RESEARCH ARTICLE ADVANCES IN COGNITIVE PSYCHOLOGY

Predicting Individuals’ Experi-

enced Fear From Multimodal
Physiological Responses
to a Fear-Inducing Stimulus
Eun-Hye Jang1, Sangwon Byun2, Mi-Sook Park3, and Jin-Hun Sohn4
1
Welfare & Medical ICT Research Department, Electronics and Telecommunications Research Institute (ETRI), 218 Gajeong-ro,
Yuseong-gu, Daejeon, 34129, Republic of Korea
2
Department of Electronics Engineering, Incheon National University, 119 Academy-ro, Yeonsu-gu, Incheon, Republic of Korea
3
Department of Rehabilitation Counselling, Seoul Hanyoung University, 290-42 Kyoungin-ro, Guro-gu, Seoul, 08274, Republic of Korea
4
Department of Psychology, Brain Research Institute, Chungnam National University, 99 Daehak-ro, Yuseong-gu, Daejeon, 34134,
Republic of Korea

ABSTRACT
Emotions are experienced differently by individuals, and thus, it is important to account for individ-
uals’ experienced emotions to understand their physiological responses to emotional stimuli. The
present study investigated the physiological responses to a fear-inducing stimulus and examined
whether these responses can predict experienced fear. A total of 230 participants were presented
with neutral and fear-inducing film clips, after which they self-rated their experienced emotions.
Physiological measures (skin conductance level and response: SCL, SCR, heart rate: HR, pulse transit
time: PTT, fingertip temperature: FT, and respiratory rate: RR) were recorded during the stimuli pres-
entation. We examined the correlations between the physiological measures and the participants’
experienced emotional intensity, and performed a multiple linear regression to predict fear inten-
sity based on the physiological responses. Of the participants, 92.5% experienced the fear emotion,
and the average intensity was 5.95 on a 7-point Likert scale. Compared to the neutral condition, the
SCL, SCR, HR, and RR increased significantly during the fear-inducing stimulus presentation whereas
FT and PTT decreased significantly. Fear intensity correlated positively with SCR and HR and nega-
KEYWORDS tively with SCL, FT, PTT, and RR. The multiple linear regression demonstrated that fear intensity was
predicted by a combination of SCL, SCR, HR, FT, and RR. Our findings indicate that the physiological
fear intensity
responses to experiencing fear are associated with cholinergic, sympathetic, and α-adrenergic vas-
experienced emotion
cular activation as well as myocardial β-sympathetic excitation, and support the use of multimodal
physiological signals
physiological signals for quantifying emotions.
autonomic responses

INTRODUCTION

Emotion recognition has been studied in affective computing and al., 2002). In addition, physiological signals can be measured us-
human-computer interaction (HCI) as a crucial machine capacity ing noninvasive and simple methods, which are important advan-
for effective communication (Choi et al., 2015; Park et al., 2013). tages in developing emotion recognition. Therefore, physiological
Previous studies have demonstrated that emotional stimuli can signals have been previously studied with the aim of recognizing
elicit spontaneous reactions from the autonomic nervous system human emotions based on a strong relationship between physi-
(ANS), which affects various physiological signals such as heart
rate (HR), respiratory rate (RR), blood pressure (BP), skin con- Corresponding author: Jin-Hun Sohn, Department of Psychology, Chungnam
ductance, and body temperature (Levenson, 2014). Physiological National University, 99 Daehak-ro, Yuseong-gu, Daejeon, 34134, South Korea.
responses to emotions are relatively similar among different soci- E-mail: jhsohn@cnu.ac.kr; Sangwon Byun, Department of Electronics Engineering,
eties and cultures since they are mostly involuntary and resistant Incheon National University, 119 Academy-ro, Yeonsu-gu, Incheon, Republic of
to deception (Drummond & Quah, 2001; Tsai et al., 2000; Tsai et Korea. E-mail:swbyun@inu.ac.kr

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

291 2020 • volume 16(4) • 291-301 http://www.ac-psych.org

DOI • 10.5709/acp-0303-x
RESEARCH ARTICLE ADVANCES IN COGNITIVE PSYCHOLOGY

ological reactions and emotional and affective states (Park et al., the specific physiological responses to fear since they used an ap-
2013). praisal model that conceptualizes emotion based on the dimen-
Detection of the fear emotion has also been studied based on sions of relevance, implication, coping potential, and normative
physiological signals, which can be applied to criminal investiga- significances.
tion, intelligent surveillance systems, and treatment of anxiety However, there are some studies examining the relationship
disorders (Choi, et al., 2015). Fear is a withdrawal-related emo- between fear intensity and physiological responses from the per-
tion with high arousal and negative valence that is elicited by an spective of discrete emotions (Aue et al., 2012; Choi et al., 2015;
event appraised as threatening. This leads to an activation of the Vidmar et al., 2016; Yoshihara et al., 2016). Aue et al. (2012) used
ANS and a possible fight-or-flight response. Emotion encom- multiple pictures displaying spiders and snakes as fear-inducing
passes both the process of emotion elicitation and the emotional stimuli and asked participants to rate their subjective fear for in-
response (e.g., Scherer & Moors, 2019). As individuals respond dividual images, which demonstrated a significant effect of fear
with varying intensity to the same emotional stimulus, it is im- intensity (high vs. low) on physiological measures. However, this
portant to consider their experienced emotions to understand study did not investigate differences in response to the same fear-
their corresponding physiological changes. Evaluating emotion at inducing stimulus, for example, a single picture of a spider, since
the individual level is important in developing user-specific HCI multiple images of different types of animals were presented.
technologies, which benefit substantially from tracking users’ To understand physiological changes in response to emotional
emotional states. stimuli, it is important to consider individual variability in expe-
Fear is also defined by the conscious emotional experience rienced fear intensity and physiological responses. In addition,
that occurs when an organism is threatened (LeDoux, 2014), emotional responses to static stimuli, such as pictures, are less ac-
which is closely bound to changes in bodily sensations (Pace- tive than the responses to film clips, which present dynamic visual
Schott et al., 2019). The direction of causality and the specificity and auditory stimuli to participants (Gross & Levenson, 1995).
of relationships between physiological responses and emotional Yoshihara et al. (2016) used a film clip to induce the fear
experience have been debated ever since James (1890) proposed response and showed a significant negative correlation between
that subjective emotion is determined by somatic responses to the subjective ratings of fear and rate of change in fingertip tem-
stimuli. Since emotion is a multicomponent process, it is crucial perature (FT). However, this study was based on a single sympa-
to investigate it at different levels, including physiology, which thetic autonomic measure. Since the physiological responses to
complements self-reported emotional experience and behavior emotion elicitation comprise an integrated pattern, it is impor-
(Pace-Schott et al., 2019). For example, Grandjean et al. (2008) tant to include a sufficient number of physiological measures to
proposed the hypothesis that conscious emotional experience identify any variations (Schneiderman & McCabe, 1989). Choi
emerges as a function of multilevel, appraisal-driven response et al. (2015) used a fear-inducing film clip and measured facial
synchronization. Meanwhile, studies on individual differences temperature, blinking rate, and brain activity (via electroen-
in physiological responses to emotional stimuli have demon- cephalography, EEG), which reflects the reactions of the central
strated that some people are highly responsive to self-produced nervous system (CNS). They proposed that facial temperature is
cues from their own behavior (e.g., facial expressions, actions), more reliable than other physiological response for the evaluation
whereas others are more responsive to situational cues from their of fear. Vidmar et al. (2016) measured the intensity of experienced
understanding of the context (e.g., norms about situations, social emotion in response to two fear-inducing video clips and related
pressures, Laird & Berglas, 1975; Laird & Lacasse, 2014). it to changes in HR, BP, and SC; however, they did not find a
Various approaches in affective sciences have been used to statistically significant relationship. This might have been due
investigate the relationship between physiological responses to familiarity with the popularly known stimuli or a self-report
and emotional experience. For example, dimensional models of bias from participants who did not want to admit to experiencing
emotion based on arousal and valence have been used widely to intense fear in an attempt to appear strong or stoic. These issues
describe the experienced emotion and relate emotional intensity might have affected the results substantially since their study was
to physiological responses (Bradley et al., 2001; Bradley & Lang, based on only 20 participants.
2000; Lang et al., 1993; Jenke & Peer, 2018). These studies have In particular, the relationship between individual variability
demonstrated that arousal and valence ratings are related sig- in experienced fear intensity and encompassing multimodal ANS
nificantly to physiological responses to emotional stimuli. Lang responses has not been studied in detail from the perspective of
et al. (1993) also examined the relationship between physiological discrete emotions. To address this issue, in the present study, we
responses and emotional intensity. However, they measured in- selected a short excerpt from a horror movie as the stimulus to
tensity as the level of arousal and valence instead of the level of the avoid using static images or familiar video clips and to reliably
experienced discrete emotion, that is, fear specifically. Similarly, elicit strong fear responses. In addition, the same stimulus was
Jenke and Peer (2018) examined the link between emotional used for all participants to evaluate differences at the individual
intensity and physiological signals, but it was difficult to define level. Instead of relying on one or two autonomic responses,

292 2020 • volume 16(4) • 291-301 http://www.ac-psych.org

RESEARCH ARTICLE ADVANCES IN COGNITIVE PSYCHOLOGY

multiple physiological signals were measured to interpret ANS Emotion-Provoking Stimuli

responses more effectively. Since the abovementioned previous
Audiovisual film clips were used to elicit the participants’ emo-
studies were based on only 16−34 participants, we used a relative-
tions, as they are more effective in presenting information related
ly large sample size for this study. We recruited 230 participants
to developing integrated and sustained emotional responses than
and asked them to identify their emotion and rate its intensity
still pictures. The film clip used as the fear-inducing stimulus in
in response to the fear-inducing stimulus, which accounted for
the present study was 60 s and excerpted from a horror movie
the independent variable of how the fear-inducing stimulus was
entitled A Tale of Two Sisters (Kim et al., 2003). We selected and
individually experienced.
excerpted the clip to elicit a strong fear response and to provide
Our study had two objectives: First, we compared the multi-
additional recovery time from the previous physiological baseline
modal physiological responses, including skin conductance level
measurement session. Therefore, the fear-inducing stimulus was
(SCL), skin conductance response (SCR), HR, FT, RR, and pulse
designed to intensify the emotion toward the end of its duration.
transit time (PTT), induced by the fear-inducing and neutral film
The clip showed a ghost suddenly appearing and approaching
clips. These physiological variables were selected to represent
a girl lying on her bed, which happened during the second half
emotion-related autonomic activation. We hypothesized that
of the clip. The first half did not contain a scene that showed
indices of peripheral physiological arousal will be higher while
the ghost explicitly. Therefore, although the entire length of the
participants watch the fear-inducing film rather than the neutral
clip was 60 s, only the second half was expected to elicit strong
film—as demonstrated by previous studies. We also examined
emotion-related physiological responses.
the relationship between the participants’ experienced intensity
We conducted a preliminary experiment to evaluate the valid-
of fear in reaction to the fear-inducing stimulus and their physi-
ity of the fear-inducing stimulus. A total of 50 college students
ological responses, and investigated whether the responses could
(25 males, Mage 22.58 ± 1.24), who did not participate in the study
predict fear intensity. As fear induces widespread sympathetic ac-
proper, were presented with the fear-inducing film clip via a 38
tivation and is an emotion with high arousal and negative valence,
in. screen in a classroom and asked to report the emotion they
we hypothesized that the participants’ experienced fear intensity
experienced during exposure to the stimulus. The participants
will statistically significantly correlate with their physiological
had to choose one emotion out of 11 (i.e., happiness, sadness,
responses related to sympathetic activation, and will be predicted
anger, contempt, disgust, fear, surprise, boredom, interest, neu-
by a linear combination of the physiological responses.
tral, and other). Then, they were required to rate the intensity of
the emotion they experienced on a 7-point Likert scale ranging
METHODS from 1 (least intense) to 7 (most intense). In this experiment, all 50
participants chose the emotion of fear with an average intensity of
5.5 ± 0.83. A 60 s neutral film clip was chosen to evoke no emo-
Participants
tion; it was excerpted from the same movie to match the contents,
We conducted an a priori power analysis to compute the neces- color, and hue with the fear-inducing film clip. The neutral film
sary sample size using G*Power, a power analysis program for clip depicted a girl walking around a lake, house, and room. The
statistical tests (Faul et al., 2009). The result indicated that 164 average intensity of neutral stimulus was 5.1 ± 1.36.
participants would be required to obtain adequate power (0.95)
for detecting a medium effect size of 0.25 at the standard .05 α Procedure
error. Since the minimum sample size was relatively large, we Prior to the experiment, the procedure was explained to the par-
recruited as many participants as possible before the end of the ticipants while they were sitting on a chair in the laboratory. To
academic semester during which the study took place, in order measure physiological signals, the electrodes for the sensors were
to account for drop-outs (Simmons et al., 2011). A total of 230 attached to the participants. Physiological responses were meas-
undergraduate students (110 males; Mage ± SD, 22.28 ± 2.05) took ured during three sessions: at baseline and in the neutral and fear-
part in the study. None of the participants reported any history of inducing film sessions (see Figure 1). The duration of each ses-
medical illness or psychotropic medication use. The experimen- sion was 60 s. During the baseline session, the participants were
tal procedure was explained to the participants, who provided instructed to rest. Then, the neutral and fear-inducing film clips
written consent to participate in the study before the experiment were presented in a random order to the participants. After both
began. In addition, $20 compensation was paid to each partici- clips were presented, the participants rested for 30 s and evaluated
pant. This study was approved by the Institutional Review Board the emotions they experienced in response to the stimuli. First,
of Chungnam National University, Daejeon, South Korea (No. the participants labeled the emotion they experienced (based on
201309-SB-004-01). the above 11 categories). Next, they rated emotion intensity on
the 7-point Likert scale. Finally, the participants identified and
described the scene in which they experienced the strongest emo-
tion during the fear-inducing film session.

293 2020 • volume 16(4) • 291-301 http://www.ac-psych.org

RESEARCH ARTICLE ADVANCES IN COGNITIVE PSYCHOLOGY

FIGURE 1.
Example of the recorded physiological data and analysis sections. The last 30 s interval of each session (dotted box) was selected for
data analysis.

Physiological Data Acquisition µS/s, maximal half recovery time of 10 s) were averaged across
and Feature Extraction the 30 s period for each session (Stemmler, 1992; Stemmler et al.,
2001). The SCR amplitude was calculated as the change from the
The MP100WS device and AcqKnowledge software (version
onset of the response to the peak of the response (Alexander et al.,
3.9.1) from BIOPAC Systems Inc. (Goleta, CA, USA) were used
2005). The mean FT was calculated by averaging FT values over
to measure and analyze the participants’ physiological signals.
30 s. As a PPG feature, PTT was used, which was defined as the
Electrocardiograms (ECGs) were recorded using a lead I configu-
time between the ECG R peak and the systolic peak in the PPG
ration, in which the electrodes were attached to the right wrist,
pulse signal. RR was defined as the number of breaths per minute,
left wrist, and left ankle. The ECG signal was low-pass filtered
which was counted from RSP signals.
(0–125 Hz) and sampled at 250 Hz. The EDA signals were re-
corded by applying a constant voltage of 0.5 V between the volar Data Analysis
surfaces of the proximal phalanx of the forefinger and middle
The self-ratings indicated that all participants in the current
finger of the nondominant hand. The EDA signal was low-pass
study experienced the most intense fear during the last 30 s of
filtered (0–10 Hz) and sampled at 100 Hz. The FT was measured
the fear-inducing clip. For each session, the last 30 s interval was
at the volar surface of the distal phalanx of the little finger of the
selected from the total length, and physiological features during
nondominant hand. The FT signal was low-pass filtered (0–0.15
this interval were evaluated. Figure 1 illustrates examples of the
Hz) and sampled at 1 Hz. The photoplethysmograph (PPG) signal
physiological signals collected and the 30 s intervals selected for
was measured from the volar surface of the distal phalanx of the
data analysis.
thumb of the nondominant hand. The PPG sensor was strapped
All data were analyzed using the Statistical Package for Social
around the thumb, which prevented interference by external light
Sciences (SPSS), version 21.0 (SPSS Inc, Chicago IL). To ana-
or a decrease in finger skin temperature that causes vasoconstric-
lyze the differences between the physiological responses during
tion. The PPG signal was low-pass filtered (0–10 Hz) and sampled
the neutral and fear-inducing sessions, paired-sample t-tests
at 250 Hz. The respiration (RSP) was measured using a respira-
were performed on the values after subtracting the baseline.
tion belt transducer, which was band-pass filtered (0.05–1 Hz)
Nonparametric Spearman’s rank correlation coefficients were
and sampled at 50 Hz.
calculated to investigate the relationships between psychological
In total, six features were extracted from the physiological
and physiological responses (i.e., fear minus baseline) as the psy-
signals. As an ECG feature, mean HR was used to represent heart
chological responses were not distributed normally. The strength
activity, which was calculated from R peak-to-R peak intervals.
of the correlation coefficients was evaluated using the following
As EDA features, SCL and SCR were used to reflect sweat secre-
criteria (Ratner, 2009): the values of 1 and -1 indicate a perfect
tion. SCL was extracted from the raw EDA signal using a 0.15
linear relationship, values between .7 and 1 (also, -.7 and -1)
Hz low-pass filter and sampled at 1 Hz. The mean SCL ampli-
indicate a strong relationship, values between .3 and .7 (-.3 and
tude was evaluated over the 30-s period for each session. The
-.7) indicate a moderate relationship, and values between 0 and
SCRs were considered valid if they showed an onset in the 1−4
.3 (0 and -.3) indicate a weak relationship. A level of significance
s poststimulus period and showed a peak during the 5 s subse-
of .05 was chosen for all tests, and 95% CI were calculated for
quent to the onset (Boucsein et al., 2012). Then, all valid SCRs
each physiological response. To correct for comparing multi-
greater than the threshold of 0.078 µS (minimal slope of 0.007

294 2020 • volume 16(4) • 291-301 http://www.ac-psych.org

RESEARCH ARTICLE ADVANCES IN COGNITIVE PSYCHOLOGY

TABLE 1.
Differences between physiological responses during the neutral and fear-inducing sessions (N = 230, df = 228)
Neutral Baseline Fear Baseline Mean difference (std.
Parameters 95% CI t p Cohen's d
M SD M SD error difference)

SCL (µS) .01 .07 .19 .20 .18(.02) .14−.21 10.58 < .001 .81
SCR (µS) .10 .45 2.06 1.24 1.95(.10) 1.75−2.15 19.24 < .001 1.56
HR (bpm) .40 3.38 7.27 11.24 6.86(.91) 5.08−8.65 7.58 < .001 .60
FT (° C) .07 .24 −.07 .22 −.14(.02) −.18−−.09 -5.63 < .001 .38
PTT (ms) -.65 6.64 −13.63 17.66 −12.98(1.45) −15.83−−10.13 −8.96 < .001 .63
RR (breaths/min) -.37 1.31 .72 1.44 .40(.09) .22−.59 4.30 < .01 .51

Note. SCL = skin conductance level; SCR = skin conductance response; HR = heart rate; FT = fingertip temperature;
PTT = pulse transit time; RR = respiratory rate.

ple physiological features, we used the Bonferroni correction the neutral film clip showed 94.8% appropriateness and 5.23 ±
method and adjusted α levels by dividing .05 by six, the number 1.22 effectiveness.
of features in the present study. Multiple linear regression with
stepwise forward entry was conducted to determine whether a Physiological Responses to the
linear combination of the physiological responses could predict Fear-Inducing Stimulus
the experienced fear intensity. The paired t-tests indicated that physiological responses changed
significantly after the presentation of the emotional stimulus in
the fear-inducing session, relative to the neutral session (see Table
RESULTS
1). All six physiological features showed significantly different
responses between the neutral and fear-inducing sessions. Skin
Psychological Responses to the conductance level (t = 10.58, p < .001), SCR (t = 19.24, p < .001),
Fear-Inducing Stimulus HR (t = 7.58, p < .001), and RR (t = 4.30, p = .001) increased in the
fear-inducing film session to a greater degree than in the neutral
The appropriateness of the two film clips was evaluated by cal-
film session. In contrast, FT (t = −5.63, p < .001) and PTT (t =
culating the percentage of the participants who experienced the
−8.96, p < .001) decreased in the fear-inducing film session to a
intended emotion. The fear-inducing film clip showed 92.5%
greater degree than in the neutral film session. The physiological
appropriateness. The effectiveness of the film clips was defined
feature values before subtracting the baseline are shown in Figure
as the average intensity ± SD measured using the 7-point Likert
3, which indicates that the physiological responses to the fear-
scale. The fear film clip showed effectiveness on a level of 5.95
inducing clip were greater than those to the neutral film clip, and
± 1.36. Figure 2 shows the distribution of the participants’ self-
that the physiological features measured from the neutral session
ratings of the intensity of the experienced fear, which was mostly
were similar to those from the baseline.
between 5 and 7. These results suggest that the experienced emo-
tions indicated by the participants’ ratings were consistent with
the intended emotion for the fear-inducing stimulus. Similarly,
TABLE 2.
Correlation Between The Intensity of Experienced Fear and
Physiological Responses (Fear Minus Baseline) to the Fear
Stimulus (Bonferroni Correction for Multiple Comparisons, N
= 230, df = 228)

Intensity SCL SCR HR FT PTT RR

Intensity 1.00 −.14* .18* .27* −.33 −.43* −.26*

SCL −.14* 1.00 .39* −.10 .08 .11 .11
SCR .18* .39* 1.00 −.01 −.11 −.00 −.12
HR .27* −.10 −.01 1.00 −.05 −.26* −.12
FT −.14* .08 −.11 −.05 1.00 −.08 −.026
PTT −.43* .11 −.00 −.26* −.08 1.00 .13
RR −.25* .11 −.12 −.12 −.02 .13 1.00

Note. SCL = skin conductance level; SCR = skin conductance re-

FIGURE 2. sponse; HR = heart rate; FT = fingertip temperature; PTT = pulse
Distribution of the self-reported intensity ratings (x-axis: inten- transit time; RR = respiratory rate.
sity, y-axis: frequency). * p < .0083

295 2020 • volume 16(4) • 291-301 http://www.ac-psych.org

RESEARCH ARTICLE ADVANCES IN COGNITIVE PSYCHOLOGY

FIGURE 3.
Physiological responses measured during the baseline, neutral, and fear-inducing sessions.

FIGURE 4.
Correlation between fear intensity and changes in physiological responses (fear minus baseline).

296 2020 • volume 16(4) • 291-301 http://www.ac-psych.org

RESEARCH ARTICLE ADVANCES IN COGNITIVE PSYCHOLOGY

TABLE 3.
Multiple Linear Regression Analysis Predicting Experienced Fear Intensity Based on Physiological Responses, N = 230, df = 223
Unstandardized Standardized
Predicted Collinearity statistics
Predictors coefficients coefficients t p 95% CI of B
variable
B SE β Tolerance VIF
SCL −.264 .073 −.211 −3.61*** < .001 −.39−−.14 .770 1.299
Intensity of SCR .165 .048 .184 3.47** < .001 .09−.24 .940 1.064
experienced HR .053 .006 .467 8.97*** < .001 .04−.06 .969 1.032
fear FT −.515 .097 -.278 −5.29*** < .001 −.68−−.35 .954 1.049
RR −.074 .023 -.191 -3.24** < .01 −.11−−.04 .754 1.325
R = .694, R2 = .482, Adjusted R2= .469, F(5, 224) = 12.039, p < .001, Durbin-Watson's d = 1.606

Note. SCL = skin conductance level; SCR = skin conductance response; HR = heart rate; FT = fingertip temperature;
PTT = pulse transit time; RR = respiratory rate.
* p < .05, ** p < .01, *** p < .001

Relations Between the Intensity DISCUSSION

of Experienced Fear and
Physiological Responses
Psychological Responses to the
Table 2 and Figure 4 show the relationships between the partici- Fear-Inducing Stimulus
pants’ intensity of experienced fear and physiological responses
In the present study, we used a film clip as an emotional stimulus
(i.e., fear minus baseline) to the fear-inducing stimulus. Fear in-
to induce intense fear. The fear-inducing film clip showed an ap-
tensity was positively correlated with SCR (rs = .18, p = .005) and
propriateness of 92.5% and scored 5.95 (out of 7) for effectiveness,
HR (rs = .27, p < .001), and negatively correlated with SCL (rs =
suggesting that it was effective in provoking fear. These results are
−.14, p = .008), FT (rs = −.33, p < .001), PTT (rs = −.43, p < .001),
consistent with previous studies, which have suggested that film
and RR (rs = −.26, p < .001). The strength of the correlation be-
clips are more effective than still pictures in delivering informa-
tween fear intensity and PTT was moderate (r = −.43) and for the
tion for inducing emotional responses.
correlations between the intensity and SCL (r = −.14), SCR (r =
−.18), HR (r = −.27), FT (r = −.23), and RR (r = −.26), it was weak. Physiological Responses to the
In addition, there were significant correlations between physi- Fear-Inducing Stimulus
ological responses, that is, a positive correlation between SCL and
Our results demonstrated that SCL, SCR, HR, and RR increased
SCR (rs = .39, p < .001) and a negative correlation between HR
in response to the fear-inducing stimulus, which is consistent
and PTT (rs = −.26, p < .001; see Table 2).
with previous findings. Unlike other physiological features, SCL
We performed multiple linear regression analysis with step-
and SCR are only affected by the sympathetic response, spe-
wise forward entry to examine whether a combination of physi-
cifically cholinergic sympathetic changes, and are not affected by
ological responses can predict the fear intensity (see Table 3). A
parasympathetic responses. Therefore, increases in SCL and SCR
significant regression equation was found (p < .001), with R =
during the fear-inducing session reflected an increased activation
.694. Except for PTT, SCL (p < .001), SCR (p = .001), HR (p <
of the sweat glands and sympathetic arousal.
.001), FT (p < .001), and RR (p = .001) were significant predic-
Cardiac activity is affected both by the sympathetic and para-
tors, which explained 48.2% of the variance in fear intensity. Fear
sympathetic nervous systems. An increased HR can be caused
intensity was influenced by the physiological responses in the fol-
by either sympathetic activation or parasympathetic inhibition.
lowing order: HR (β = .467), FT (β = -.278), SCL (β = −.211), RR
Therefore, it is not certain which mechanism underlies the in-
(β = −.191), and SCR (β = .184). Skin conductance response and
creased HR unless other physiological signals are accounted for.
HR were positively related to the intensity, whereas SCL, FT, and
In this study, both HR and RR increased during the fear-inducing
RR were negatively related. The Durbin-Watson test showed that
session, which may indicate activation of the sympathetic nerv-
there was no first order auto-correlation (d = 1.606). Tolerance
ous system. Similarly, in previous studies, increased HR was
(.7–.97) and variance inflation factors (VIF, 1.03–1.33) indicated
observed with a shorter period of respiration during the fear
that there was no multicollinearity. The analysis yielded the fol-
state . However, conflicting results on fear-induced HR changes
lowing linear model:
have also been reported, for which a decreased HR was observed
Intensityfear = 5.212 − .264(SCL) + .165(SCR) + .053(HR) −
during the fear-inducing session. Furthermore, changes in HR
.515(FT) − .074(RR).
induced by fear-inducing stimuli may be dependent on the ap-
praisal processes or stimuli types (e.g., slide, cognitive task,

297 2020 • volume 16(4) • 291-301 http://www.ac-psych.org

RESEARCH ARTICLE ADVANCES IN COGNITIVE PSYCHOLOGY

and imagery). For example, if a person perceives a stimulus as is associated with sympathetic activation, but threat imminence
threatening and imminent, their cardiac activity increases as a is characterized by immobilization and sympathetic inhibition
reaction to fear. In contrast, HR decreases if certain types of stim- (Kreibig, 2010).
uli, such as attentional task demands, are used for fear-inducing Since the film clip used in this study included the scene where
stimuli. Thus, it is important to account for appraisal processes a ghost suddenly appears, we cannot rule out the possibility that
and stimuli types when interpreting HR changes. In the present the participants who reported intense fear also experienced the
study, the participants’ self-reports of the emotion induced by the emotion of surprise. Kragel and LaBar (2013) showed decreased
fear-inducing film suggest that they experienced the stimulus as RR in response to surprise induced by music and film clips. Feleky
threatening and fearful, as if they were in the protagonist’s place. (1916) used autobiographical recall to elicit wonder (surprise)
When considering the process of appraisal of the fear-inducing and found decreased RR. She suggested that a “decided inspira-
stimulus based on these results, the increased HR found in the tory pause” was a distinct feature of the characteristic breathing
present study may indicate activation of the sympathetic nervous curve of wonder and was observed similarly, but to a lesser
system’s fight-or-flight reaction. The RR defines how often the degree, in the fear breathing curve. Nonetheless, other previous
respiration cycle repeats itself each minute. Fear is usually asso- studies investigating surprise have shown increased RR (Kreibig,
ciated with increased arousal and induces a respiratory pattern 2010; Siedlecka & Denson, 2019), and no clear conclusion can
characterized by faster and deeper breathing. Similarly, in the be derived on respiratory response to the emotion of surprise.
present study, the fear-inducing stimulus provoked an increased Therefore, a more in-depth study is required to understand and
RR. As previous studies demonstrated, we observed decreases explain decreased RR with increasing fear intensity. Moreover,
in FT and PTT during the fear-inducing session. Pulse transit our results indicate that FT decreased with increased experienced
time is an indicator of arterial BP, and a decrease in PTT is sig- fear intensity, which was previously reported by Yoshihara et al.
nificantly correlated with increased BP. A simultaneous decrease (2016). As mentioned, PTT decreased compared to the baseline
in FT and PTT has been reported widely in previous research on while watching the fear-inducing stimulus clip and was nega-
fear, which suggests α-adrenergic vascular changes or increased tively correlated with experienced fear intensity. Based on these
vasoconstriction. results, we can conclude that as fear intensity increases, there are
increased α-adrenergic vascular changes and increased vasocon-
Relations Between the Fear striction.
Intensity and Physiological
Responses Limitations
The psychological responses during the fear-inducing session Our results demonstrated a significant relationship between
were positively correlated with SCR and HR, while they were fear intensity and physiological responses to the fear-inducing
negatively correlated with SCL, FT, PTT, and RR. In addition, we stimulus. However, the present study had some limitations. First,
demonstrated that the intensity of experienced fear can be pre- since fear intensity ratings were not distributed normally, we used
dicted using a multiple linear regression of features derived from Spearman’s rank correlation as a nonparametric statistical meth-
physiological measures (SCL, SCR, HR, FT, and RR). Fear is an od. As shown in Figure 2, the fear intensity ratings mostly scored
emotion with high arousal and negative valence. Previous stud- above 5 on a 7-point Likert scale. Second, although we have dem-
ies have shown that SCR increases when arousal increases, and onstrated the validity of the fear-inducing stimulus in the prelimi-
HR increases when an emotion becomes more unpleasant (more nary study (see Methods section) and the study proper, we did
negative valence). These findings are consistent with our results not consider the other emotions experienced by the participants
showing positive correlations between experienced fear intensity and to what extent the clip had achieved specificity for the target
and SCR and HR. However, the negative correlation between emotion. This is because all 50 participants in the preliminary
experienced fear intensity and SCL and RR was not consistent study reported fear as the emotion they experienced during ex-
with previous findings. Bradley and Lang (2000) and Lang et al. posure to the fear-inducing stimulus, and based on these results,
(1993)demonstrated that SCL and RR increased linearly with we decided to evaluate only the fear response. Third, we evaluated
the arousal intensity of emotional stimuli. However, a negative fear intensity, but did not account for the effect of the participants’
relationship between fear intensity and SCL has been reported motivational states. Previous studies have suggested that changes
from previous studies using a real-life induction context (radio in physiological responses depend not only on the emotional
play, announcement of uncontrollable event, and sudden light stimuli (e.g., threatening, violent death, etc.) but also on cognitive
outage) and music excerpts (Stemmler, 1989; Krumhansl, 1997) factors such as the participants’ fear reactions based on motiva-
as fear-inducing stimuli. Kreibig (2010) suggested that fear para- tional states, that is, defensive and appetitive (Bradley et al., 2001;
digm elicits a stronger degree of self-involvement, which leads to Lang et al., 1997). In addition, some participants may experi-
higher perceptions of threat imminence (Bradley & Lang, 2000; ence phobic fear depending on the context of the stimuli. Since
Craske, 1999; Fanselow, 1994; Lang et al., 1997). In general, fear we used a single fear-inducing stimulus, we did not investigate

298 2020 • volume 16(4) • 291-301 http://www.ac-psych.org

RESEARCH ARTICLE ADVANCES IN COGNITIVE PSYCHOLOGY

how the participants would respond to different contexts of the cated a negative correlation between subjective fear ratings and
stimuli, which limits comparisons to other studies on physiologi- the rate of change in FT. In addition, to the best of our knowledge,
cal responses to fear-inducing stimuli. Fourth, we were unable to there has been no study on the correlations between experienced
investigate the interaction between the sympathetic and parasym- fear intensity and PTT or RR. We also discovered individual dif-
pathetic activations in the physiological responses. In particular, ferences in psychological responses during the presentation of
our results did not differentiate between the influences or laten- the fear stimuli by studying a large sample of 230 individuals.
cies of the sympathetic and parasympathetic nervous systems The present findings suggest that physiological changes during
in response to the fear-inducing stimulus. Therefore, additional the experience of fear are associated with the cholinergic sym-
analyses of physiological responses such as heart rate variability pathetic and α-adrenergic vascular activations and myocardial
(HRV) are needed to differentiate the roles of the sympathetic and β-sympathetic excitation. Finally, we suggest that the psychologi-
parasympathetic nervous systems in regulating fear responses. cal responses of fear can be derived from a multiple linear regres-
For example, HRV spectral analysis combined with respiration sion model of peripheral physiological measures, which supports
frequency data can be used to evaluate sympathetic and parasym- the use of multimodal physiological signals for recognizing and
pathetic activities by calculating low- and high-frequency areas quantifying human emotions.
in the HRV power spectrum, respectively. This analysis has been
used to investigate correlations between emotional intensity in ACKNOWLEDGEMENTS
response to stimuli and HRV spectral components. However, we This work was supported by the Institute for Information
did not conduct HRV analysis in the present study, because the & Communications Technology Promotion (IITP) grant fund-
length of the data (30 s) was too brief to obtain reliable results, for ed by the Korea government (MSIT) (No. 2015-0-00062).
which at least 5 min duration is generally recommended. Lastly, This work was also supported by the National Research
despite the positive correlation between SCR and SCL, we found Foundation of Korea (NRF) grant funded by the Korea gov-
an opposite relationship with fear intensity ratings for these two ernment (MSIT) (No. 2020R1F1A1049236).
skin conductance features. SCR, which represents phasic re-
sponse, increased as fear intensity increased. Changes in phasic REFERENCES
activity are characterized by a steep incline and a slow recovery, Alexander D. M., Trengove C., Johnston P., Cooper T., August J.
and this pattern would similarly occur in SCRs during the fear- P., & Gordon E. (2005). Separating individual skin conduct-
inducing session. Further analyses of specific features, such as the ance responses in a short interstimulus-interval paradigm.
number of SCRs (nSCR) and temporal characteristics of the SCR Journal of Neuroscience Methods, 146, 116–123. doi: 10.1016/j.
waveform, including onset latency, rise time, and half-recovery jneumeth.2005.02.001
time, are needed to understand the increase in SCR. Although Aue, T., Hoeppli, M.-E., & Piguet, C. (2012). The sensitivity of
these temporal characteristics are not as commonly quantified as physiological measures to phobic and nonphobic fear intensity.
amplitude (Dawson et al., 2007), half-recovery time, for example, Journal of Psychophysiology, 26, 154–167. doi: 10.1027/0269-
is defined as the time it takes for an SCR to decrease to half its 8803/a000078
peak value and reflects the rate at which an emotional response, Boucsein, W., Fowles, D. C., Grimnes, S., Ben-Shakhar, G., Roth,
particularly fear or anxiety, dissipates (Newrith et al., 2006). W. T., Dawson, M. E., & Filion, D. L. (2012). Publication recom-
Therefore, slower recovery rates are related to slower dissipation mendations for electrodermal measurements. Psychophysiology,
of autonomic activation (Newrith et al., 2006). An increase in 49, 1017–1034. doi: 10.1111/j.1469-8986.2012.01384.x
SCR can be induced by high nSCR with short recovery time but Bradley, M. M., Codispoti, M., Cuthbert, B. N., & Lang, P. J.
also by low nSCR with long recovery time. In addition, instead (2001). Emotion and motivation I: Defensive and appetitive
of taking the average, the 30-s interval can be subdivided into a reactions in picture processing. Emotion, 1, 276–298. doi:
sequence of short moving windows (e.g., 5-s or 10-s) using the 10.1037/1528-3542.1.3.276
participants’ self-reports. Then, temporal changes in fear inten- Bradley, M. M., & Lang, P. J. (2000). Measuring emotion: Behavior,
sity during stimulus presentation can be tracked and related with feeling and physiology. In R. Lane & L. Nadel (Eds.), Cognitive
SCRs, which would provide more information on how experi- neuroscience of emotion (pp. 49–59). Oxford University Press.
enced fear is associated with the ANS response. Choi, J. S., Bang, J. W., Heo, H., & Park, K. R. (2015). Evaluation of
fear using nonintrusive measurement of multimodal sensors.
Implications Sensors, 15, 17507-17533. doi: 10.3390/s150717507
Although previous studies have examined the differences in Coan, J. A., & Allen, J. J. B. (2003). Frontal EEG asymmetry and the
physiological changes in response to basic emotions using mainly behavioral activation and inhibition systems. Psychophysiology,
HR, SCR, and BP during fear-inducing sessions, they did not find 40, 106–114. doi: 10.1111/1469-8986.00011
significant relationships between fear intensity and physiological Craske, M. G. (1999). Anxiety disorders: Psychological approaches
responses (Vidmar et al., 2016). Yoshihara et al. (2016) only indi- to theory and treatment. Boulder, CO.

299 2020 • volume 16(4) • 291-301 http://www.ac-psych.org

RESEARCH ARTICLE ADVANCES IN COGNITIVE PSYCHOLOGY

Dawson, M. E., Schell, A. M., & Filion, D. L. (2007). The elec- Sensory and motivational processes (pp. 97–136). Lawrence
trodermal system. In J. T. Cacioppo, L. G. Tassinary, & G. G. Erlbaum Associates.
Berntson (Eds.), Handbook of psychophysiology (pp. 159–181). Lang, P. J., Greenwald, M. K., Bradley, M. M., & Hamm, A. O.
Cambridge University Press. (1993). Looking at pictures: Affective, facial, visceral, and be-
Drummond, P. D., & Quah, S. H. (2001). The effect of express- havioral reactions. Psychophysiology, 30, 261–273. doi: 10.1111/
ing anger on cardiovascular reactivity and facial blood flow in j.1469-8986.1993.tb03352.x
Chinese and Caucasians. Psychophysiology, 38, 190–196. doi: LeDoux, J. E. (2014). Coming to terms with fear. PNAS, 111,
10.1111/1469-8986.3820190 2871–2878. doi: 10.1073/pnas.1400335111
Fanselow, M. S. (1994). Neural organization of the defensive be- Levenson, R. W. (2014). The autonomic nervous sys-
havior system responsible for fear. Psychonomic Bulletin and tem and emotion. Emotion Review, 6, 100–112. doi:
Review 1, 429–438. doi: 10.3758/BF03210947 10.1177/1754073913512003
Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009). Statistical Newrith, C., Meux, C., & Taylor, P. (2006). Personality disorder
power analyses using G*Power 3.1: Tests for correlation and re- and serious offending: Hospital treatment models. CRC Press.
gression analyses. Behavior Research Methods, 41, 1149–1160. UK.
doi: 10.3758/BRM.41.4.1149 Pace-Schott, E. F., Amole, M. C., Aue, T., Balconi, M., Bylsma,
Feleky, A. (1916). The influence of the emotions on respiration. L. M., Critchley, H., Demaree, H. A., . . . VanElzakker, M. B.
Journal of Experimental Psychology, 1, 218–241. doi: 10.1037/ (2019). Physiological feelings. Neuroscience and Biobehavioral
h0073754 Reviews, 103, 267–304. doi: 10.1016/j.neubiorev.2019.05.002
Grandjean, D., Sander, D., & Scherer, K. R. (2008). Conscious
emotional experience emerges as a function of multi-level, Park, B. J., Jang, E. H., Chung, M. A., & Kim, S. H. (2013).
appraisal-driven response synchronization. Consciousness and Design of prototype‐based emotion recognizer using physi-
Cognition, 17, 484–495. doi: 10.1016/j.concog.2008.03.019 ological signals. ETRI Journal, 35, 869–879. doi: 10.4218/
Gross, J. J., & Levenson, R. W. (1995). Emotion elicita- etrij.13.0112.0751
tion using films. Cognition and Emotion, 9, 87–108. doi: Ratner, B. (2009). The correlation coefficient: Its values range
10.1080/02699939508408966 between +1/−1, or do they? Journal of Targeting, Measurement
James, W. (1890). The principles of psychology. Dover. and Analysis for Marketing, 17, 139–142 doi: 10.1057/jt.2009.5
Jenke, R., & Peer, A. (2018). A cognitive architecture for modeling
emotion dynamics: Intensity estimation from physiological Scherer, K. R., & Moors, A. (2019). The emotion process:
signals. Cognitive Systems Research, 49, 128–141. doi: 10.1016/j. Event appraisal and component differentiation. Annual
cogsys.2018.01.004 Review of Psychology, 70, 719–745. doi: 10.1146/annurev-
Kim, J. W. (Director), Oh, J. W. (Producer), & Oh, K. M. psych-122216-011854
(Producer). (2003). A Tale of Two Sisters. South Korea. Schneiderman, N., & McCabe, P. M. (1989). Psychophysiologic
Kragel, P. A., & LaBar, K. S. (2014). Advancing emotion theory strategies in laboratory research. In N. Schneiderman, S. M.
with multivariate pattern classification. Emotion Review, 6, Weiss, & P. G. Kaufmann (Eds.) Handbook of research methods
160–174. doi: 10.1177/1754073913512519 in cardiovascular behavioral medicine (pp. 349–364). Plenum
Kreibig, S. D. (2010). Autonomic nervous system activity in Press.
emotion: A review. Biological Psychology, 84, 394–421. doi: Siedlecka, E., & Denson, T. F. (2019). Experimental methods for
10.1016/j.biopsycho.2010.03.010 inducing basic emotions: A qualitative review. Emotion Review,
Krumhansl, C. L. (1997). An exploratory study of musical emo- 11, 87–97. doi: 10.1177/1754073917749016
tions and psychophysiology. Canadian Journal of Experimental Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2011). False-positive
Psychology, 51, 336–353. doi: 10.1037/1196-1961.51.4.336 psychology: Undisclosed flexibility in data collection and
Laird, J. D., & Berglas, S. (1975). Individual differences in the analysis allows presenting anything as significant. Psychological
effects of engaging in counter-attitudinal behavior. Journal Science, 22, 1359–1366. doi: 10.1177/0956797611417632
of Personality, 43, 286–304. doi: 10.1111/j.1467-6494.1975. Stemmler, G. (1989). The autonomic differentiation of emo-
tb00707.x tions revisited: Convergent and discriminant validation.
Laird, J. D., & Lacasse, K. (2014). Bodily influences on emotional Psychophysiology, 26, 617–632. doi: 10.1111/j.1469-8986.1989.
feelings: Accumulating evidence and extensions of William tb03163.x
James’s theory of emotion. Emotion Review, 6, 27–34. doi: Stemmler, G. (1992). Differential psychophysiology: Persons in
10.1177/1754073913494899 situations. Springer.
Lang, P. J., Bradley, M. M., & Cuthbert, M. M. (1997). Motivated Stemmler, G., Heldmann, M., Pauls, C. A., & Scherer, T. (2001).
attention: Affect, activation and action. In P. J. Lang, R. F. Constraints for emotion specificity in fear and anger: The con-
Simons, & M. T. Balaban (Eds.), Attention and orienting: text counts. Psychophysiology, 38, 275–291. doi: 10.1111/1469-

300 2020 • volume 16(4) • 291-301 http://www.ac-psych.org

RESEARCH ARTICLE ADVANCES IN COGNITIVE PSYCHOLOGY

8986.3820275 Vidmar, K., Pittala, K., Rewais, H., McLaughlin, K., & Bond, Q.
Tsai, J. L., Chentsova-Dutton, Y., Freire-Bebeau, L., & Przymus, D. (2016). An analysis of physiological arousal in response to sad
E. (2002). Emotional expression and physiology in European and fearful video stimuli. Journal of Advanced Student Science,
Americans and Hmong Americans. Emotion, 2, 380–397. doi: 6, 1–24.
10.1037/1528-3542.2.4.380 Yoshihara, K., Tanabe, H. C., Kawamichi, H., Koike, T., Yamazaki,
Tsai, J. L., Levenson, R. W., & Carstensen, L. L. (2000). Autonomic, M., Sudo, N., & Sadato, N. (2016). Neural correlates of fear-
subjective, and expressive responses to emotional films in older induced sympathetic response associated with the periph-
and younger Chinese Americans and European Americans. eral temperature change rate. NeuroImage, 134, 522-531. doi:
Psychology and Aging, 15, 684–693. doi: 10.1037/0882- 10.1016/j.neuroimage.2016.04.040
7974.15.4.684
RECEIVED 26.02.2020 | ACCEPTED 18.08.2020

301 2020 • volume 16(4) • 291-301 http://www.ac-psych.org