes fixed on the target. The amplitude of P1 elicited by the selected photograph was significantly larger than that elicited by unselected ones only in the concealed condition. The result demonstrates that spatial attention to incoming visual information was modulated by the intent to conceal and indicates that EFRPs are a useful measure in CIT.

A concealed information test (CIT) is a memory detection technique that is commonly used in criminal investigations [1]. During this test, examinees are presented with both crime-relevant and crime-irrelevant information. The crime-related information is presented as a stimulus known only to those involved in the crime, while responses of the autonomic nervous system (ANS) and/or central nervous system (CNS) are measured. If the examinees can distinguish between the information (i.e., if they have a memory of the crime-related information), their responses to the information will differ.

In criminal investigations, ANS responses are commonly measured [2], but event-related brain potentials (ERPs), which are brain potentials that are time-locked to the onset of various events, are not commonly used for this purpose, despite being the most common measure of CNS in CIT studies. This is partly due to ERPs being susceptible to artifacts caused by eye movements. According to a report that measured ERPs in criminal investigations [3], over 20% of examinees’ data was difficult to analyze due to eye movement contamination. Despite this, CIT measuring ERPs can detect individuals with a crime-related memory as well as or better than CIT measuring ANS [4]. Consequently, proposing methods to address eye movement contamination could make brain potentials a valuable index for CIT in the future.

The current study aimed to address the issue of eye movement contamination by utilizing eye-fixation-related brain potentials (EFRPs). EFRPs are a type of brain potentials that are time-locked to the onset of eye-fixations (i.e., the offset of saccades) [5]. Thus, the use of EFRPs might help reduce artifacts caused by eye movements as they are calculated using eye movements. Since the responses in the EFRP waveform are similar to those in the ERP waveform, previous ERP studies are useful in understanding result of EFRP responses. Therefore, in CIT, it is possible that EFRPs could be a useful measure that is not affected by eye movement contamination.

To investigate whether EFRPs are a useful measure to detect crime-related memory in the future, we conducted a modified CIT task. The CIT procedures that measure ANS and ERPs differ from each other. To explore the potential use of EFRPs in actual criminal investigations, we conducted an experiment in the former paradigm as much as possible. However, as a basic study to examine the usefulness of EFRPs, this study was purposely conducted in a paradigm that produced eye movements, as described below. In the experiment, participants selected a photograph of one person out of five and created a fictional story about a crime they committed with the person in the selected photograph. The participants told information about the selected photograph to an experimenter in a no-concealed condition, but they did not disclose any information about it in a concealed condition. Subsequently, the participants were presented the selected (or an unselected) photograph (probability = 0.89) and a black circle as a target (probability = 0.11) continuously on alternating sides of the monitor while their electroencephalogram (EEG) and electrooculography (EOG) were recorded. They were asked to follow the target with their eyes and to press the button when their eyes fixed on the target. After the presentations, they were asked whether the person in the photograph displayed was an accomplice, and they had to deny it. As a previous ERPs study has shown, the amplitude of P300 at parietal site elicited by crime-relevant information was larger than that elicited by crime-irrelevant one when the participants disclosed the crime-relevant information to the experimenter because of enhanced attention to one [6]. However, in the current study, we presented selected (or unselected) photographs continuously in a trial, whereas in previous ERP study, the crime-relevant and crime-irrelevant information were presented in a random order. It was reported that continuous presentation of stimuli that elicit P300 leads to attenuation of this amplitude [7-8]. Therefore, we expect that results similar to the ERP study are unlikely to be obtained. Rather, when the EFRPs are used as a measure, an attentional modulation to continuously presented stimuli might be reflected in a P1 amplitude because time-course changes in attention to a stimulus are reflected in the P1 amplitude [9-11]. Thus, we expected that the P1 amplitude in the concealed condition would be different for selected and unselected photographs.

2.1. Participants

To determine the sample size, we calculated the power (1-β) needed to detect a two-way interaction using PANGEA (https://jakewestfall.shinyapps.io/pangea/). Assuming that we recruited 16 participants and obtained at least 20 epochs of EFRPs, the power was found to be 0.81 (effect size (d) = 0.45, variance of error = 0.33, variance of two-way interaction = 0.08, repetitions = 20). To meet this criterion, we recruited 19 individuals between the ages of 19 and 24 (Mage = 20.6, SDage = 1.1). Three participants were excluded from the analysis (see 2.4. Recording and analyses). All participants were right-handed with corrected vision, and they received an honorarium of 4,000 Japanese yen in the form of an Amazon gift card. This experiment was approved by the ethics board of Bunkyo Gakuin University and was carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki). Written informed consent was obtained from all participants.

2.2. Stimuli

Twenty color photographs of people were used as a stimulus set. Each photograph showed a different person. The persons were Asian, and their face, eyes, and body were facing forward. Their visual angle was approximately 8º (w) × 8º (h). A set of five photographs, randomly selected from the set of 20, was used in each of the conditions described below. The set used in one condition was not used in the other condition. A black circle figure was also used as a target, with the visual angle of approximately 2º (w) × 2º (h).

2.3. Procedure

The participants sat in a chair, in front of a computer display, keyboard, and mouse. Before an EEG was recorded, the participants did a story creation task, where they were presented with file names of photographs in the set. Each file name was assigned a random number between 1 and 5. Each participants selected and opened only one photograph on their own. Following the selection, the participants were asked to create a fictional story of a crime they committed with person in the photo as an accomplice, including details such as the accomplice's name and the specifics of the crime, all within a five-minute time limit. For this procedure, two conditions were set: a no-secret condition and a concealed condition. In the no-secret condition, the participants disclosed the number of the selected photograph, and information about the accomplice and crime to the experimenter. In the concealed condition, they did not disclose any information to the experimenter. The computer operation during this task was monitored in a separate room, without the participants' knowledge, allowing the experimenter to record the number of the selected photograph in both conditions.

Following the story creation task, participants performed a modified CIT task while their EEG was being recorded (Figure. 1). The selected (or an unselected) photograph and a target were presented continuously on alternating sides of the display. The photograph was randomly selected from the set used in the story creation task, and this photograph and target were presented 16 and 2 times, respectively, in random order. Each stimulus was presented for 1,200 ms, and no inter-stimulus-interval was set. The participants were instructed to follow the stimuli with their eyes and press a button for the target. After 18 presentations, participants were asked whether the presented person was the accomplice and were told to always answer "No." The same procedure was repeated for the other photographs in the set, and the procedure from stimulus presentation to question and answer was performed twice for each photograph. In the concealed condition, participants were instructed to conceal their accomplices successfully so that their brain potentials would not indicate their guilt. After completing one condition, a five-minute break was taken, and the other condition was performed.

2.4. Recording and analyses

EEG was recorded using an actiCHamp amplifier (Brain Products, Germany) and Ag/AgCl electrodes from six sites (Fp1, Fp2, Fz, Cz, Pz, and Oz) with an electrode cap (Easy-cap (Asian cut), Brain Products, Germany). A ground and a reference electrode were placed on the nose tip and AFz, respectively. EOG was recorded from electrodes placed on the outer canthus of both eyes. Electrode impedances were kept below 10 kΩ. The sampling rate was 1,000 Hz with 0.01 - 100 Hz bandpass filter.

To select the participants who looked at the photographs appropriately, individuals whose rate of the response to target (pressing the button within 1,000 ms after eye fixation to it) was 90% or higher in each condition were included in the analysis. Two participants who did not meet this criterion and one participant who fell asleep during the experiment were excluded from the following analyses.

The EEGLAB toolbox [12], ERPLAB toolbox [13], and EYE-EEG extension [14-15] in MATLAB (The MathWorks, Inc., USA) were used for the EFRPs analyses. The saccade offsets were detected from right EOG and Fp2 signals using the EYE-EEG extension. Missed detections were corrected manually. All physiological responses were filtered using a noncausal Butterworth filter (half-amplitude cutoff at 30 Hz, 12 dB/oct roll-off). Based on the detected events, epochs of 700 ms, including 200 ms before the eye-fixations, were extracted. Baseline correction was performed in reference to before the eye-fixation. Epochs in which the EEG signal exceeded ± 100 μV were excluded. The average number of remaining epochs for selected and unselected photographs were 28 and 104 in the concealed condition and 28 and 107 in the no-secret one. No subject had fewer than 24 epochs, and we determined from previous reports [10,16] that all data could be analyzed for P1 and P300.

A positive wave at the occipital site predominance was observed around 100 ms after the eye fixation. This feature indicated that this response was P1 [5,17]. Although a distinct P300 was not confirmed from the EFRP waveforms at parietal site, late positive component (LPC) differed between photographs. Based on previous reports [5-6,17] and obtained waveforms, the amplitudes of P1 at Oz (80 – 100 ms) and LPC at Pz (250 – 450 ms) were analyzed.

In addition, it was possible that each EFRP amplitude was affected by EOG because EFRPs were calculated using this. To check this influence, mean EOG amplitudes were calculated from the same latency window of the P1 and LPC amplitudes. Then, the EOG waveforms caused by saccades from right to left (EOG_RtoL) and vice versa (EOG_LtoR) were separately calculated because the responses would cancel each other out if they were averaged. Two-way repeated measures ANOVAs were conducted for each amplitude: Condition (no-secret and concealed conditions) × Item (selected and unselected photographs). When significant interaction was obtained, a simple main effect test was conducted. We reported η_p²as an effect size and the significance level was set at .05.

The EFRP waveforms at Fz, Cz, Pz, and Oz are shown in Figure 2a. It is important that no participant was excluded from the EFRP analyses because of eye movement contamination. The grand average waveforms obtained from all participants show that the P1 amplitudes at Oz for each photograph differed in the concealed condition. The LPC amplitude at Pz elicited by the selected photograph were slightly larger than that elicited by the unselected one. For the P1 amplitudes (80 – 100 ms at Oz, Figure 2b), the interaction of Condition × Item [F (1, 15) = 7.02, p = .02, η_p² = .32] was significant. The post hoc comparisons revealed that the P1 amplitude elicited by the selected photograph was significantly larger than that elicited by the unselected one in the concealed condition (p = .02). The main effects of Condition [F (1, 15) = 0.45, p = .51, η_p² = .03] and Item [F (1, 15) = 4.70, p = .05, η_p² = .24] were not significant. For the LPC amplitudes (250 – 450 ms at Pz, Figure 2c), the main effect of Item [F (1, 15) = 1.51, p = .24, η_p² = .09], Condition [F (1, 15) = 3.84, p = .07, η_p² = .20], and their interaction [F (1, 15) = 0.71, p = .41, η_p² = .05] were not significant.

The EOG_RtoL and EOG_LtoR are shown in the right side of Figure 2a. The figure shows that the EOG waveforms for each stimulus overlapped. For the EOG amplitudes during the P1 latency window, the main effects of Condition [EOG_LtoR: F (1, 15) = 2.14, p = .16, η_p² = .13, EOG_RtoL: F (1, 15) = 0.37, p = .55, η_p² = .02], Item [EOG_LtoR: F (1, 15) = 0.11, p = .74, η_p² = .01, EOG_RtoL: F (1, 15) = 0.47, p = .50, η_p² = .03], and their interaction [EOG_LtoR: F (1, 15) = 0.22, p = .65, η_p² = .01, EOG_RtoL: F (1, 15) = 2.79, p = .12, η_p² = .16] were not significant. For the EOG amplitudes during the LPC latency window, the main effects of Condition [EOG_LtoR: F (1, 15) = 2.19, p = .16, η_p² = .13, EOG_RtoL: F (1, 15) = 0.84, p = .37, η_p² = .05], Item [EOG_LtoR: F (1, 15) = 0.03, p = .86, η_p² = .002, EOG_RtoL: F (1, 15) = 0.33, p = .57, η_p² = .02], and their interaction [EOG_LtoR: F (1, 15) = 0.26, p = .62, η_p² = .02, EOG_RtoL: F (1, 15) = 1.66, p = .22, η_p² = .10] were not significant. Thus, the effects on P1 and LPC were not significantly affected by the eye movements.

The current study investigated whether EFRPs could be a useful measure for CIT. According to a previous report, data from more than 20% of examinees were difficult to analyze when ERPs were used in criminal investigations because of eye movement contamination [3]. However, in our study, no participants were excluded from the analysis due to eye movement contamination. This result suggests that EFRPs might serve as a valuable alternative to address the limitations associated with ERPs in CIT.

As we expected, the P1 amplitude, rather than the P300, elicited by the selected photograph was significantly larger than that elicited by the unselected one only in the concealed condition. The result of P1 amplitude suggests that the participants discriminated between selected and unselected photographs and that their spatial attention was modulated by the intent to conceal. A previous ERPs study showed that P1 amplitude elicited by objects presented at attended location is significantly larger than that elicited by objects presented at ignored one [18]. This P1 modulation reflects an early attentional sensory gain control [19]: the P1 amplitude increases as more attentional resources are allocated to incoming visual information. Similar results were reported by EFRPs studies [10,17,20]. In our data, this modulation was confirmed only in the concealed condition. This result suggests that more attentional resources were allocated to information that had to be concealed. In criminal investigations, CIT is conducted on individuals who might be concealing crimes. Thus, based on these results, the P1 amplitude in EFRPs might be a useful measure that can detect crime-related memory in individuals with the intent to conceal.

The reason why a distinct P300 did not occur might be due to the experimental procedure. As described at the Introduction, it was reported that continuous presentation of stimuli that elicit P300 leads to attenuation of this amplitude [7-8]. In the current study, selected photographs were presented continuously in a trial, might explain why it was not clearly observed and why there was no significant difference in LPC in the current study. Therefore, this result suggests that it might be difficult to use P300 in EFRPs as a measure in criminal investigations.

The current study was conducted as a basic study to confirm the usefulness of EFRPs in CIT. However, there are still some points to consider when applying EFRPs to actual criminal investigations. In CIT measuring ANS, which is commonly used in criminal investigations, the examinees are presented crime-related or non-crime-related information for a few seconds and are asked questions about the item right after the presentation [21]. However, in our study, we modified the CIT procedure by continuously presenting a person's photograph (and a target) on alternating sides of the monitor to elicit eye fixations. If EFRPs are to be used in criminal investigations, it would be necessary to measure EFRPs elicited by spontaneous eye fixation to a photograph presented in a fixed location on the display. In addition, it is also possible that the P1 modulation might be influenced by a temporary change in arousal, rather than attention, due to the continuous presentation of the selected (or unselected) photograph. If the participants' arousal (i.e., their state) influences the P1 amplitude, it is reasonable to expect a similar effect on other EFRP responses, such as the LPC amplitude. However, since this was not observed, we believed that the observed effect is more likely related to attention rather than arousal. Nevertheless, further investigations with more precise experiments and analyses are needed to confirm this hypothesis.

In conclusion, the current study succeeded to provide evidence about the utility of EFRPs, especially the P1 amplitudes, in CIT. The findings of this study constitute a significant contribution, as they unveil a novel measure that is anticipated to be used in criminal investigations. In addition, this measure has the potential to shed more light on the psychological mechanisms of concealment that occur in situations close to everyday life. Further studies are needed to further clarify the usefulness of this approach.

Funding

Part of this study was supported by Collaborative Research Subsidy of 2023-2024 funds from Bunkyo Gakuin University awarded to KF.

Conflicts of interest/Competing interests

The authors declare no competing interests.

Data availability

The data set presented in this research is partially available upon request to the corresponding author, limited to the data that the participant has agreed to make publicly available.

Code availability

Not applicable.

Authors' contributions

KF and NS designed the experimental procedures. All authors prepared the experiment. KF and NS collected and analyzed the data. KF wrote the first draft of the manuscript. All authors revised the draft and approved the final version.

Verschuere B, Ben-Shakhar G, Meijer E. Memory detection: Theory and application of the Concealed Information Test. Cambridge University Press; 2011.
Matsuda I, Ogawa T, Tsuneoka M. Broadening the use of the concealed information test in the field. Front. Psychiatry. 2019; https://doi.org/10.3389/fpsyt.2019.00024.
Miyake Y, Mizutanti M, Yamahura T. Event related potentials as an indicator of detecting information in field polygraph examinations. Polygraph. 1993; 22: 131–149.
Hira S. A review of recent brain function studies of the concealed information test (written in Japanese). Jpn. J. Physiol. Psychol. Psychophysiol. 2009; https://doi.org/10.5674/jjppp.27.57.
Yagi A. Eye fixation related potential as an index of visual function. In: Kikuchi T, Sakuma H, Saito I, Tsuboi T, editors, Biobehaivioral Self-Regulation in the East and West, Springer-Verlag; 1995. pp.177–181.
Kubo K, Nittono H. The role of intention to conceal in the P300-based concealed information test. Appl. Psychophysiol. Biofeedback. 2009; https://doi.org/10.1007/s10484-009-9089-y.
Gonsalvez CJ, Polich J. P300 amplitude is determined by target-to-target interval. Psychophysiology. 2002; https://doi.org/10.1017/S0048577201393137.
Squires KC, Wickens C, Squires NK, Donchin E. The effect of stimulus sequence on the waveform of the cortical event-related potential. Science. 1976; https://www.science.org/doi/10.1126/science.959831.
Fischer T, Graupner ST, Velichkovsky BM, Pannasch S. Attentional dynamics during free picture viewing: Evidence from oculomotor behavior and electrocortical activity. Front. Syst. Neurosci. 2013; https://doi.org/10.3389/fnsys.2013.00017.
Fuseda K, Suzuki S, Iwakawa M, Katayama J. Time-course change in attentional resource allocation during a spot-the-difference task: investigation using an eye fixation-related brain potential. Curr. Psychol. 2023; https://link.springer.com/article/10.1007/s12144-021-01623-9.
Rämä P, Baccino T. Eye fixation–related potentials (EFRPs) during object identification. Vis. Neurosci. 2010; https://doi.org/10.1017/S0952523810000283.
Delorme A, Makeig S. EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. J. Neurosci. Methods. 2004; https://doi.org/10.1016/j.jneumeth.2003.10.009.
Lopez-Calderon J, Luck SJ. ERPLAB: an open-source toolbox for the analysis of event-related potentials. Front. Hum. Neurosci. 2014; https://doi.org/10.3389/fnhum.2014.00213.
Dimigen O, Sommer W, Hohlfeld A, Jacobs AM, Kliegl R. Coregistration of eye movements and EEG in natural reading: analyses and review. J. Exp. Psychol. Gen. 2011; https://doi.org/10.1037/a0023885.
Engbert R, Mergenthaler K. Microsaccades are triggered by low retinal image slip. Proc. Natl. Acad. Sci. U.S.A. 2006; https://doi.org/10.1073/pnas.0509557103.
Cohen J, Polich J. On the number of trials needed for P300. Int. J. Psychophysiol. 1997; https://doi.org/10.1016/S0167-8760(96)00743-X.
Yagi A. Averaged cortical potentials (lambda responses) time-locked to onset and offset of saccades. Physiol. Psychol. 1981; https://doi.org/10.3758/BF03326985.
Hillyard SA, Münte TF. Selective attention to color and location: An analysis with event-related brain potentials. Percept. Psychophys. 1984; https://doi.org/10.3758/BF03202679.
Hillyard SA, Vogel EK, Luck SJ. Sensory gain control (amplification) as a mechanism of selective attention: electrophysiological and neuroimaging evidence. Philos. Trans. R. Soc. Lond., B, Biol. Sci. 1998; https://doi.org/10.1098/rstb.1998.0281.
Ries AJ, Touryan J, Ahrens B, Connolly P. The impact of task demands on fixation-related brain potentials during guided search. PloS one. 2016; https://doi.org/10.1371/journal.pone.0157260.
Matsuda I, Nittono H. Physiological responses in the concealed information test: a selective review in the light of recognition and concealment. Detecting Concealed Information and Deception 2018; pp.77–96.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Applying eye-fixation-related brain potentials for a concealed information test

Status:

Version 1

Abstract

Figures

1. Introduction

2. Methods

3. Results

4. Discussion

Declarations

References

Additional Declarations

Status:

Version 1