Alexithymia and Help-Seeking Attitudes in College-Aged Students

Aim: This epidemiological study focused on prevalence of the personality trait, alexithymia – the inability to recognize and describe one’s own emotions – among college-aged students. Method: Levels of normative alexithymia (measured by the NMAS), clinical alexithymia (measured by the TAS), as well as attitudes towards help seeking behavior (measured by the ATSPPH) were assessed. Results: A total of 547 participants were collected from over 48 colleges and universities, primarily in Western New York. Results indicated that 89.5% of college students displayed moderate to high levels of normative alexithymia and 48.7% displayed moderate to high levels of clinical alexithymia. Additionally, there were no significant differences between the scores of men and women on either alexithymia measure. Significant findings indicated that there was a positive relationship between clinical alexithymia scores and attitudes toward helpseeking. Conclusion: These results call into question previous literature, which states that there is no correlation between alexithymia and helpseeking behavior (Berger et al., 2005, https://doi.org/10.1037/1524-9220.6.1.73). The results of this study suggest a shift in culture, as well as a need for more research regarding alexithymia.

negative attitudes towards help-seeking. Consequently, additional research is needed to determine whether help-seeking behavior is impacted by characteristics of alexithymia. Furthermore, with the exception of a few examples, concrete, empirically-based techniques to help people who have alexithymia are lacking. There is very little research on how therapists can help their clients who show signs of this trait. However, to help his patients overcome alexithymia, Levant (1992) focused on expanding their vocabulary with regard to vulnerable emotions. More recent research suggests that helping individuals with alexithymia construct words to describe difficult situations, understand feelings of the situation, and then, learn how to respond emotionally have been effective in improving their ability to deal with problems (Vanheule, Verhaeghe, & Desmet, 2011). In turn, this may result in more effective treatment of related mental illnesses or other difficulties since the clients will be able to more effectively talk about and describe what they are feeling.

Gender and Emotions
As is implied by the socialization process accounted for in the normative male alexithymia hypothesis (Levant et al., 2006), it is not surprising that alexithymia is more commonly detected in men (Levant et al., 2009). It is debated, however, whether men and women differ in their ability to identify feelings, a common indication of alexithymia. Pérusse, Boucher, and Mylene (2012) report no difference while Levant, Hall, Williams, and Hasan (2009) indicate that men have a much more difficult time identifying feelings than women.
Identification of feelings is likely to be impacted by the emotional socialization of young boys by their parents and peers (Feder, Levant, & Dean, 2010;Levant et al., 2006) and is sometimes referred to as the boy code: the rules and expectations for boys that come about due to gender stereotypes (Pollack, 1998). These spoken or unspoken rules seem to be heavily influenced by paternal figures. Fathers teach their children about sex-roles much more than mothers do, and they start these lessons at early ages (Langlois & Downs, 1980). Along the same lines, Devine (2018) found that when looking at the interactions with sons versus daughters, both parents were more elaborative and encouraging with their daughters than sons except when it came to negative emotions. Research shows that boys block themselves from feeling vulnerable emotions because they are made to feel ashamed for having these emotions by their parents and peers (Feder et al., 2010). Consequently, boys learn to be less expressive and more controlling of emotions as they develop. By the ages of four to six years, young boys learn to hide their responses to emotion (Buck, 1977).
Given the emotional development of most boys throughout their childhood and adolescence, it is not surprising that men often find it difficult to find words to describe their emotions. Research shows that this development continues as boys get older. In a sample of college-aged students, men were more emotionally restricted and possessed less open attitudes to emotions than women (Komiya, Good, & Sherrod, 2000). Consistent with when they are young, men continue to be socialized to restrict their vulnerable emotions as they grow older. Lynch, Long, and Moorhead (2018) found that there is a lack of conversation with family and friends about mental health with this same age group which in turn leads to perceptions that mental health problems are socially undesirable. Additionally, inhibiting and regulating emotions when around others is a social norm for men and when a man does not follow these norms, they experience a threat to their manhood (Vandello & Bosson, 2013). As a result, instead of outwardly showing emotions, men tend to logically think about what they are feeling. They view vulnerable emotions (i.e., sadness or fear) as meant only for women (Levant et al., 2006).

Help-Seeking Behavior
Research has shown that the contrast between the roles and responsibilities of men and women, as well as their biological differences, contribute to the variation in mental health problems they suffer from and whether they actively seek help or not. Specifically, men are more likely than women to seek care at a later point after the onset of their symptoms. They might even delay until the symptoms become extremely severe (World Health Organization, 2002).
Traditional ideas of the man's role in society, concern over expressing affection toward other men, and concern about expressing emotions in general are all reasons why men have more negative attitudes about seeking professional psychological help (Good, Dell, & Mintz, 1989). Age also seems to play a role in attitudes toward help-seeking behavior. The older men are, the more positive attitudes they have towards seeking psychological help (Berger et al., 2005). Low mental health literacy, stigma, and not being able to physically see or touch what they are feeling are additional barriers to help seeking for men (Lynch et al., 2018). Contrastingly, the need for success, power, and competition does not seem to play a role in men's help-seeking attitudes (Good et al., 1989).
Researchers suggest that men might be more likely to seek psychological help if it is thinking-focused rather than feeling-focused (Berger et al., 2005). Since men associate feelings with being feminine, cognitive-oriented, rather than emotion-oriented, counseling techniques could help men be more open and have positive attitudes towards therapy (Wisch, Mahalik, Hayes, & Nutt, 1995). Additionally, cognitive approaches focus more on problem solving skills than other approaches, which is preferable to men, especially those who have high gender role conflict (Wisch et al., 1995). Recent research has reported that increasing mental health literacy and small group based programs may additionally encourage more men to seek help (Lynch et al., 2018).
Other factors, besides gender, may also contribute to attitudes toward help-seeking. Depending on a person's diagnosis, there are different reported rates of treatment use. According to the NIMH (2015), more than half of individuals diagnosed with a mood disorder (i.e., depressive and bipolar disorders; 56.4%) or post-traumatic stress disorder (57.4%) are receiving treatment. The rates are less for anxiety disorders (42.2%), personality disorders (39%), and eating disorders (anorexia nervosa, 33.8%; binge eating disorder, 43.6%; bulimia nervosa, 43.2%). Obsessive-compulsive disorder, dissociative disorder, attention deficit hyperactivity disorder, and autism spectrum disorder treatment rates were not reported (NIMH, 2015).
Among college students specifically, only 40.4% of individuals seek treatment (NIMH, 2008). Students report that lack of time, privacy concerns, lack of emotional openness, and financial constraints as barriers to seeking treatment (Hunt & Eisenberg, 2010). Stigma has also been reported as a key factor in why students may have more negative attitudes toward seeking help (Eisenberg, Downs, Golberstein, & Zivin, 2009).

Present Study
It is important to provide more research regarding alexithymia and its possible connection with help seeking.
Doing so may help improve treatment for individuals who are seeking counseling and also have this trait. Additionally, it may help us understand whether people with this trait may be more or less likely to seek counseling.
The first necessary step, which was addressed in this study, was to determine the prevalence of alexithymia in college-aged students and the relationship between alexithymia and attitudes towards help-seeking. To our knowledge, only two previous studies report prevalence rates of alexithymia in the general population in the United States. One, using the Toronto Alexithymia Scale (TAS), reported prevalence rates of clinical alexithymia (13.4% high alexithymia and 18.4% moderate alexithymia; Lane et al.,1996). The second, also using the TAS, reported that out of 234 college students, 12.4% had alexithymia (Schuetz & Multon, 2017). It is important to consider studies that have reported prevalence rates of alexithymia in countries outside the United States as well when creating a hypothesis. Research conducted in Finland showed that alexithymia is estimated to effect 13% of the general population (Salminen et al., 1999). An 11-year follow-up study reported alexithymia to be a stable personality trait in the Finnish population (Tolmunen et al., 2011). Additionally, research in Spain showed that there was a significant percentage of young people who have high levels of alexithymia (24.1%) although there were not significant differences between men and women (Galván, 2014). Alexithymia has also been shown to be more prevalent in women in clinical samples (de Barros, Furlan, Marques, & de Araujo Filho, 2018) but to our knowledge, has not been studied in a college sample. In this study, it was hypothesized that alexithymia, at both the normative and clinical levels, would be prevalent among college students and rates would be similar to the United States rates mentioned above. Additionally, while there is a plethora of research that states that normative alexithymia is more common in men (e.g., Levant et al., 2006), there have not been any exact prevalence rates reported. When comparing genders, it was hypothesized that there would be a larger number of men than women at both levels, which is based on previous research findings, as well as the nature of the trait.
The relationship between alexithymia and help-seeking behavior was varied in the literature; so, further assessment was warranted. The current study assessed attitudes, not behaviors, but attitudes are an indication of whether or not an individual will perform that behavior (Fazio & Williams, 1986). The present researchers predicted that higher alexithymia scores would result in negative attitudes towards help-seeking since therapy is often perceived to be focused on emotions and people with higher levels of this trait have difficulty identifying and describing emotions (Berger et al., 2005).

Method Participants
Data was collected for the present study from 619 college-aged students. The total sample size was reduced to 547 due to exclusion of participants who: had missing data, completed the survey twice, were not between the ages of 18 to 24 years, or did not identify as a man or woman. The participants (ages 18-24, M = 19.77, SD = 1.544) were from various postsecondary institutions, mostly located in the Western New York region. The majority of students (47.1%) were enrolled in medium program master's colleges and universities as according to the Carnegie Classification system (Carnegie Foundation for the Advancement of Teaching, 2015). Of the total sample, 171 participants identified as men and 376 as women. Nearly 80% identified as white, 7.3% as black or African American, 6% as Hispanic or Latino, 3.7% as Asian/ Pacific Islander, 2% as bi-racial, 1.1% as Native American, and 0.4% as other. Class year of participants was indicated as follows: 32.3% first year, 20.7% second year, 24.8% third year, 17.8% fourth year, 2% fifth+ year of undergraduate coursework, and 2.4% graduate students. Using the New York State Department of Education (2010) college major classification system, psychology was the most popular major (30.2%) among our sample. Over 7% of students reported being diagnosed with a mental illness in the previous 12 months and 24.2% reported being diagnosed in their lifetime for a total of 31.7% answering yes to having a diagnosis. The remaining participants answered no (66.6%) or unsure (1.7%) to the question regarding diagnosis. The most prevalent diagnoses were anxiety (22.9%) and depressive disorders (18.5%). Approximately 7% of the participants had ADHD, 3.5% disclosed an eating disorder, and 2.7% endorsed the presence of PTSD. Each of the remaining disorders was endorsed by less than 2% of the total sample (see Table 1). Participants were recruited through a variety of methods. At one small western New York university, recruitment involved the Introduction to Psychology participant pool and an e-mail was sent through the university's residence life office to all of the students living on campus requesting participation. Additionally, several other western New York postsecondary institutions were asked to send the survey to their students. The survey link was also shared on social media (i.e., Twitter and Facebook) as well as posted on the Psi Chi National Honor Society website and on a national Social Psychology database.

Measures Toronto Alexithymia Scale (TAS)
The TAS (Bagby et al., 1994) is a 20-item inventory used to assess clinical alexithymia by measuring 3 factors: difficulty identifying feelings, difficulty describing feelings, and externally-oriented thinking. It uses a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree) and scoring for 5 of the questions is reversed. This results in a total score and is interpreted as follows: higher scores indicate higher levels of alexithymia and lower scores indicate lower levels of alexithymia. Scoring at or above 61 means the participant has high alexithymia (i.e., alexithymia), and a score at or below 51 means low alexithymia (i.e., non-alexithymia). The TAS has good consistency and reliability (α = .86, r = .77; Bagby, Ayearst, Morariu, Watters, & Taylor, 2014).

Normative Male Alexithymia Scale (NMAS)
The NMAS (Levant et al., 2006) is a 14-item self-report measure that uses a 7-point Likert scale to assess normative male alexithymia, which is known as a milder form of alexithymia. The scale ranges from 1 (strongly disagree) to 7 (strongly agree). It was initially developed for assessing men, but can also be used on women, and has been used on women several times in the past (Levant et al., 2006). To obtain the total score, seven items must be reversed. The sum of the responses is then divided by 20 for the mean item score, or total score.
Higher scores on the NMAS indicate higher levels of normative male alexithymia and lower scores indicate lower levels of normative male alexithymia. The NMAS has shown good internal consistency (α = .92) and good test-retest reliability (r = .91; Levant et al., 2006).

Attitudes Toward Seeking Professional Psychological Help (ATSPPH)
The ATSPPH (Fischer & Turner, 1970) is a 29 item self-report measure used to assess help-seeking attitudes.

Procedure
After the study was approved through the University's Institutional Review Board, data collection began through an online program. A hyperlink to the study was provided in the invitations to participate. Before beginning, participants were asked to consent online by typing their name after reading the form. After consenting to participate, college-age students answered several demographic questions and were asked if they had ever been diagnosed with a mental illness. Specification of particular diagnoses was asked if the respondents reported the presence of a mental illness. Participants were then asked to complete three self-report measures respectively: the NMAS, TAS, and lastly, the ATSPPH. After completion of all three surveys, a debriefing statement and the opportunity to receive compensation was presented. The total amount of time to complete the forms and questionnaires was 20 to 30 minutes.
Upon completion of the survey, participants were asked if they would like to receive a $1.00 gift card to Amazon as compensation. Psychology students from particular institutions were offered either class credit or the gift card. Regardless of whether they received class credit or the gift card, all participants were also asked if they wanted to be entered in a drawing for a $5.00 Amazon gift card.

Results
Nearly 50% of the participants in this study had moderate or high alexithymia, as measured by the TAS.
Additionally, approximately 90% had moderate or high levels of normative alexithymia, as measured by the NMAS (see Table 2). The relationships between both alexithymia measures and gender yielded similar results. Using cross tabulation, results indicate little variation between the scores of men and women on the TAS or the NMAS (see Table   3). A chi-square test showed that neither normative alexithymia, χ 2 (2) = 2.135, p = .344, nor clinical alexithymia, χ 2 (2) = 2.063, p = .356, showed a significant gender difference. Analyses regarding the relationship between alexithymia scores and help-seeking attitudes indicated that NMAS total scores were not significantly associated with attitudes towards help seeking (β = -.068, p = .227). Surprisingly, higher TAS scores significantly indicated more positive attitudes towards help seeking (β = .596, p < .001, Cohen's f 2 = .429).
Additional multiple regression analyses were done with mental health diagnoses to examine the relationships between diagnoses and attitudes towards help-seeking (see Table 4). The only diagnosis that reached statistical significance was depression (β = -.160, p < .05, Cohen's f 2 = .041), which was significantly associated with negative attitudes. A sensitivity analysis was performed to determine if there was an interaction effect between the reported presence of a mental illness and alexithymia scores on help-seeking attitudes. It was determined that there was a nonsignificant interaction between diagnosis and score on the TAS, F(1, 533) = 3.027, p = .082. Additionally, there was a nonsignificant interaction between diagnosis and NMAS score, F(1, 529) = 3.650, p = .057.

Discussion
The current study measured the prevalence of both clinical and normative alexithymia among college-aged students. Results from the present analyses show that the highest prevalence rates are in the low clinical alexithymia and the moderate normative alexithymia categories. This means that many college students do have difficulty, though usually at a mild level, identifying and describing their emotions. These results were expected though previous literature had reported differently. In one previous study using the TAS, 13.4% of the population had high alexithymia and 18.4% had moderate alexithymia (Lane et al.,1996). In another study, 12.4% of college students had alexithymia (Schuetz & Multon, 2017). The current study reports rates of 27.2% for high alexithymia and 21.5% moderate alexithymia. While previous research does not provide prevalence rates for normative alexithymia, rates were highest for the moderate normative alexithymia level (69.3%). One possible explanation for this could be due to the fact that people do not verbally communicate as much as they used to. With the introduction of technology and texting, students most often communicate with others through use of their phones. In fact, technology-based communication is dominant among millennials (Botterill, Bredin, & Dun, 2015). As a result, there appears to be a cultural shift where discussion about sensitive topics, such as our emotions, occurs through technology rather than in face-to-face conversations (Dolev-Cohen & Barak, 2013;Green et al., 2005). Therefore, it is possible that technology could make it more difficult for individuals to identify and verbalize how they feel.
In addition to assessing overall prevalence in college students, the current study aimed to also look at prevalence based on gender. The analyses showed that there were no significant differences between the scores on both the clinical and normative tests for men and women. This leads the researchers to question the hypothesis created in the 1970s stating that alexithymia is more common in men than women and may even be normative in men (Levant et al., 2009). Social norms and roles have changed since the 1970s, and we are breaking what has been known for so long as normal gender roles. In the 1970s, roles were just beginning to change.
While there was growing support for gender equality and a less restrictive view of gender roles, men were still expected to be dominant, attend college, and make money for their families, while it was the women's job to care for the children and the home (i.e., cook and clean before their husband came home; Brooks & Bolzendahl, 2004). Today, however, women are embarking on more professional roles than in the past and joining men in the workforce at relatively similar rates (United States Census Bureau, 2015). In fact, it is now the norm for women to get a job and contribute to the family's finances which can be seen by the increase in their rate of participation in the workforce (Gardner, 2017). As women are becoming more professional, they are also being taught to be less emotional. Women are told to act like men do, meaning that they need to be stoic and unemotional, especially at work (Wajcman, 2013). It appears women are possibly being socialized in the same way that men are, but it is unclear at what age this is happening, whether it is when they are young or in high school and college. Young boys are often instructed by caregivers early in childhood to hide their vulnerable emotions (Feder et al., 2010;Levant et al., 2006), and it appears as though women are given the same messages at some point in life. If they want to be taken seriously, they are told to not cry or express any sad emotions. With women and men both taking on the professional role, women are also being taught to hide their vulnerable emotions (Perrone et al., 2009).
The current study also sought to look at the connection between alexithymia and attitudes toward help-seeking.
Results indicated that higher scores on the NMAS were associated with more negative attitudes towards seeking professional psychological help. This finding, however, was not statistically significant, which supports Berger et al.'s (2005) claim that there is no connection between alexithymia and help-seeking behavior. In contrast, higher levels of clinical alexithymia were significantly associated with more positive attitudes towards seeking professional psychological help, which supports Levant et al.'s (2009) claim. This finding was significant and leads us to believe that people with this trait may not be able to communicate their emotions well, but they understand that this is a difficulty that can be addressed through counseling. It is possible that individuals with characteristics of alexithymia are aware that something is wrong and although they cannot describe it, they recognize that seeking a therapist's support could be helpful.
Overall, the present findings do not reveal whether there is a direct relationship between alexithymia and attitudes toward help-seeking. It is possible that there may be other factors not measured in this study that have contributed to these results. For example, students on most college campuses have access to free mental health care, which may give them more positive attitudes towards seeking help. While stigma is still evident on a college campus, slowly decreasing rates of stigma may also play a role in students' attitudes towards seeking help. Previous research has found that if there is less judgment and disgrace surrounding seeking help, especially within people's own perceptions, individuals would develop more positive attitudes towards it (Eisenberg, Downs, Golberstein, & Zivin, 2009). Also, about 32% of the students in the current sample were psychology majors, which may also play a role in positive attitudes, due to the increase increased knowledge of mental health services and its benefits. Additionally, because attitudes do not necessarily mean that they will seek help, it is unclear whether individuals with higher levels of clinical and normative alexithymia will actually seek help if they need it. While the current findings cannot answer that question, further research including these variables may be able to.
Additional analyses were done to look into the association between college students' diagnosis of mental illness and help-seeking attitudes. The present study found that individuals who have been diagnosed with depression have more negative attitudes towards seeking help. This is supported by previous research, which states that participants diagnosed with depression are not likely to seek help, due to concerns about disclosing their feelings to informal sources as well as professionals (Barney, Griffiths, Christensen, & Jorm, 2009). This may also contribute to why only a little over 50% of people with mood disorders seek treatment (NIMH, 2015).
There were no other significant results for associations between diagnoses and attitudes towards help-seeking.

Limitations and Future Directions
One limitation to this study is that the measures were provided in an online survey format. Thus, the diagnosis question could have been misinterpreted. Although participants were asked to only endorse diagnoses that had been made by mental health professionals, some participants may have endorsed diagnoses that they thought they had or that had been self-diagnosed. Another limitation is the sample, which was mostly comprised of students at colleges and universities in Western New York. Though the sample size was large, a better representation of students in the United States is needed for future studies. Other limitations include not measuring income or socioeconomic status of the participants.

Carosa & Button
Future research could also assess the connection between alexithymia and help-seeking behavior, not just attitudes. Someone may have positive attitudes toward seeking help but may not necessarily seek help. The current study was limited to only attitudes, so, an accurate representation of college students' help-seeking behavior was not depicted. Alexithymia and gender roles could also be studied to see if there is as strong a correlation as previous research states (Fischer & Good, 1997). Since the current research shows no significant connection between gender and alexithymia rates, looking at alexithymia and gender roles may show whether or not gender roles are changing, as is hypothesized. The association between technology and alexithymia could also be researched. The platform people feel most comfortable talking about their feelings on could lead to results that show whether individuals are primarily using their smartphones or other technology to communicate about their emotions rather than face-to-face. This would likely be related to symptoms of alexithymia. Lastly, different age groups and alexithymia should be considered. Would the same results be found in children or adolescents? By looking at this, research may be able to pinpoint whether there is a shift in women and their emotional expression. Since the current results show that there is not a significant gender difference in levels of alexithymia, but previous research says that only boys are socialized to hide their vulnerable emotions, it is important to determine when women are being taught this (Feder et al., 2010).
Overall, there needs to be more research done on alexithymia, especially since it has been shown to be prevalent in college-aged students. Students need to learn how to identify and communicate their emotions particularly since hiding vulnerable emotions can lead to negative effects (Feder et al., 2010).

Competing Interests
The authors have no conflicts of interest.

Funding
Funding was received from Psi Chi: The International Honor's Society in Psychology and the Alfred University Honors Program.