Articles

The Effect of Trait Anxiety on Recognition of Threatening Emotional Facial Expressions: A Study among High School Students.

Roslinda Mustapha, Md. Azman Shahadan, Hazalizah Hamzah,
Article Date Published : 28 February 2019 | Page No.: 5312-5318 | Google Scholar

Downloads

Download data is not yet available.

Abstract

Previous studies indicated that sensitivity to facial expressions of threat is related to anxiety in children, adolescents and adults. A small amount of anxiety often improves students' performance, but a high level of anxiety can interfere the learning process. The feeling of being threatened by particular stimuli would cause them to perceive many daily situations as threatening and this will result in more frequent experiences of fear of what may happen, especially for the high anxiety students. This research will explore the threat perception that the secondary school students might have in relation to negative facial expression and examine the sensitivity towards anger expressions as threatening stimuli. 49 students (25 low anxiety and 24 high anxiety) age between 16 to 18 years old have been recruited to answer a set of anxiety questionnaires and they were also required to identify the facial expression to explore the threat perception by looking at images posing facial expression in 2 and 3 dimensions. These images have been transformed into 5 levels of anger using FaceGen Modeller 3.5. Results demonstrated that the high anxiety students can identify threat stimuli from faces more accurately and faster than the low anxiety students. It is suggested that angry faces may be perceived as particularly threatening amongst students and play a significant role in their emotional well being. It is hoped that this research will increase our understanding of the relationship between anxiety and threat perception and this unique visual stimulus can generate a wealth of other research in Malaysia.

1.0 Introduction

Face perception has long been an interesting field of study. A lot of previous research has found that anxiety significantly correlated with threat perception (Gutierrez-Garcia & Calvo, 2016; Schofield, Johnson, Inhoff & Coles, 2012). Beside faces, other ambiguous stimuli such as body gesture, stories, voices, words, animals or threat related situation were also used (LoBue & Perez-Edgar, 2015; Koizumi, Tanaka, Imai, Hiramatsu, Hiramoto, Sato and Gelder, 2011; Bar-Haim et al., 2007).

  1. Individual with anxiety trait are more sensitive to a variety of threatening stimuli when compared to non-anxious individuals, and they tend to interpret others' emotional state in a negative manner ( Richards et al, 2002) which include the mixed emotion such as fear and surprise and neutral faces that are ambiguous in their thinking ( Mohlman et al. ; Yoon and Zinbarg ) . While many studies examined the bias of anxious individuals in recognizing facial expression, few studies have examined the sensitivity in detecting threatening facial expressions. The ability and the sensitivity towards recognizing facial emotions is an important component in our nonverbal communication system. It is also a crucial social skill needed for every human being from infancy stage to adult stage to successfully adapt and manipulate the environment. Unable to correctly interpret the emotional facial expression will Adolescent stage is especially important as this stage of development has long been recognized as a period of heightened risk-taking, a stage that need oversight from adults. Individual with anxiety trait are more sensitive to a variety of threatening stimuli when compared to non-anxious individuals, and they tend to interpret others' emotional state in a negative manner ( Richards et al, 2002) which include the mixed emotion such as fear and surprise and neutral faces that are ambiguous in their thinking ( Mohlman et al. ; Yoon and Zinbarg ). While many studies examined the bias of anxious individuals in recognizing facial expression, few studies have examined the sensitivity in detecting threatening facial expressions. The ability and the sensitivity towards recognizing facial emotions is an important component in our nonverbal communication system. It is also a crucial social skill needed for every human being from infancy stage to adult stage to successfully adapt and manipulate the environment. Unable to correctly interpret the emotional facial expression will Adolescent stage is especially important as this stage of development has long been recognized as a period of heightened risk-taking, a stage that need oversight from adults.

What students had experienced daily at schools might be

perceived as threatening especially by the anxious students. They probably even interpreted words and neutral facial expression around them in a negative way or as a sign of threat or danger. According to Wieser & Moscovitch ( 2015), neutral faces may produce the feeling of anxiety for socially anxious individuals due to their ambiguous nature. There are many different kinds of anxiety that can be identified among schools' student as well as the causes of anxious feeling experienced by them. Anxiety can be caused by different factors such as biological, cognitive and psychological factors that are intertwined in a complicated ways such as examination, test, pressure of school work, peers, family problems or parents divorcing can all be difficult events for students and can be a sources of anxiety for them ( Nadeem, Ali, Maqbool, & Zaidi, 2012).

However, to our knowledge, what does not seem to have been investigated is the effect of threat perception on anxious students. Therefore, this study aims to gain a better picture and understanding of students' anxiety and how it's effect their perceptions of negative facial expression. The aims of the study can be stated specifically into four objectives:

  1. To investigate if there is a significant interaction between anxiety, in a way that students with high anxiety will readily recognize angry expression accurately than students with low anxiety. 1. To investigate if there is a significant interaction between anxiety, in a way that students with high anxiety will readily recognize angry expression accurately than students with low anxiety.

  2. To investigate whether there is a significant interaction between groups of anxiety and threat perception across five morphs levels of anger expressions. 2. To investigate whether there is a significant interaction between groups of anxiety and threat perception across five morphs levels of anger expressions.

  3. To investigate if there is a significant interaction between anxiety, in a way that students with high anxiety will detect angry expression faster than students with low anxiety.3. To investigate if there is a significant interaction between anxiety, in a way that students with high anxiety will detect angry expression faster than students with low anxiety.

  4. To investigate whether there is a significant interaction between groups of anxiety and response time across five morphs levels of anger expressions. 4. To investigate whether there is a significant interaction between groups of anxiety and response time across five morphs levels of anger expressions.

2.0 Methodology

The study has been conducted by using experimental method. The STAI-trait anxiety questionnaire was used for pre-screen to classify the high and low anxiety students. The data were collected from the face perception tasks to examine whether there are differences between the high and low anxiety students in perceiving the threat from face stimuli.

Participants were selected among students from three different secondary schools in Selangor area. The schools were selected because of the availability of form 4 and form 6 students that is consistent with the study's model background that targeted adolescent participants age 15 to 19 years old. 137 (53 male and 84 female) students age predominantly 16 and 18 years old were recruited in order to conform to the restrictions imposed by the Ministry of Education regarding involvement of exam year students (which inadvertently excluded Form Three and Five students). Only participants with STAI-Trai anxiety scores over 75 and under 25 percentile were selected. The final samples were composed by 25 high anxiety students and 24 low anxiety students. We were helped by school teachers and counselor in providing samples, preparing the room for the tests and gathering other information needed for this study.

All participants completed the demographic questionnaires and State Trait Anxiety Inventory (STAI-state & trait; Spielberger, 1983). The Instrument consists of 40-item self-report questionnaire with two subscales namely as state anxiety and trait anxiety. This questionnaire was used to measure participants’ social anxiety level. STAI is a self-report measures that can assess how respondents “feel right now, at this moment” (y1) and how respondents “generally feel” (y2). According to Spielberger ( 1989), the STAI can also identify the high and low anxiety and has appeared in over 3,000 studies and has been translated into 30 languages (ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1037/1040-3590.19.4.369", "ISSN" : "1040-3590", "PMID" : "18085930", "abstract" : "The State-Trait Inventory for Cognitive and Somatic Anxiety (STICSA; M. J. Ree, C. MacLeod, D. French, & V. Locke, 2000) was designed to assess cognitive and somatic symptoms of anxiety as they pertain to one's mood in the moment (state) and in general (trait). This study extended the previous psychometric findings to a clinical sample and validated the STICSA against a well-published measure of anxiety, the State-Trait Anxiety Inventory (STAI; C. D. Spielberger, 1983). Patients (N=567) at an anxiety disorders clinic were administered a battery of questionnaires. The results of confirmatory factor analyses (Bentler-Bonnett nonnormed fit index, comparative fit index, and Bollen fit index>.90; root-mean-square error of approximation<.05); convergent and discriminant validity analyses; and group comparisons supported the reliability and validity of the STICSA as a measure of state and trait cognitive and somatic anxiety. In addition, compared with the STAI (anxiety: rs/=.64), the STICSA was more strongly correlated with another measure of anxiety (rs>/=.67) and was less strongly correlated with a measure of depression (rsGros, Antony, Simms, & McCabe, 2007)", "manualFormatting" : " Gros, Antony, Simms, & McCabe, 2007)", "plainTextFormattedCitation" : "( Gros, Antony, Simms, & McCabe, 2007)", "previouslyFormattedCitation" : "(Gros, Antony, Simms, & McCabe, 2007)" }, "properties" : { "noteIndex" : 0 }, "schema" : "https://github.com/citation-style-language/schema/raw/master/csl-citation.json" }Gros, Antony, Simms, & McCabe, 2007).

2.1 Selection of angry facial expression

There are numbers of study have found the advantage of angry or threatening face over other facial expression such as happy or neutral faces ( LoBue, 2009; Harvey, Thrasher, LoBue, & Matthewa 2014) or sad and fearful faces ( Calvo, Avero, and Lundqvist, 2006). Those studies had covered all group of ages, from infancy, children, adolescent and adults. Impressively, even infants as young as 9 months can detect angry faces more quickly than happy, neutral and sad faces ( LoBue, 2009, LoBue, 2010). Many researchers have suggested that the rapid detection of threatening faces from angry expression is because of its threatening message produced by the angry face stimuli ( LoBue et al., 2013). Even more, the anxious individuals have interpreted the neutral facial cues as conveying anger emotion regardless the threat condition (speech condition), while the non-anxious individuals have interpreted the neutral faces in a negative manner only when they were in the threat condition ( Yoon & Zinbarg, 2008).

2.2 Generating facial stimuli

The material for the face task was actually duplicated from study by Mustapha, Shahadan, Abdul Wahab Khan, Stephan & Archer ( 2017). The face stimuli were 6 faces consist of 3 male faces and 3 females' Malay faces showing neutral expressions. Those photos of real life face showing neutral expressions of emotion were converted into virtual avatars using the “photo fit” application in the FaceGen. FaceGen is a 3D face-generating software (http://facegen.com). We used this FaceGenModeller software, version 3.1 because it can generate a large number of stimuli easily and the users are allowed to randomize, normalize, generate emotional facial expression systematically and adjust certain characteristic such as race, gender, age, eye gaze and head position ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1073/pnas.0805664105", "ISSN" : "1091-6490", "PMID" : "18685089", "abstract" : "People automatically evaluate faces on multiple trait dimensions, and these evaluations predict important social outcomes, ranging from electoral success to sentencing decisions. Based on behavioral studies and computer modeling, we develop a 2D model of face evaluation. First, using a principal components analysis of trait judgments of emotionally neutral faces, we identify two orthogonal dimensions, valence and dominance, that are sufficient to describe face evaluation and show that these dimensions can be approximated by judgments of trustworthiness and dominance. Second, using a data-driven statistical model for face representation, we build and validate models for representing face trustworthiness and face dominance. Third, using these models, we show that, whereas valence evaluation is more sensitive to features resembling expressions signaling whether the person should be avoided or approached, dominance evaluation is more sensitive to features signaling physical strength/weakness. Fourth, we show that important social judgments, such as threat, can be reproduced as a function of the two orthogonal dimensions of valence and dominance. The findings suggest that face evaluation involves an overgeneralization of adaptive mechanisms for inferring harmful intentions and the ability to cause harm and can account for rapid, yet not necessarily accurate, judgments from faces.", "author" : [ { "dropping-particle" : "", "family" : "Oosterhof", "given" : "Nikolaas N", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Todorov", "given" : "Alexander", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Proceedings of the National Academy of Sciences of the United States of America", "id" : "ITEM-1", "issue" : "32", "issued" : { "date-parts" : [ [ "2008", "8", "12" ] ] }, "page" : "11087-92", "title" : "The functional basis of face evaluation.", "type" : "article-journal", "volume" : "105" }, "uris" : [ "http://www.mendeley.com/documents/?uuid=83a64f16-e82c-4dee-ad5e-079ab429c036" ] } ], "mendeley" : { "formattedCitation" : "( Oosterhof & Todorov, 2008)", "plainTextFormattedCitation" : "( Oosterhof & Todorov, 2008)", "previouslyFormattedCitation" : "( Oosterhof & Todorov, 2008)" }, "properties" : { "noteIndex" : 0 }, "schema" : "https://github.com/citation-style-language/schema/raw/master/csl-citation.json" }( Oosterhof & Todorov, 2008). The models are of medium age (25+ years old), bare of hair on the external features. We then generated 4 levels of anger from each of the neutral faces using an automated procedure in FaceGen. The morphing degree was adjusted into anger expression and the intensity of emotion was added linearly started with a neutral face at 0% of anger to 25%, 50%, 75% and 100% of anger resulting in 4 morphed images from each neutral face. Thus, there were 6 sets of facial images that had been generated into 30 emotionally angry faces of male and female stimuli (6 faces x 5 level of anger emotion strength). The example of one set of the generated faces is presented in figure 1. The face stimuli were then presented randomly in the PsychoPy program for the experiment. We presented the facial images for 2000 ms. The facial images were 400 X 400 pixel size and the images were presented in the middle of the screen with a black background.

Figure 1Example of unfamiliar facial images as stimuli that have been generated into five levels of anger (0 percent, 25 percent, 50 percent, 75 percent and 100 percent). Adopted from Mustapha et al. (2017)

Figure 1.Distribution in groups of patients by the stroke type.

2.3 Procedures

The experiment was run using a psychoPy program and was set up on the laptop as it will be easier for us to conduct the experiment in a school setting. The face stimuli were presented randomly and the measure of the answer is based on the proportions of “yes” responses. The study was approved by the Ethics Committee of University of Nottingham Malaysia Campus and granted permission by the Ministry of Education, Selangor State Department of Education and the respective school principal.

The nature and purpose of the study were explained to participants briefly and how the experiment will be conducted. They were then asked to complete the set of questionnaires consisting of: section 1 = Demographic Questionnaire, section 2 = State-Trait Anxiety Inventory (STAI); and doing the experiment of face perception. Approximately 20 to 30 minutes is needed by the participants to complete the questionnaires. The questionnaires were made available in Malay Language because many of the students are not good in English language.

The facial stimuli task was administered upon completing the questionnaires. Participants were instructed about what they will see on the screen and what they have to do. Participants were presented with images of 30 faces showing facial expressions, one facial image at a time. They were presented with a fixation point (+) for 1000 ms at the center of the screen, followed by facial image for 2000 ms. After that, the participants were asked if the facial image shows angry emotion or not angry. The task was self-paced where the question remained on the screen until the participant’s response. Participants made their responses by pressing the “Z” key on the keyboard which was labeled “yes” if the answer is angry and “M” key which was labeled “No” if the answer is not angry. The order of the trials was randomized for each participant. Participants were instructed to press the “yes” key or the “no” key as quickly as possible while still being accurate. RTs were recorded automatically by the computer. The average time needed to complete the task is less than 20 minutes. All the instruction given to the participants were in Malay language.

3.0 Result

The Analysis was conducted using the Statistical Package for Social Sciences version 21 (SPSSv21). Initial analyses examining the relations between demographic variables and student' self-reported anxiety included independent t-test for categorical variables. For the face experiment, we used a standard visual search procedure in which the accuracy in identifying emotional expressions had been measured based on the proportion of 'yes' responses when the participant chooses the target emotions per level of morphs. The mean of reaction time had been calculated as a function of sensitivity toward threat stimuli. In order to examine whether there is differences between the mean scores of identifying angry expression from five levels of anger expressions, a one-way within-group ANOVA was computed ( Field, 2009). Finally, repeated measure analysis of variance (ANOVA) with group (high vs low anxiety) by emotion valence (five morph levels) have been conducted to test the predictions about the ability of students in recognizing angry facial expression as well as its response time. The dependent variables are the proportion of the correct answers and response time.

3.1 Preliminary analysis

The participants were divided into high and low anxiety groups. For the prescreen session, 137 (53 male and 84 female) students age between 16 to 18 years old with a mean age of 17.4 (SD=1.43) were recruited. Only participant s with STAI-trait anxiety scores over the 75 and under 25 percentile were selected. The final sample was composed by 24 participant (10 male and 14 female and 25 participants (7 male and 18 female). The two groups differ significantly on STAI-trait scores t( 47)=-21.2, P<0.01, STAI-state score t( 47)=-3.46, P<0.01. However, they did not differ significantly in age t( 47)=6.71,P>0.01) respectively (refer to table 1).

Table 1: Mean scores and standard deviation of STAI- Trait and STAI-State of anxiety.

STAI-trait 57.2 5.3 28.1 4.2
STAI-state 48.7 8.3 39.3 10.5
Age 16.4 1.04 18.4 1.01

Therefore, a one-way within-group ANOVA was computed The result showed that there was a statistically significant differences between the score of identifying angry faces across five morph levels of angry expression, Wilks' Lambda=088, F(4, 46)= 26207, p<001

Since we have found a significant result for overall differences, we then computed additional paired samples t-tests to make post hoc comparisons between each levels. A first paired samples t-test indicated that there was a significant difference in the scores for neutral (M=.89, SD=1.37) and 25% (M=1.45, SD=1.36) morph levels; t( 49)=-4.210, p < .001. A second paired samples t-test indicated that there was a significant difference in the scores for 25% and 50% (M=2.67, SD=1.35) morph levels; t( 49)=-7.212, p<.001. A third paired samples t-test indicated that there was no significant difference in the scores for 50% and 75% (M=4.78, SD=1.45) morph levels; t( 49)=-13.050, p>.001 and finally, a forth Paired Samples T-Test indicated that there was a significant difference in the scores for 75% and 100% (M=6.23, SD=1.39) morph levels; t( 49)=-8.764, p<.001. Therefore, this stimuli has showing a strong effect of all morphs level an applicable to predict the differences in perceived threat.

Ref er ences

3.2.1 Emotional Recognition Task

There was a statistically significant difference for STAI-trait anxiety score between high (M=17.52, SD=3.82) and low (M=15.21, SD=3.71) in terms of total angry scores, t( 47)= -2.15, p=.037 (see figure 2a). This means that in general, students with high anxiety can identify the anger expression of the facial image more accurately than the students with low anxiety. Further analysis was conducted in order to test whether there are also differences between the anxious groups in perceiving threat perception across five morph levels. Repeated measure ANOVA of accuracy of facial expression identification showed no significant main effect of group F (1, 47) =3.71, p>0.05. However, there is a significant effect of morph levels F (4,3.21) =158.4, p<0.001; and a significant interaction of anxiety group x morph levels, F (4, 3.21) =2.95, p<0.05. There were no other interactions of variable as a function of anger expression identification. According to several pairwise comparison, there are significant differences between all level of facial morphs with p<0.05. Specifically between neutral and 25% of morph levels (p<0.001), between 25% and 50% of morph levels (p<0.001), between 50% and 75% of morph levels (p<0.001) and between 75% and 100% of morph levels (p<0.001). The interaction effect suggested that the high anxiety students showed more accuracy in detection of angry face when presented with ambiguous stimuli, specifically towards 50% and 75% of morphs levels.

Figure 2EThe differences in accuracy to detect anger facial expression between students with high and low anxiety

Figure 1.Distribution in groups of patients by the stroke type.

Figure 3The differences in accuracy to detect anger facial expression between students with high and low anxiety across five morph levels

Figure 1.Distribution in groups of patients by the stroke type.

3.2.2 Respond Time

There was a statistically significant difference between high (M=19.5, SD=8.48) and low (M=39.31, SD=17.91) stai-trait anxiety score in terms of total time needed to identify anger expression, t( 47)= 4.96, p=.000. it was shown that students with low anxiety spent more time to recognize the anger expression from the images as compared to the high anxiety students (figure 3a).

Further analysis was conducted to examine whether there is also differences in response time between the two anxious groups across five levels. Repeated measure analysis of ANOVA have showed no significant main effect of group F (1,47) =.003, p>0.001. and no significant interaction of anxiety group x morph levels, F (4,2.71) =.481, p>0.05. However, there is a significant main effect of morph levels F (4,2.71) =4.005, p<0.005 (see figure 3b). The main effect of morph levels showed that there were significant differences between the five morph levels. According to several pairwise comparisons, there is significant differences between neutral and 50% of morph level (p<.05), between neutral and 100% of morph level (p<0.05). In other words, There are no other significant differences between other morph levels.

Figure 4The differences in reaction time to detect anger facial expression between students with high and low anxiety

Figure 1.Distribution in groups of patients by the stroke type.

Figure 5The differences in reaction time to detect anger facial expression between students with high and low anxiety across five morph levels

Figure 1.Distribution in groups of patients by the stroke type.

4 . Discusion

4.1 The accuracy of detecting angry expressions

The results from the analysis of accuracy in detection of angry emotions appeared to be consistent with the hypothesis. Based on the total scores of accuracy, students with high anxiety can identify anger expressions of the facial images more accurately than the students with low anxiety (hypothesis 1). Even, when participants were displayed with different levels of anger intensity (from neutral to 100% of anger expression), the results also indicated a significant differences between the anxious groups in regards with accuracy of the detecting the threat related expressions. The anxious participants had also been able to identify angry expression accurately at a 50% of morph level which was expected to act as ambiguous stimuli. The results showed that individuals with high social anxiety overestimate perceived social danger even from subtle facial cues.

This finding seems compatible with the intentional bias towards threatening faces in anxious participants found in several other studies. For example, Cisles & Koster ( 2010) in their study had found that anxious individuals are more likely to allow threat-related information (in the form of words or faces) to unduly control their thoughts and actions. A study by Mohlman, Carmin & Price ( 2006), also found that participants with generalized social anxiety disorder (GSAD) sorted angry cards with greater accuracy as compared to the non-anxious control samples. There was also evidence that the GSADs tend to misclassified neutral cards as angry. Similarly, Schofield et. al. ( 2012), also found that social anxiety was associated with difficulty disengaging attention from threatening facial expressions as they combined a dot-probe task that required participant to disengage their attention by tracking the eye-movement. The result indicated that this response strategy may be related to adaptive strategy for preparing for potential threat before accounting it ( Mobbs et. al. 2009). Meaning that, the high anxiety students have a tendency to be cautious about potential threat in advance regardless of the actual threat evidence in the stimuli.

According to Koizumi, Tanaka, Imai et. al. ( 2009), individuals with high trait anxiety were found to have a tendency to interpret others’ emotional states negatively when the participants were presented with dynamic faces and voices in a synchrony pattern. The anxious individuals were also showed bias in their recognition of the face and voice stimuli if the images displayed in parallel, but not if they were presented in isolation. This recognition bias in high anxiety students is not generalized to recognition of emotional faces regarding to neural expression, but it is specific to angry. As regards to the high anxiety individuals, Pollak et al. ( 2005) suggested that individuals who had been exposed to the multiple life aversively events require more intense facial expression of anger due to the low intensity of angry stimuli because they are normally accustomed to more severe levels of anger expression. While the above studies were all conducted in a non-realistic situation, there were at least three studies examining the effect of anxiety on perception under a more naturally occurring condition. MacLeod, Mathews, & Tata, ( 1986), MacLeod and Mathews ( 1988) and Mogg, Bradley, and Hallowell ( 1994) had examined the attentional bias under condition of exam stress. The high anxiety students were tested one week before the exam and low anxiety students 12 week before the major exam. The result showed that the high anxiety students attended to the exam related words presented to them, while the low anxiety students avoided such words. In regards to developmental aspect, the result had strongly supported by other study from Lobue et al. ( 2014) whom had found that both children and adults detect angry adult faces as a threatening emotions faster than sad and happy faces regardless of whether they are posed by adults' or children's face stimuli.

However, several research have revealed contradict finding such as a studies by Schofield, Coles, &Gibb ( 2007) has explained that anxiety doesn’t have any impact on the detection of negative facial expressions and in a study by Philippot & Douilliez ( 2005) they have found that anxiety is also not biased in evaluating facial expressions more negatively. Despite of no differences between non-anxious participants and socially anxious participants found by the above researchers, the socially anxious participants systematically used additional visual information from eyes, nose and mouth regions to reach the same performance level as non-anxious participants ( Langner, Becker, and Rinck, 2009). In addition, Richards, Hadwin, Benson et. al. ( 2011) found that the high level of anxiety are associated with the increased processing capacity to detect multiple angry faces, even with the absence of threat, the anxious individuals tend to focus their attention in order to facilitate the detection of threatening stimuli in various conditions or locations to minimize the potential danger. The study also highlight that the high anxiety students have no deficit for discriminating neutral and angry expressions based on the correct response. Moreover, according to Quigley et al. ( 2012), the high anxious individuals might readily appraise the possible threat as the situation itself was not objectively threatening. Thus, the accuracy of detecting threat from angry expression may be more evident if the threat value of the stimuli is more difficult to appraise such as unpredictable or ambiguous stimuli.

4.2 The response time in detecting angry expressions

The hypothesis predicted that high anxiety students will detect anger facial expression faster than the low anxiety students. The result had also confirmed that the students with low anxiety spent more time to recognize the anger expression from the facial images as compared to the high anxiety students (hypothesis 3). A large number of previous studies reported that the unpleasant stimuli, including angry facial expression have indicated that anxious individuals took less time to identify threatening stimuli. For example, a study by Langner et al. ( 2015) in which the high anxious participants were faster overall in recognizing expressions across lower frequency band. However, according to Heimberg, 1995 in Langner et al, 2015, the participants have speed advantage in their performance, partly because they may become so eager to perform well in the face experiment. This phenomena is known as floor effect, in which socially anxious individuals seem to be more susceptible to the demand characteristic of experiments. Whereas, based on four experiments conducted by Fox, Lester, Russo, Bowless, Pichler & Dutton, 2000, presented kind of fairly clear evidence that angry faces detected more efficiently and rapidly compared to happy and neutral faces regardless of the used of schematic faces as stimuli through visual search experimental paradigm. The research also found that the detection of angry faces had speed up when they appear among three to seven neutral or happy distractors. The results also supported by Leber, Heidenreich, Stangier & Hofmann, 2009, confirming that individuals with social anxiety were faster than non anxious individuals when they were confronted with social-evaluative threat. However, both groups did not differ in response time to detect angry faces when they were under the non-threat conditions.

When the data analyzed further across the morphs levels, it seems that there is no significant effect of groups and no interaction between the anxious groups and the morphs levels. However, the significant effect was found in morph level, specifically between the neutral faces with 50% and between neutral faces and 100% of morph level. The data showed that the low anxiety students were faster in detecting angry expression at 50% of morph levels, while the high anxiety students were faster in detecting angry expression at 100% of morph level. Interestingly to learn that at 50% of morph level which is considered as ambiguous facial expressions had been perceived rapidly as threatening among the low anxiety students. It could be explained based on several studies that have already shown that the more aversive or negatively adults rate threatening stimuli, the more quickly they detect them (Beaver et al., 2005; Lundqvist & Öhman, 2005). Looking at the pattern of responding across different facial expressions, LoBue et al ( 2014) found that both adults and children detected angry faces the most quickly compared to happy and sad faces. Other explanation might due to the unnatural face stimuli generated through the process of morphing into different levels of angry expressions. It may produce intermediate facial construction that would not naturally present to the participants. Thus, there is no evidence that can be referred distinguishing the effectiveness of morphed vs. natural transitions of facial expressions. This explanation needs further investigation. Several limitations of the present study should be highlighted. First, the study was done with pre-selected students who scored not as high as expected on the STAI trait and STAI state. Naturally, the missing of selective effect on threatening facial expressions could be due to this fact. It could be suggested for further research with more socially anxious students in the future. Second, another limitation is that the female participant in this research is outnumbers the male participants due to the over-representation of female students. This of course hampers the generalizability of the results. Altogether, the present results showed that anxious students' exhibits greater perceptual sensitivity and response times to detect angry expression as threat related stimuli. This pattern of results broadly construed high anxiety students are having a tendency to interpret stimuli as threatening.

4.0 Conclusion

These study are crucial to our understanding about how facial expression could also be a source of threat for the anxious students. Finding of threat perception in Malaysia context allows us to reconstruct our own notions about how to interpret the data in terms of culturally driven ideas about the way adolescent might perceive people around them. This area of research would benefit from a more detailed look at the research design and the development of facial stimuli by using different face generated software that can produce better facial images for the facial emotional stimuli. This study has also provided some insights to future study. The finding should be taken as a guideline for future study rather than as definite answers to the research questions. Further study need to be conducted in order to verify the present findings by taking into accounts the recommendations for future study which has been suggested.

Acknowledgements

This work was funded by University Grant (Grant code: 2017-0269-107-01) from the Sultan Idris Education University. We are grateful to Research Management & Innovation Centre (RMIC) for all the assistance given in completing this research.

References

  1. Beaver JohnD, Mogg Karin, Bradley BrendanP. Emotional Conditioning to Masked Stimuli and Modulation of Visuospatial Attention. 2005;:67-79. CrossRef | Google Scholar
  2. Calvo ManuelG, Avero Pedro, Lundqvist Daniel. Facilitated detection of angry faces: Initial orienting and processing efficiency 2006-sep;:785-811. CrossRef | Google Scholar
  3. Fernandes Orlando, Portugal LianaCL, Alves RitaCS, Campagnoli RafaelaR, Mocaiber Izabela, David IsabelPA, Erthal FátimaCS, Volchan Eliane, Oliveira Leticiade, Pereira MirtesG. How you perceive threat determines your behavior 2013. CrossRef | Google Scholar
  4. Behnke Joachim. Felix Brosius: SPSS-Programmierung. Effizientes Datenmanagement und Automatisierung mit SPSS-Syntax$\mathsemicolon$ Andy Field: Discovering Statistics Using SPSS 2006-dec;:751-753. CrossRef | Google Scholar
  5. Fox Elaine, Lester Victoria, Russo Riccardo, Bowles RJ, Pichler Alessio, Dutton Kevin. Facial Expressions of Emotion: Are Angry Faces Detected More Efficiently? 2000-jan;:61-92. CrossRef | Google Scholar
  6. GrĂśs DanielF, Antony MartinM, Simms LeonardJ, McCabe RandiE. Psychometric properties of the State-Trait Inventory for Cognitive and Somatic Anxiety (STICSA): Comparison to the State-Trait Anxiety Inventory (STAI). 2007;:369-381. CrossRef | Google Scholar
  7. Gutiérrez-García Aida, Calvo ManuelG. Social anxiety and trustworthiness judgments of dynamic facial expressions of emotion 2016-sep;:119-127. CrossRef | Google Scholar
  8. Koizumi Ai, Tanaka Akihiro, Imai Hisato, Hiramatsu Saori, Hiramoto Eriko, Sato Takao, Gelder Beatricede. The effects of anxiety on the interpretation of emotion in the face–voice pairs 2011-apr;:275-282. CrossRef | Google Scholar
  9. Langner Oliver, Becker EniS, Rinck Mike. Social Anxiety and Anger Identification 2009-jun;:666-670. CrossRef | Google Scholar
  10. Langner Oliver, Becker EniS, Rinck Mike, Knippenberg Advan. Socially anxious individuals discriminate better between angry and neutral faces, particularly when using low spatial frequency information 2015-mar;:44-49. CrossRef | Google Scholar
  11. Leber Simone, Heidenreich Thomas, Stangier Ulrich, Hofmann StefanG. Processing of facial affect under social threat in socially anxious adults: mood matters 2009-feb;:196-206. CrossRef | Google Scholar
  12. LoBue Vanessa. More than just another face in the crowd: superior detection of threatening facial expressions in children and adults 2009-mar;:305-313. CrossRef | Google Scholar
  13. LoBue Vanessa. What Are We So Afraid of? How Early Attention Shapes Our Most Common Fears 2012-dec;:38-42. CrossRef | Google Scholar
  14. LoBue Vanessa, Pérez-Edgar Koraly. Sensitivity to social and non-social threats in temperamentally shy children at-risk for anxiety 2013-nov;:239-247. CrossRef | Google Scholar
  15. LoBue Vanessa, Matthews Kaleigh, Harvey Teresa, Thrasher Cat. Pick on someone your own size: The detection of threatening facial expressions posed by both child and adult models 2014-feb;:134-142. CrossRef | Google Scholar
  16. Lundqvist Daniel, Ohman Arne. Emotion regulates attention: The relation between facial configurations, facial emotion, and visual attention 2005-jan;:51-84. CrossRef | Google Scholar
  17. MacLeod Colin, Mathews Andrew, Tata Philip. Attentional bias in emotional disorders. 1986;:15-20. CrossRef | Google Scholar
  18. MacLeod Colin, Mathews Andrew. Anxiety and the Allocation of Attention to Threat 1988-nov;:653-670. CrossRef | Google Scholar
  19. Mobbs D, Marchant JL, Hassabis D, Seymour B, Tan G, Gray M, Petrovic P, Dolan RJ, Frith CD. From Threat to Fear: The Neural Organization of Defensive Fear Systems in Humans 2009-sep;:12236-12243. CrossRef | Google Scholar
  20. Mogg Karin, Bradley BrendanP, Hallowell Nina. Attentional Bias to Threat: Roles of Trait Anxiety, Stressful Events, and Awareness 1994-nov;:841-864. CrossRef | Google Scholar
  21. Mohlman Jan, Carmin CherylN, Price RebeccaB. Jumping to interpretations: Social anxiety disorder and the identification of emotional facial expressions 2007-mar;:591-599. CrossRef | Google Scholar
  22. Karaca Celal, Ocak MehmetAkif. Effects of Flipped Learning on University Students’ Academic Achievement in Algorithms and Programming Education 2017. CrossRef | Google Scholar
  23. Oosterhof NN, Todorov A. The functional basis of face evaluation 2008-aug;:11087-11092. CrossRef | Google Scholar
  24. Philippot Pierre, Douilliez Céline. Social phobics do not misinterpret facial expression of emotion 2005-may;:639-652. CrossRef | Google Scholar
  25. Pollak SethD, Cicchetti Dante, Hornung Katherine, Reed Alex. Recognizing emotion in faces: Developmental effects of child abuse and neglect. 2000;:679-688. CrossRef | Google Scholar
  26. Quigley Leanne, Nelson AndreaL, Carriere Jonathan, Smilek Daniel, Purdon Christine. The effects of trait and state anxiety on attention to emotional images: An eye-tracking study 2012-dec;:1390-1411. CrossRef | Google Scholar
  27. Richards HelenJ, Hadwin JulieA, Benson Valerie, Wenger MichaelJ, Donnelly Nick. The influence of anxiety on processing capacity for threat detection 2011-jul;:883-889. CrossRef | Google Scholar
  28. Ibrahim Abdul-Wahab. The Students’ Perception of Teachers’ Classroom Effectiveness on Their Self-Concepts in Lagos Metropolis 2014-jan;:133-141. CrossRef | Google Scholar
  29. Schofield CaseyA, Johnson AshleyL, Inhoff AlbrechtW, Coles MeredithE. Social anxiety and difficulty disengaging threat: Evidence from eye-tracking 2012-feb;:300-311. CrossRef | Google Scholar
  30. Schofield CaseyA, Coles MeredithE, Gibb BrandonE. Social anxiety and interpretation biases for facial displays of emotion: Emotion detection and ratings of social cost 2007-dec;:2950-2963. CrossRef | Google Scholar
  31. Spielberger CharlesD. State-Trait Anxiety Inventory for Adults 1983. CrossRef | Google Scholar
  32. Wieser MatthiasJ, Moscovitch DavidA. The Effect of Affective Context on Visuocortical Processing of Neutral Faces in Social Anxiety 2015-nov. CrossRef | Google Scholar
  33. Yoon KLira, Zinbarg RichardE. Interpreting neutral faces as threatening is a default mode for socially anxious individuals. 2008;:680-685. CrossRef | Google Scholar

Author's Affiliation

Copyrights & License

The International Journal of Social Sciences and Humanities Invention, 2019.
Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Article Details


Issue: Vol 6 No 2 (2019)
Page No.: 5312-5318
Section: Articles
DOI: https://doi.org/10.18535/ijsshi/v6i2.07

How to Cite

Mustapha, R., Shahadan, M. A., & Hamzah, H. (2019). The Effect of Trait Anxiety on Recognition of Threatening Emotional Facial Expressions: A Study among High School Students. The International Journal of Social Sciences and Humanities Invention, 6(2), 5312-5318. https://doi.org/10.18535/ijsshi/v6i2.07

Download Citation

  • HTML Viewed - 204 Times
  • PDF Downloaded - 53 Times
  • XML Downloaded - 0 Times